eCommerce Search Filters – 13 Best Practices You Need to Implement ASAP to Increase Conversion Rate

eCommerce Search Filters – 13 Best Practices You Need to Implement ASAP to Increase Conversion Rate

As a shopper, finding a product that you need should be easy in any online shop, but more often than not, it isn’t. You have to deal with clumsy filters, outdated search functions, 0 result pages, and irrelevant results.

Using eCommerce search filters can fix all of these challenges, but there are a number of things retailers have to pay attention to in order to create a great shopping experience.

In this article, we will show you:

  • filter design best practices
  • what filters you should use
  • how you should present them

and even give you a few SEO tips.
Let’s start with the basic question…

What Are eCommerce Search Filters?

Filters are essential tools for refining large amounts of information effectively by providing certain criteria and narrowing down the results by excluding the items that do not match the set criteria. eCommerce filters follow this exact logic, many times adding real-time results, and other convenient features to make filtering easier.

As Nielsen rightly states, filters and faceted search are often used as interchangeable concepts today, however, this is, in fact, not the reality.

Now, what is faceted search? While still quick and easy to use, it is a more complex process that uses a multitude of site search filters that can be automatically generated by the algorithm reading the different attributes present in the database.

For example, if in an online store you type “black shoes” in the search box, you will likely get results containing either or both of those words. But if you want to narrow your search, you have to type in additional phrases and browse by relevance.

Faceted search on the other hand will provide results containing either or both phrases, and then offer you additional filters which you can use in parallel. So you keep the results for “black shoes”, but then you filter for size, materials, brand, the shade of color, etc.

A good example for this is Zappos: if you put in the initial phrase, you will then be able to filter further by Women’s Size, Women’s Width, Men’s Size, Men’s Width, Kids’ Sizes, Width, Heel Height, and so on.

eCommerce Search Filters on Zappos

Why Use Product Filters?

Online shopping problems are in many cases caused by the customer’s inability to find what they are looking for in a quick and easy fashion.

There may be a lot of reasons for this, from a slowly loading site to horrendous design choices, but in a great number of cases, it can be distilled down to problems with the site search function and especially the use of filters – or the lack thereof.

But let’s dig in a bit deeper and talk about some of the exact benefits of using filters.

  • Providing too many choices can confuse and overwhelm your customers. They can easily fall victim to the paradox of choice and leave your site without making a purchase, even if you have what they want.
  • Filters narrow down the choices while still offering comparisons.
  • Filtering also makes finding the ideal products much quicker than scrolling through endless category pages, and thus, also improve user experience.

By including search filters in your eCommerce store, you can greatly provide a much better user and customer experience to your visitors than even some multi-million dollar online retailers.

According to the Baymard Institute, which conducted a thorough analysis of many of the largest online stores:

Analyzing [71 Major eCommerce Sites] we’ve found the average site to perform mediocre at best, and 36% of sites to have such severe design and feature flaws that it was downright harmful to their users’ ability to find and select products.

In fact, according to Smashing Magazine, as many as:

  • 42% of top e-commerce websites lack category-specific filter types,
  • 20% of top e-commerce websites lack thematic filters,
  • 32% of websites either have insufficient truncation design

and the list could be even longer. In fact, Baymard concluded that only 16% of the analyzed sites provided a good filtering experience.

In short: this is an area where you can easily compete with the largest online retailers in the world, and win.

Learn more about the most important eCommerce search engine features to increase conversion rate by 15% & revenue by 47%.

E-Commerce Product Filter Design Strategies

First, let us look at a few basics about how you should present your filters to ensure the best user experience.

The most important things to think about are:

1. The best location for your filters

When thinking about the placement of your filters, you basically have two options: a horizontal filter bar on top of the page, or a vertical filter bar on the left side of the screen.

There are pros of each option.

  • A horizontal filter bar will be more effective in focusing the attention of the shopper and in many cases will be more convenient to use.
  • A vertical sidebar on the left is where users are used to finding filters, so it offers a familiar experience, and it can also contain a much larger number of filtering options.

Amazon is one of the best-known examples for the vertical filter bar, offering even dozens of options depending on the products.

Search Filters on Amazon

Victoria’s Secrets uses both: when you enter a search term, you are given a sidebar for faceted filtering, however, on category pages they opt for the simple horizontal bar, given there are fewer options inside a certain category:

Search Filters on Victoria's Secret

Learn more about the tpyical beauty industry challanges in eCommerce and important search features to increase conversion rate & revenue by 6%

2. Horizontal Search Bar: Design and Customer Experience

In addition to the two solutions above, Algolia has tIf you opt for a horizontal search bar, you have to pay very close attention to how you design it to keep the customer experience excellent.

  • You will have limited space, so you might want to hide some of the filtering options – you can read about how to do it in the next few paragraphs.
  • Make sure the drop-down menus don’t automatically disappear when the cursor is not hovering over them.
  • Use icons to help shoppers quickly understand the options and to streamline navigation.

IKEA opts for a hybrid solution: on category pages, they present the most important filters horizontally, while also providing an option to open a vertical filter bar on the right for additional options.

And their main bar is also simplified, you can only see the first 4-5 options right away, so the drop-down menu doesn’t cover your entire screen.

Search Filters on IKEA

3. Present your filters wisely

If you have a large number of filtering options, you have to make some compromises when presenting them to your shoppers. Shoppers have to be able to access all options while focusing mainly on the most popular and most frequently used ones.

There are a few ways to present the filters:

  • You can offer all filtering options at once, which can be a bit overwhelming if there are a lot of them.
  • You can add scrolling to the individual categories. Moving the mouse in and out of the different boxes where you want to scroll however can be a bit inconvenient and can cause accidental scrolling on the page itself.
  • You can opt for showing only the filter categories or headers as drop-down options.
  • You can use truncated filters, where you present some of the options – preferably the most popular ones – and provide a “show more” option.

The main goal is to restrict the length of the sidebar in order to present as many of the filter categories as possible while also hiding some of the specific options.

Book search filters on

Pay attention to UX: if you decide to hide options behind a scroll, a drop-down menu, or a “show more” link, make sure that it is clear to the user how they can access the additional options. For example, you can use different colors for links and arrows.

4. Make sure applied filters are obvious

It is always important that the users understand exactly where they are in any process. This is also true for filtering, so the applied filters should always be obvious.

Shoppers should understand exactly which filters are applied and why they are viewing the given set of results. They should be able to easily make changes to the filtering options if they are not satisfied with the results.

You can make applied filters obvious in multiple ways, for example:

  • With checked or filled checkboxes.
  • Displaying cancellable tags.
  • Providing a back arrow for selected subcategories to go back one level.
  • Displaying the currently applied filters in a horizontal bar above the results.

The main goal is to make it easy for your shoppers to navigate amongst the applied (and non-applied) filters.

5. Your filtering decisions are SEO decisions, too

Apart from the user experience, you also have to take into account how different approaches to filtering affect your organic traffic.

For starters: filtering, and specially faceted filtering, are great tools, but they don’t in any way replace the need for category and subcategory pages.

For starters: filtering, and specially faceted filtering, are great tools, but they don’t in any way replace the need for category and subcategory pages.

This mainly has to do with your URLs – which are not primary ranking factors normally but bear weight in eCommerce. Here is how it works: you can’t put in-depth written content on category or product pages, your product descriptions can only be so detailed before they become counterproductive.

So what you do is create a category page. For example, if you want to rank for “leather jeans”, you will have a category page in your store with the URL

This is more likely to rank with all the products listed on it.

But this is only the smaller part.

You see, faceted filtering, and basically any kind of dynamic filtering, even just using a simple search function in an online store tends to create a large number of result pages.

Every time someone enters a phrase, a new result page will be generated for the results, with the URL something along the lines of

and so on. And then the filters get added to that.

The first problem is duplication: now you have dozens, hundreds, or thousands of URLs with the same phrase. This can be mitigated by telling the developers to add canonical tags to the category pages for the phrases you want to rank for.

But this kind of endless page-creation presents another problem, albeit not a huge one. You can very easily end up with millions of pages under your domain, with 99% of them bearing basically the same, repeating thin content, like search results in a list in different orders.

To avoid this problem,  it makes sense to have you developers indicate in the robots.txt that Google should ignore all search-related URLs.

Basically, you don’t want the majority of your crawled URLs to look like this:

eCommerce Search Filters URLs can cause SEO problems

Ecommerce Product Filter Types

Based on the type of products you are selling, there can be a myriad of attributes that your shoppers can filter for.

The most common ones used across eCommerce sites are:

  • Brand
  • Price (or price range)
  • User ratings
  • Color
  • Material
  • Size
  • Theme (e.g. occasion)
  • Popularity

Also, you can include filters for ongoing promotions.

Apart from these, if you use the right faceted search solution, you can automatically create any number of filters based on the attributes your products have, which makes it easier for your customers to find what they’re looking for. 

Keep in mind that it also matters how you visualize these filtering options and how useable they are. To give you a good idea of the basic requirements for a good filtering experience, let’s talk about…

Product Filtering Implementations

6. Use category-specific filters

There are general eCommerce filters that can apply to a larger range of products, and which are very handy on category pages. They can narrow down the results without the need to specifically search for something. For example, in a store selling clothes, there will be a set of site search filters that can be generally applied to most products like gender, size, color, or price range.

But some of the eCommerce filters are very specific, they refer to the attributes of a given set of products – like when you search for books on Amazon, a filter may be offered so you can narrow down the results to specific award-winning books. This filter naturally won’t appear if you search for leggings.

These category-specific eCommerce filters should be treated uniquely. You should bring your shoppers’ attention to them to encourage filtering and thus provide a better customer experience.

So how can you do that?

7. Use horizontally displayed promoted filters

As we have seen in IKEA, they use a hybrid solution of horizontal and vertical filter bars, which makes sense because there is a great difference between eCommerce filter types even if they seem similar…

It makes sense to put category-specific filters horizontally on the top of the category page even if your faceted filtering solution appears in a sidebar. This way, they will not be overlooked.

You may call these “promoted filters”, which encourage shoppers to select one or more to get to their desired products faster.

But as with all horizontally displayed filters, they have to be curated carefully, because again, you have very limited space to present them.

If you use horizontally displayed promoted filters, these are some things you should keep in mind:

  • Choose the most important filters: promoted filters don’t have to be consistent across the entire site. Choose the ones that are relevant in the given category, and display a handful of the most important ones like size, brand, price, etc.
  • Display the promoted filters in the sidebar. A number of customers will instinctively look for filters in the sidebar, so don’t exclude your promoted filters. Display them in both positions to maximize their visibility.
  • Promoting doesn’t mean overdoing design. The horizontal bar should be noticeable and easy to use, but avoid banner-like graphics, because they can be counterproductive and dissuade users from paying attention to the promoted filtering options.

8. Hide filters that lead to 0 result pages

We have talked about how to optimize 0 result pages in length before. However, ideally, you can prevent shoppers from ever reaching this dead-end in the first place.

With dynamic filtering, you should show the customer only the products that are in stock. So for example, if you are selling shirts, you may have a few category-specific filters pre-set, like color. Let’s say they have the option to choose if they want to see red, green, or blue shirts.

If you currently have no green shirt in store, two things can happen.

One: they select the “green” filter and are faced with a 0 result page.

Two: you automatically hide the “green” option, and only show the red and blue. This way, you avoid the frustration they might feel if they set a product filter that is unintentionally too narrow to return actual results. eCommerce Filters UX is as important as the overall user experience like the search autocomplete feature!

Learn more about the most important search autocomplete features to decrease zero-result pages by 31% and increase cart value by 19%.

For example, here is Oakley’s store – they are a little more direct about this; they don’t hide the options, but instead make sure you understand that they are not active, as the Holbrook model family has no yellow or green models:

Automatically hide search filters related to out-of-stock-products

9. Use real-time interactive filtering

As a general rule, every process in your store should be as short and quick as possible. Not because you want to rush your customers to the check-out page, but because they should feel that the shopping process is easy and streamlined.

This also applies to search and filtering. Most stores use batch filtering, which means the user has to choose the desired filters then take an action – click on a button, usually – for the results to be displayed.

A much better solution is to use interactive filtering. This means that the results are instantly changed on the result page as soon as a filter is applied, no further action from the shopper is required.

This is especially handy if you have a great number of filtering options, as clicking a button every time the shopper wants to narrow down a search can get old quickly.

Batch filtering on the other hand can also have its own advantages, mainly if your site has issues with speed. This is of course an entirely different problem, but if you absolutely have to cut some corners, using batch filtering can help.


10. Allow your customer to apply multiple filter values

While shopping, people often want to see a wide range of options that cannot be described by one filter.

Let’s say they want to buy a book written by Isaac Asimov. Now, Asimov published almost 500 books in his lifetime, and on wildly different topics. He authored science-fiction and fantasy books, textbooks, mystery, thriller and horror books, and even cooking books.

They might be interested primarily in his fiction, but that means the customer has to look at a number of different genres because this includes sci-fi, fantasy, horror, and so on.

Interestingly, many online stores don’t offer the opportunity to select multiple filter values in certain filter categories. Many choices are mutually exclusive, so as a shopper if you want to broaden your search a bit, you have to discard the initial value and apply another – while memorizing the results from the first value.

This can be complicated and frustrating for shoppers.

So, always make it possible for people to select multiple values of a filter, and display all the results that match those values. If they want to narrow down their search, they can always discard some of the selected options.

11. Offer thematic filtering options

For many products, plain attributes are not enough to filter results.

For example, if I am shopping for shoes, I already know what size I want to filter for, I also know what colors I like, if I want it to be waterproof, and so on. But mainly I want to filter by activity.

I might need it for running, hiking, for everyday commuting or for a certain social occasion. There might not be a separate category for all of these, but as a shop owner, you can take the products in a category and assign thematic filters to them.

This will help your customers locate the specific type of product they want and improve their experience on your site.

Macy’s for example offers an entire filter where you can select the exact occasion you need the dress for:

12. Make filter changes separate browser history events

If a customer narrows their search too much and wants to get back to the previous set of results, they may just discard the filter value they applied last. But many of them will do something else: hit the back button of the browser.

This is because when the displayed results change in front of their eyes, they tend to look at this as a new page – even if there was no apparent page reload.

And if they hit the back button, that can lead to an unpleasant experience if they are taken out of their search, and taken back to a category page for example.

The solution is to register every single filter application as a separate event in their browser history. You may think this can only be done if the page is reloaded each time, but luckily with the HTML5 History API you can preserve the seamless experience and still treat every filtering action as a separate event.

13. Display the applied filters separately

Make sure that your customers can always see which filters have they already applied – this makes it easier for them to navigate the results, discard filters, or apply more to further narrow down the results.

If you don’t display the applied filters separately, shoppers may misinterpret what they are seeing – especially if there are many filtering options as they might not remember what they already applied.

Applied filters should…

  • Be displayed separately, while also being displayed in their original position.
  • Be displayed on the top, in the horizontal bar, preferably.
  • Be easy to deselect.

This way you can make it clear to shoppers what criteria affect the results they are seeing and provide convenient options to change them.

For example, eBay displays the selected filtering options right above the results, where you can deselect them with a single click:

Key Takeaways

Using filters that are optimized not just for your products, but for the behavior of your customers will increase your profits. It’s that simple.

Exits and cart abandonments will decrease, while average order value and overall conversion rate will increase.

This is because you make it quicker and easier for shoppers to find exactly what they are looking for, providing them with a great experience.

In order to do that, you must pay attention to the best practices and design your filtering solution in a way that can actually help.

If you do that, you can successfully outperform some of the largest brands and online retailers in the world.

Paige TyrrellHead of Marketing – Prefixbox

Paige is the Head of Marketing at Prefixbox, a leading eCommerce site search solution. She’s an American who’s been living in Budapest since 2017 and loves giving #alwayslearning sessions to help people optimize their online stores.

10 Autocomplete Search Best Practices – How Predictive Search Will Generate More Revenue for Your Store

10 Autocomplete Search Best Practices – How Predictive Search Will Generate More Revenue for Your Store

One thing is certain: around 30% of your online store visitors will use your on-site search function – and 25% of them will click on a search suggestion.

If you provide suggestions, that is.

This is why predictive autocomplete search is an essential tool for any online retailer.

And we’re here to show you how to leverage it to maximize conversions and revenue.

We need something here that we want to clarify the basics first.

Chapter 1

What is Autocomplete Search?

Autocomplete is the function of search engines that displays keyword and product suggestions in real-time, based on what search query the user is typing into the search field.

The feature works in a very simple way: it detects what a customer is typing, and matches the query with data in the search index. If there are keywords or phrases stored in the index that matches what’s being typed in, it suggests those.

In reality, it’s a bit more complicated because they also take into consideration the popularity of different products and keywords when ranking suggestions.

For example: in an online store selling clothes, if you type in “sh”, it might suggest shoes or shirts. If the store is selling bathroom appliances, it might suggest shower curtains or shelves.

Chapter 2

How does Autocomplete Search work?

Apart from using your site’s search index, autocompletes also takes previous user behavior into account, as it keeps track of what queries were previously searched and clicked, and which led to a purchase.

There can be other factors that determine exactly which suggestions appear, and in what order, like the frequency certain terms are entered or click-through rates of individual suggestions.

Autocomplete Search takes previous user behavior into account

Autosuggest vs Autocomplete – What’s The Difference?

Autosuggest is an input field that suggests words based on what search term you have typed so far. Autocomplete is an input field (search box) that automatically completes your entry for you, based on its own internal dictionary.

Chapter 3

Achieve Better Sales via More Effective Autocomplete & Search Suggestions

There are many advantages of using autocomplete suggestion. If we want to summarize it, it helps shoppers make successful searches.

Here are some additional ways it improves the customer experience in your store:

  • It decreases search time, as the user is presented with relevant suggestions and quickly arrives at what they are searching for.
  • It makes the users more confident in the search and encourages them to add more details, leading to more precise matches.
  • It decreases the number of times a user arrives to a 0 result page, which provides a better experience overall. If relevant queries that are connected to existing product pages,are suggested, it effectively guarantees a successful search.
  • This also leads to the reduction in the number of people leaving your site, which means you have more time and opportunities to convert visitors to buyers.
  • It also educates potential customers about your product range, working as kind of a soft cross- or upsell opportunity.
List of autocomplete features that improve better search and UX

In short: by making search faster and ensuring relevant results, autocomplete will decrease exit rate, increase conversion rate and likely even your average order value.

We know from the B2C Retail Benchmark Report that

“Conversion rates are significantly higher where consumers have higher intent, i.e. they are searching for products.”

Now that we got through the basics, let’s dive into actual best practices to ensure those results we just talked about.

We included 10 of these with detailed descriptions here – let’s start with…

Chapter 4

10 Search Autocomplete Strategies

1. How Ranking the Suggestions Should Work

Even with the best autocomplete functionality, you have a very limited number of chances to show the user the right queries, which makes ranking essential.

As we mentioned before, there can be many factors influencing the ranking of suggestions. Based on user behavior you can show popular searches (queries that entered most frequently) first.

You can also opt for ranking queries that are more frequently purchased. Or you can rank queries and products related to ongoing promotions or special offers first. (Learn more about digital merchandising strategies if you want to boost your sales with product promotion.)

Prefixbox’s Re-Ranking AI can take weight off your shoulders, and automatically ensure that the most relevant results appear first for search queries, by taking numerous kinds of historical data into account. Check our video on the importance of dynamic re-ranking as well:

2. Personalization Makes Autocomplete Predictions More Effective

There are three basic ways you can personalize the autocomplete suggestion in on-site search:

  • Consider the location of the customer, and show them queries that are popular in their area – or exclude ones that are irrelevant.
  • Consider the language: if your online store is multilingual, show each user suggestions in their preferred language for a better user experience.
  • Include their search history: make suggestions that are relevant to what they previously searched for on the site.

3. Keep Search Suggestions Simple and Few

Your suggestions should not extend the available space or clutter it – which is even more important on a mobile screen, considering that keyboards usually take up 30% of the screen.

a. Keep the Autocomplete List Manageable

The best practice is keeping your suggestions at 10 items or less (and this is why raking them correctly is essential).

If your suggestion list is longer than that, a number of unpleasant things can happen:

  • It makes the search time longer, as the user scrolls through them.
    Suggestions that appear off-screen may be ignored altogether.
  • The paradox of choice might kick in, leading to choice paralysis. This basically means that if we are presented with too many options, our brains often choose to just opt out entirely instead of wasting energy on weighing all of them.
  • On a mobile screen, the preferred number of suggestions is 4-8, and may even be less if they include not only queries, but precise products with photos and descriptions.

If you want to display more keywords and products, consider a 2-column layout instead.

b. Avoid Scroll Bars

If your suggestions extend into an area accessible only by scrolling, again, there are many problems that can arise.

First, these initially hidden suggestions are likely to be ignored, but if they are not, search time increases again.

Also requiring an additional task from the user also deteriorates the experience. As does the fact that they can’t get a quick overview of their choices right away.

A search bar in the suggestion field can also cause technical problems with design, but we are not going to detail those here.

c. Reduce the Visual Noise

With modern on-site search solutions, the suggestion field can be filled with a large variety of elements like text, prices, photos, short descriptions etc.

While these can help shoppers, be careful not to include too many additional elements as it can drive the focus away from the actual suggestions and confuse the customers more than it helps them.

Our suggestion is to keep the visual noise to a minimum. Include both keyword and product suggestions, prices, and photos where relevant.

4. Highlighting Autocomplete Suggestions

Highlighting certain elements of your predictive suggestions helps the user keep their focus and use the feature more naturally. Here is what to keep an eye on.

a. Highlight the Differences

Instead of highlighting (which usually means bolding) what the customer already typed in, it’s much more effective to instead highlight the predictive part of the suggestions.

This way they can focus on that more easily and determine the difference between suggestions, leading them to a faster decision.

Don't highlight what customer already typed in, highlight the predictive part instead for better UX

b. Highlight the Active Suggestion

When choosing between suggestions, you should clearly indicate which one the users’ mouse is hovering over or which one is active for shoppers using keyboard navigation (which must be supported in your autocomplete feature).

This provides clarity and helps in eliminating mistakes: choosing the wrong suggestion and having to go back.

Also, the active suggestions should be copied into the search bar. This helps users understand how the autocomplete box works and makes it possible to expand upon the suggestion by providing more details, which leads to more precise results.

Highlighting the active suggestion is usually best done with a simple background shading.

c. Style Different Search Suggestions Differently

As we have discussed, multiple different suggestion types may appear in your field, and you must help the user understand the difference between them.

Let’s say that besides predictive queries you also include products and/or product categories. If you group these together, make sure to differentiate them. For example, style the text a differently. e.g. giving it another color.

If all different types of suggestions are displayed as the same, the users might not understand the difference, and ignore them or choose ones that aren’t really relevant.

Small changes in style make it easier for the user to scan the presented suggestions, and focus on those they are interested in.

Help search users to understand the difference between the search results for example with styling and different colors.

d. Style for Readability

Especially on mobile devices, it’s very important shoppers can actually read the suggestions, and also easily select the ones they are interested in.

Keeping readability in mind, the suggestions should be presented in a large enough font size, and with enough spacing, maybe even with separators, so that tapping on them doesn’t lead to accidentally selecting another option.

5. Provide Clear Instructions

The exact functionality of an autocomplete feature may not be completely clear to users right away, especially considering how different on-site search solutions can be.

In order to help them use the feature, providing instructions and labels can be greatly useful. These may include headings in the list, e.g. separating “Search suggestions”, “Categories”, “Articles” and so on.

Provide instructions and labels in search autocomplete, like "categories" and "articles".

This will help users understand how the list is structured, scan it more easily and direct their attention to the ones that are relevant to them. Without clarifying these details, mixing the different suggestions into one list, the user may pick an article instead of a product, or go to a category page instead of a more general results page despite their initial intention.


6. Visual Focus and Simplicity

When the customer is using the search feature, the autocomplete field together with the search box should be given absolute priority in terms of visual attention.

a. Design for Visual Depth

Giving the autocomplete field priority can be easily achieved with darkening the rest of the site – the background in this case.

Help search user to focus on search results instead of other parts of the online store.

This will help fade the elements on the site that are fighting for shopper attention – CTA buttons, banners, product photos and so on. This way, the customer can easily keep their focus and not get distracted.

b. Reduce Visual Competition

On mobile, directive elements like navigation or shortcuts may appear beside or even above the autocomplete field, which actually ends up making navigation problematic and distracting.

Be careful where you place your live chat option, an icon for the shopping cart, or even sticky headers, to make sure it doesn’t detract from the search experience.

By minimizing these distracting elements, you can minimize mis-clicks and provide a much smoother experience.

7. Support Both Mouse Interaction and Keyboard Navigation

Customers should be able to see which suggestion they are hovering over. This can be done by highlighting the given row. You may also invoke the hand cursor to make it clear that they can click the suggestions and it will take them to a result page.

It’s important to provide keyboard navigation (especially since we mostly use Google for off-site search, and it has trained us to use this functionality).

Using the up and down arrows, customers should be able to switch between the suggestions, and select one by hitting Enter.

8. Mobile Specific Optimizations

There are a few things that are very, very important to keep an eye on when designing autocomplete for a small screen.

Here are the most important ones:

a. Use Text Wrapping

We already mentioned how you shouldn’t try to expand your suggestion field with a vertical scroll bar. We also recommend not using a horizontal one. Many suggestions, including long or multiple keywords, won’t be able to fit in their row, given the limited screen size and that you must use a big enough font for readability.

If you just shorten the suggestion by including “…” at the end, you confuse the customer who won’t know exactly what they might be clicking on as part of the info is missing.

So how do you solve this problem?

Use text wrapping and expand suggestions to multiple rows as needed, even if this means fewer will be visible.

Providing adequate information before the customer has to commit to a click is essential.

b. Partially Obscuring the Last Visible Suggestion

On mobile, scrolling is unavoidable in many cases, especially if you have to wrap search suggestions.

Instead of adding a scrollbar, there are a few other things you can do.

The best one is probably partially obscuring the last result – which is a clear indication that the list continues below.

Now, considering the different screen sizes, browsers, fonts etc. it’s nearly impossible to get this right every single time because there are just too many variables.

So what you might want to do is check your analytics, find out which devices are used the most to access your store, and optimize for those.

Most users will still be selecting one of the suggestions that are visible right away, but with this method, you can provide a smooth experience for those who want to explore the whole list.

c. Make it Easy to Exit Autocomplete (and Remove the Search Input)

If a customer decides they no longer want to use the search bar, and wants to return to browsing the site, it should be easy for them to do that.

Providing an “X” icon to delete the query and search bar suggestions instead of asking shoppers to do it manually, is a handy solution to enhancing shoppers browsing journey.

9. Provide Category Search Suggestions

When a customer starts to type a query in the search bar, they might just want to explore what you have in a given product category.

Providing them with category suggestions is a very convenient way to enable them to browse your products and find the one they want to buy.

This saves time for the customer and provides a clear path, making the user experience smoother.

However, you want to avoid visual clutter and confusion – as we have discussed before – clearly indicate that category suggestions are not keyword suggestions. Do this with a simple heading and/or by using a different style.

10. Speed is Essential: Real-Time Autocomplete

Your autocomplete should always provide suggestions in real time.

If it’s slower than that, it can be downright irritating for the user, as they can visually perceive the lag – like with a website slowly loading individual elements.

If the autocomplete is not real-time, all of the above best practices are worthless as no one will be using a search bar they perceive as not useful.

Suggestions have to appear right when the first character of the query is entered, and should change with every subsequent keystroke to be updated to show relevant options.


Providing the highest level of quality during a customer’s journey will help you directly increase your store revenue. A better search experience leads to higher satisfaction, better experience, and overall higher conversion rates and average order value.

By using the best practices in your store, you can achieve exactly that.

Provide the highest level of quality during customer's journey

Remember the most basic guidelines:

  • Autocomplete should work in real-time.
  • It should provide relevant results – which can be personalized.
  • It should provide clear visual guidance as to how to use it.
  • If it’s in use, all other site elements should be faded into the background.

Of course it can quickly get complicated, and fine-tuning can be time-consuming to keep up with. This is why, to achieve the best results, we recommend working with on-site search experts.

As always, if you have any questions, or suggestions for the expansion of this list, reach out to us!

Learn more about the most important search autocomplete features to decrease zero-result pages by 31% and increase cart value by 19%.

Paige TyrrellHead of Marketing – Prefixbox

Paige is the Head of Marketing at Prefixbox, a leading eCommerce site search solution. She’s an American who’s been living in Budapest since 2017 and loves giving #alwayslearning sessions to help people optimize their online stores.

How Data Integration Can Aid In Hyper-Personalized Ecommerce Site Search

How Data Integration Can Aid In Hyper-Personalized Ecommerce Site Search

In 2020 having a hyper-personalized site search is key for any eCommerce business. On its most basic level hyper personalization is defined by Business2Community as “the use of data to provide a more personalized and targeted products, services and content.

In today’s fast-paced and increasingly competitive eCommerce industry, having a strong targeted personalized site search will allow you to reach your customer base quickly and show them exactly what they are looking for.

Personalized eCommerce Site Search

But that just isn’t enough in today’s world, Merchants who are truly successful in the space are using data gained through personalization and using data integration to automate their applications together in a future-proof headless environment.

Let’s take a deeper dive into what types of hyper-personalization are available and how data integration in a headless environment can level up your eCommerce performance and compete with the best.

What is Hyper-Personalized Ecommerce Search?

With all ecommerce efforts, whether that be improving website design, personalization of eCommerce site search and even data integration of applications – the goal is to make business processes easier and efficient for Merchants and improve the overall customer experience. Some examples of website personalization that can help focus your customer experience journey can include increasing metrics like average order value, bounce rate, conversion rates, customer retention rates and overall profitability.

personalized customer experience journey in eCommerce

Shoppers increasingly expect merchants to deliver contextually-relevant experiences that take into account their past interactions with the brand. Every interaction a customer has with your brand is a data point that can be used to improve their customer experience. 74% of customers feel frustrated when website content is not personalized. Customers expect to be made to feel special and have thoughtful interactions with Brands.

Given that statistic, it’s clear that adding in site personalization applications like Nosto or Klaviyo to your eCommerce application mix will allow you to adapt product recommendations to every individual customer behavior level and create unique customer experiences for every customer and increase conversion rates, average order value etc.

How To Future Proof: Hyper-Personalization and Headless Commerce

There has been continuous focus on Headless Commerce this past year. Touted as the next stage in e-commerce, Headless Commerce is in essentials; API-driven. The main function is to remove the content management layer from the front-end of a website and it’s checkout. The content management system thus becomes hyper-personalized.

API-driven Headless Commerce

Having a headless system allows you to individualize every aspect of the content and products on the front-end of the site. This includes focused site search that can remind a previous customer of a past purchase, showing customers detailed relevant products similar to their past purchases as well as improving average cart value with individualized add-on products options.

Benefits of A Strategy Driven Hyper-Personalized Site

These are just a few simple examples of what is possible with a well rounded, strategy driven hyper-personalized site focused on customer experience. Remember,

People don’t buy objects, they buy experiences. Especially now because no one is shopping in-store anymore. Personalizing the experience, being everywhere your customer shops online from marketplaces, to your Shopify store is now more important than ever.

Merchants that are winning at the game of eCommerce are future-proofing their businesses by implementing headless technology that allows them to have a hyper-connected customer through implementing individualized customer experience across their e-commerce shopping journey.

Headless Commerce: The Future of Commerce

VL OMNI has written extensively on Headless in their latest Guide: Headless Commerce: The Complete Integration Guide This integration-first ebook looks at what it will take to make headless commerce a reality for many. From different ways to approach headless commerce, to major issues and implications a headless future will have on the ecommerce and commerce industries, readers will come away with a true technical understanding of what it takes to become headless.

Data Integration: Agile and Scalable Automation

So, with a hyper-personalized website and marketing processes, combined with headless technology that allows you to efficiently and effectively individualize products, orders and the overall customer experience for each customer. How on earth do you tie it all together?

Data integration and automation between applications is the easiest and most effective process for businesses at scale to eliminate costly and time consuming manual data entry and manual business processes.

But what is Data Integration and why is it important?

Data integration connects disparate applications together, allowing for consistent automated movement and delivery of data to create business processes across a wide range of applications.

Data integration plays an important role in the world of eCommerce because it provides users with a real-time outlook of business performance and transforms data into valuable, usable information by combining it from different sources.

As commerce in the physical world and online continues to evolve, your understanding of data integration can help inform your business’ ability to scale and build a more robust future for itself.

Data integration between different applications can be created in an agile and scalable manner, allowing you to automate hyper-personalized processes and individualized products at scale.

eCommerce Data integration between different applications

This allows you to focus on what’s most important to Merchants: creating a great customer experience. Data automation’s real value is connecting disparate systems together to eliminate manual processes.

In a hyper-personalized headless environment that can mean integrating your eCommerce store with your CRM, ERP, loyalty or personalization program.

This allows all the order and customer data collected from your eCommerce store, for example Shopify to be integrated into your CRM, allowing you to keep track of customers personal information and ordering habits.

This information can then be integrated into the content layer of a website in a headless environment to get specific personalized information and individualized offers to each and every valuable customer.


Hyper-Personalization in Site Search: The Role of Data Integration

Those who are succeeding in the ecommerce industry today are those who look at every aspect of their business, including their applications and business processes and look at how they can leverage all of their data into providing a better personalized experience for their customers.

Creating a hyper-personalized experience starts with the back-end data. The way that data is used and leveraged into providing valuable insights for your business will inform how successful your personalization targets perform.

By automating processes and integrating all applications together, your business can create a full picture of each customer, their habits and their shopping persona, allowing you to make educated decisions about how to personalize your website and site search for them. Ecommerce growth is truly open to those Merchants that take their data into their own hands and with it, create the best personalized shopping experience possible.

Vanessa MatosVanessa Matos, Marketing at VL OMNI

Vanessa is the marketing Coordinator at VL OMNI, a serverless platform for agile and scalable iPaaS ecommerce integration. From ecommerce platforms, ERPs, EDI, or marketplaces, we integrate your applications together to create a seamless experience at the operational level and enhance your business’ customer experience.”

What is a Search Index and How Does It Generate Revenue for eCommerce Businesses?

What is a Search Index and How Does It Generate Revenue for eCommerce Businesses?

No modern search engine can work without a proper search index. Everywhere, but on the smallest of sites, running a search would take seconds, minutes or longer, without an index. It varies based on how many pages the search solution would have to crawl.

This is absurd and would lead to huge drops in your revenue, as a 2 second delay in load time results in an 87% abandonment rate and even a 1 second delay instantly drops customer satisfaction by 16%.

eCommerce Search Index Revenue Generator Guide

So, let’s talk about search indexes and how we create them, so split-second searches are possible.


What is a Search Index?

A search index is a regularly, and automatically, updated database of all your products and related information in a simple form. It’s something that enables users on a site to perform fast and accurate searches.

It makes it possible for the search engine to perform any search in microseconds, because it doesn’t have to crawl hundreds or thousands of pages every time, it only has to search in the index.

What is a Search Index > Web Crawling > Search Index > Search Engine

It’s basically like an index at the end of the book where you can find the most important terms and phrases with the indicated pages where you can learn more about them.

What is a Search Algorithm?

A search algorithm is used in searching for information on the Internet. There are two types of it: deterministic (or exact) and probabilistic (or approximate). Deterministic algorithms return an answer immediately; they do not require any probability calculations. Probabilistic algorithms use some form of calculation to determine how likely a particular result is correct.

Why Search is important in eCommerce – and why is it hard?

In short: because it’s very complex. You have to pay attention to a lot of details: things like providing the right attributes and tags, the natural language your customers use, seasonality, and the design of your product pages. There are many eCommerce search best practices you can and should try out, and fine-tuning is a never ending task.

There is a massive thought process behind every single purchase decision, even if most of it is subconscious, and there are so many little details on your site that can alienate a possible customer.

The user experience on your site is integral to the overall customer experience – and thus your revenue. Note that the conversion rate for visitors who use your on-site search is 1.8 times higher than the average – meaning they provide more revenue with every purchase.

And this isn’t possible without a proper search index.

Good to know: Searching for business data within a company is very similar to eCommerce search processes, however it requires a completely different approach and solution, so it’s worth getting a line in advance on how to implement effective enterprise search in practice.”

Creation of a Search Index

In the case of books, indexes are created manually, usually by the authors and editors, which makes sense: the information included on a couple hundred pages is comprehensible for a single human, especially if they compiled it in the first place.

Now, if we are talking about websites, especially eCommerce websites, where it is not uncommon that thousands of pages are created for different products, attributes, content and so on – it’s essential that this indexing work is automated.

So, online search indexes are created using algorithms. Crawlers (or web spiders, as they are also known), which are basically automated algorithms, visit every page of the website regularly, scanning and collecting the information on every single one and assigning those to a simple database.

How Indexes Return Search Results

This is (or seems to be) a fairly easy process, in its essence, it isn’t very different from when you hit Ctrl+F in a document or on a website to find a certain word.

Of course, there is more to it, as modern site search engines have to account for typos, misspellings, relevant keywords based on semantics and user behavior, natural language and so on.

Of course, there is more to it, as modern site search engines have to account for typos, misspellings, relevant keywords based on semantics and user behavior, natural language and so on.

But these algorithms sort themselves out, and essentially as a user, you only see the results after you hit search on your query. These results usually include product names or content titles, some attributes or an excerpt, a picture and a price.

The Benefits of a Search Index for Your eCommerce Business

It will improve the user experience and click-through rates of your site search.

First of all, the faster you return relevant results derived from a proper search index, the more satisfied your customers will be.  Today, people expect instant results when typing a search query. A search index allows the search query to be processed without using resources from the server and therefore produces faster results.

And secondly, you can promote search results based on certain factors like relevancy, freshness, ongoing promotions and so on. If you want to learn more about this, have a look at our article on Searchandising Strategies.

Let’s have a look at some basic methods on…

Managing and Adjusting Site Search Results and Ranking

This enables you to define certain parameters, which influences the results shown to your customers when they search.

Utilize Searchandising to Show what is Important

Searchandising in itself is a vast topic, so you should really give a look at our article on it (link in the previous paragraph).

Utilize Searchandising

To summarize and simplify, you can boost certain products to appear at the top of the SERP based on how recently they were uploaded, how many are in stock, if there are any promotions connected to them, if there are available bundle offers that are relevant to the query, or any other criteria you want.

Guide Customers with Pinned Search Results

You can even temporarily, or permanently, pin certain results on the top of result pages – like offers in your holiday campaigns or evergreen content

Pinned Search Results

Customers won’t notice the pin as it appears as normal results, however it may become apparent if they run more searches.

Promote Certain Site Areas

You can also direct your customers to certain areas of your site, for example if you want to highlight a certain type of content like holiday gifts or products from a certain category before others.

Personalized Search Results

A big part of enhancing your user experience is giving everyone what they personally need the most. Since we can’t read their minds, we have to prioritize search results for them based on their previous behavior on our site: the products they looked at, the searches they performed, and the purchases they made.

Personalized Search Results

If you provide user-specific results, you can improve overall click-through and conversion rates together with overall user experience.

Analytics: Track Customers and Learn From Them

If you track the data about how people use your site search engine, and fine-tune results accordingly, you have a better chance of giving them what they actually want.

The main metrics to track are:


Search UI and API

A search index, as we clarified, is the memory bank behind the site search function, which improves your user experience on its own.

But you can, in many ways, design your own search UI. For example, you can make sure that your 0 result pages are built in a way that encourages users to keep shopping, or you can use badges on certain results to highlight discounts.

But you can, in many ways, design your own search UI. For example, you can make sure that your 0 result pages are built in a way that encourages users to keep shopping, or you can use badges on certain results to highlight discounts.

It’s important that your search index is updated regularly and indicates the most recent results. You can do this by including an indexing API in the crawler.

Building a Product Database for a Well-functioning Search Index

The crawlers need data to work from, so it’s your (and your developers’) responsibility to organize your existing data in a way that the bots will understand and be able to translate to results as fast as possible.

This is why you should pay attention to…

Cleaning Your Data

Creating clean data for the search index (data cleansing) is the process of making sure that your data is actually valid. This can become especially problematic if you have a great number of products and the data that you get from the manufacturer is not compatible with your database.

If there is no regular cleansing process, you can end up with prices that are off by magnitudes or invalid values for product dimensions, among many other things.

Use Convertible Units of Measurement

Your data has to be uniform and in convertible measures, like from centimeters to millimeters, or centimeters to inches.

Imagine that you get screen sizes for TVs or smartphones from the manufacturer in millimeters, but on your product page you have to display those values in inches, because your customers are used to that measurement with these products. By having uniform and easily convertible units of measurement, you save yourself a lot of trouble.

Include Inventory Stock Data

Your stock should be continuously updated in your online store for multiple reasons. This helps you avoid situations in which you’re actively selling a product you don’t have in stock, which means your customers would have to cancel their orders. Or the flip side, your inventory could be piling up, but no orders are coming through.

Stock data can also indicate whether you should give more prominence to a product in the search results via Searchandising or push it back when stocks are low.

Consider the Product Variants

Your search index must be capable of handling different variants.. There are many instances in which you’re selling products that are similar, but have slight variations – shirts with different size and color, phone cases with different art, smartphones with different memory sizes, etc.

Product Variants handling in eCommerce Search

These variants have a tendency to appear very quickly – let’s stay with shirts for a moment and say that you have a T-shirt in 4 colors and 4 sizes. That is already 16 variants for the same product. So make sure that everything is in order in your database – and test it in your search solution, to ensure shoppers can find the exact product they want as easily, and quickly, as possible.

Deduplicate Products

If there are hundreds, or thousands, of products in your store, which is common, especially if you take product variations into account (e.g. different sizes, colors), it is very likely that duplications will occur. These sometimes will go unnoticed, for example if the same product is uploaded into different categories.

These duplications need to be found and eliminated. This is most easily done by running an automated algorithm to check photos, product descriptions and attributes.

Support Multi-language eCommerce

Finally, if your target market is wider than one country, you should have dedicated fields in your product database for every language and pay close attention to filling them in correctly. This enables each of your customers to use their own language and find the products they’re looking for.

Also look for an enterprise ecommerce search provider that has a built-in language identification feature.

What to Pay Attention When Building Your On-Site Search

Search Bar and What You Indicate With it

When considering your search bar, the thing you want to pay attention to is visibility: users should be able find it right away and understand how to use it.

This is why it can be beneficial to put text in the bar itself. For example, some stores allow people to search by article number, but in others this returns no results. This is why you have to let shoppers know how they can use your search bar.

Two things to note here:

  • If you have some kind of search criteria that is impossible to translate to relevant results, let shoppers know this immediately.
  • Include every important attribute in your search index, so you don’t have to steer shoppers away from certain search criteria.

Learn more about the most important search autocomplete features to decrease zero-result pages by 31% and increase cart value by 19%.

The Redirect Function

Your search solution could include a redirect function where certain queries take the user to a landing page or category page directly.

Without getting into the technical details, this is something your developers would have to look into – but determining the pages where you want to redirect users is yours.

Search-as-you-type Function

If your index is built the right way, the site search solution should be able to suggest queries in real-time – you can see this every time you use Google. Their solution is a mix of frequently used terms (user behavior) and terms that show up frequently in the index. You should follow a similar path.

Title of your sub-chapter

Based on user behavior and some custom product tags, you can provide cross/upsell opportunities on your result pages.

You could order results in a way that only a certain number of results are directly related to the query itself, and the rest are related products or bundled offers.

Faceted Search in Your Store

In order for faceted search to work properly, this detailed product database is essential, because results are edited in real-time based on the filters selected (which also need to update based on available products). These search filter options are based on the various detailed attributes in your database.

Faceted Search in Your Store

We could go on for an entire article about this topic, and we already have: if you are interested about it, check out our guide on faceted search!

Your Product Pages are Important

The way you structure your product pages greatly impacts how effectively the crawler is able to gather information. To get the best results, your product pages should be uniform and fairly simple when displaying details and data like sizes, colors, etc.

Use Analytics to Learn About Your Search Index Performance

Analytics can say a lot about the health of your search index. Some of the basic metrics that can indicate a problem are:

  • Loading time of your result pages
  • The percentage of no result pages
  • Conversion rates of your result pages
Search Index Health Metrics


As you can see, it’s important that your search index stays up-to-date and is full of well-structured data in order to work properly. When you have a high-quality search index, you can provide the best possible search and customer experience via providing fast results.

A lot of what goes into a great search engine is very, very technical and complex – but some decisions cannot be made by developers alone. This is why it is important to always work together across units – and for you to pay close attention to the details.

  • Make sure your databases are compatible with each other
  • Make sure all information and data is up-to-date
  • Create pages where a crawler can easily find information
  • Design your result pages so they don’t only load faster, but also provide more opportunities for conversion

And you won’t have any problems!

Balazs Vekony profile picture - Prefixbox
Balazs VekonyOnline Marketing Manager – Prefixbox

Balazs is an Online Marketing Manager at Prefixbox, a leading AI search solution provider for E-commerce sites. He is a marketing enthusiast from Budapest, who is interested in new technologies and solutions and believes in the power of search.

13 Searchandising Strategies To Generate More Revenue For Your eCommerce Business

13 Searchandising Strategies To Generate More Revenue For Your eCommerce Business

Promoting products offline is a common and successful business practice.

But for online businesses, this is often neglected, because it can be difficult to help shoppers navigate to promotions. This guide will teach you how to effectively leverage searchandising on your webshop to increase revenue.

You’ll find out:

  • How to transform your result pages to increase your revenue,
  • How you can offer exact products in real-time, like autocomplete,
  • And which factors should impact your searchandising strategy.

Ready to start?

13 Searchandising Strategies for eCommerce

Think about how often people use the search bar on a site, be it a news site, an online store, a social platform or a message board. I bet nearly every visit.

As Econsultancy puts it: 

“Typically, up to 30% of visitors will use the site search box, and each of these users is showing a possible intent to purchase by entering product names or codes.“

This means that your search can make or break the success of your business. Your search function gives you a great opportunity to apply marketing and sales tactics to your search pages. 

Chapter 1

What is Searchandising?

In traditional brick-and-mortar stores, merchandising is basically the practice of guiding the shoppers’ attention to the offers and products they want them to notice.

This can be done in a large number of ways – hanging signs, rearranging the products on the shelf or the entire store, putting price tags with different colors on items that have a discount, and so on.

In eCommerce, there are many different ways to do this: putting large CTAs on your site, colorful offers above-the-fold on your main page, pop-up windows, and so on.

Where eCommerce merchandising becomes difficult is on the search page. Banners or pop-ups are not going to be very effective if a customer has a specific product in mind and is actively using the search bar to find it.

And this is where the power of search + merchandising, searchandising for short, comes into play.

As with every online marketing and sales process, searchandising isn’t a single step, but a whole process.

Let’s look at it now.

Chapter 2

The Process of Effective Searchandising

There are three very basic golden rules of searchandising.

  1. Understand your business’ basic priorities and goals. This will help you choose from the abundance of searchandising strategies available, many of which we describe in this article – and their potential benefits for your business.
  2. Have your KPIs ready. As with all business decisions, you should know what metrics you want to use to measure your success or failure. When using searchandising, click-through rate, conversion rate, average order value can all be important. But also, to better define what products and offers to showcase, you have to know your profit margins, inventory, and availability.
  3. Implement, watch closely, monitor, and refine. Monitor the performance of the rules you set, and fine-tune your strategy as you go. Incorporate searchandising analytics and decisions in your regular meetings and brainstormings, and define intervals for changing up things.

Does this all sound a little vague? Don’t worry.

Here are 13 very specific things you can do…

Chapter 3

Searchandising Strategies

In this section, we’ll show you some specific ways to highlight promotional offers to your customers, while maintaining their sense of control over the search process.

Keep in mind: the goal is not to bombard the customer with offers until they give up and convert. It’s to find the best offer for them by paying attention to what they’re doing, and what they’re searching for. 

Giving them what they want (even when they don’t know what that is), helps create grateful customers who are likely to return to your store for future purchases.

To do this, the first thing you want to do is…

1. Boosting Products: new arrivals, on-brand products

Simply placing certain products on the top of the results page can be successful, but it’s important you make sure it benefits your customers. 

When selecting products to move to the top of the results page, make sure they’re relevant, popular and are something your customer wants to buy.

This is a great thing to do with new models or products. 

Think about this: by searching for something specific, shoppers might miss a better, upgraded version, or they might not know a new product is available, so when you highlight a relevant, related product the benefit is clear.

If you have your own brand products, you might also want to showcase these first. 

We recommend you do this when these products are already preferred by your customers, validated by positive feedback, and you have clearly defined unique benefits that the competition doesn’t.

2. Custom collections and bundles

A great way of creating a cohesive shopping experience  while guiding your customers toward different products, is to not just place solitary products in front of them.

Creating custom collection offers can be particularly effective during holidays and special events. 

To do this, pick items that are especially popular during these periods or are frequently purchased together..

Then create Christmas bundles, Super Bowl offers or anything you want.Display these when your user includes the relevant keywords in their search query (like “christmas”). By displaying these bundles beside the product results, you retain relevance while guiding shoppers’ attention toward other products you can upsell and cross sell.

3. Boosting your highest-margin products

It makes sense to tweak your results pages in a way so that your best high-margin products are displayed first – as products at the top of the page are purchased more frequently. 

If you decide to boost products this way, make sure to monitor user behavior as this method can backfire. Keep an eye on your click and conversion rates, because altering results in this way can create a negative search experience and hurt your business in the long term.

4. Embedded Product Listings (Autocomplete on Steroids)

Suggesting keywords and phrases that your customer might be interested in, while they are still typing, is a great way to increase conversions.

Why not take it a step further, and display products that best match the query right there in the suggestions field too?

You can often do this through searchandising functionality.Like faceted search, this is a real-time solution that can greatly reduce a shopper’s path-to-purchase, which improves their overall experience.

ikea site search autocomplete feature

Learn more about the most important search autocomplete features to decrease zero-result pages by 31% and increase cart value by 19%.

5. Promotional Banners Near the Search Bar

Banners may not be as popular as they once were, but if used correctly, they can still be powerful tools in increasing conversion rates.

Placing banner near the search bar and at the top of the results page effectively directs your customers’ attention to the categories, products, or discounts you want them to check out. Faceted Filter function

6. Promotional Badges

Adding badges is a great way to communicate discounts to customers without interrupting their shopping experience. 

It’s one of the basic and most effective merchandising methods because you don’t have to alter the search rankings in any way.

You can use badges to announce: 

  • New product arrivals
  • Upgrades or new available products (e.g. new colors or sizes for clothes)
  • Available-in-bundle offers
  • Limited-time discounts
  • Last-chance products or sales

For a simple example, look at how Google clearly indicates which pictures are licensable when you perform an image search.

7. Optimizing and Re-ranking Product Results

When your eCommerce store has many products, most search queries yield dozens or even hundreds of equally relevant results

In order to effectively guide shoppers to the products they’re looking to buy, the next thing you should do is set up additional ranking rules to display the most popular and relevant products first.  

However, with searchandising, you can edit these ranking rules any way you want.

For example, you can choose to display relevant products that are on sale at the top of the search results page, in order to quickly clear out your stock. 

Or, you could opt to show your own brand of products first. Whatever you decide to do though, make sure that you keep relevancy and popularity as important criteria, so you don’t disrupt the shopping experience too much.

Learn more about the most important eCommerce search engine features to increase conversion rate by 15% & revenue by 47%.

8. Personalized Result Optimization

Continuing that thought, what you should also take into account with ranking is the behavior of the given customer on your site.

If the customer has searched for men’s sneakers before, you could direct them to other men’s sportswear. 

Try to avoid the mistake that many retargeting campaigns make and don’t offer the customers the same exact thing that they previously bought.

This way, you provide a better shopping experience to the customer while increasing your cross-sell and up-sell opportunities.

9. Personalize Your Category Pages

Category pages are useful for shoppers who have some kind of idea of what they’re looking to buy, but aren’t looking for an exact product. 

By adding elements of personalization to these pages, you can quickly increase your conversion rate

As a pro move, personalize these pages based on personal preferences, behavior and their search intent – these yield the most relevant results.


10. Boost Products Based on Availability

One of the main things to consider when boosting products is their availability.

If you have a lot of a particular product in stock and want to move them quickly, boost them on the results page. 

However, if you have new arrivals and releases that are not yet in your inventory, you could also boost them so you can achieve nice pre-sales numbers. 

This is a highly effective strategy if you are using a search that ranks products based on popularity. 

New products don’t yet have the clicks and purchases to boost them to the top of the results page, so if you know they’re going to be a hit, you can adjust the positioning to get them in front of shoppers immediately.

11. NLP Autocomplete

As we wrote in our previous article, 12 Best Machine Learning Strategies for eCommerce Businesses, natural language processing is an essential part of any site search solution. 

Here’s why:

Natural language processing and machine learning have to understand the kind of language and phrases your customers use, how often they use them, if the results are satisfactory for those phrases, and even common typos along with the correct spelling.

This way, every user can feel that the search engine is truly there to help and pays attention to the smallest details, all without being annoying.

Using the same language as your customers automatically improves any search function while providing a subconscious positive affirmation for the user about the usability and overall the customer experience of your site.

Oftentimes, you can also boost product and keyword suggestions in your autocomplete search, while taking into consideration NLP relevancy.

12. No-result Pages

Your no results pages present a huge opportunity to direct attention of the customers. 

While there might be 0 results for precisely what they’re looking for, you can use this opportunity to promote other, relevant products or things like: your current discounts and bundle offers, the most popular products (even better when personalized) or simply the products you want to move.For more advice regarding how to design no-result pages, take a moment and check out these 12 Awesome “No results found” Pages!

Sears No Search Result Found Pages

13. Updating product catalogs

Make sure your product catalogs continually stay up-to-date. Without accurate products and attributes, it’s nearly impossible to provide shoppers with a good search experience – whether or not you leverage searchandising!Look for merchant tools that allow you to easily update your product catalogs. Batch updating attributes is the easiest way to do this.

Chapter 4


Searchandising is not actually that hard to do – however, you have to know what you are doing, and why.

The key takeaways from this article if you are looking for searchandising techniques that will definitely improve your conversion rates are these:

  • Personalize: always pay attention to the history and behavior of your users, not just their location and demographics. Try to give them what they actually want, direct their attention in a way that is useful for them.
  • Use Search Rules: don’t just trust the algorithm to make the choices for you. You have your business plan, your cash flow, your inventory to consider, marketing campaigns that are set, holidays that are coming up. Based on these, make the rules yourself, and tweak the results just enough to make the most of your offerings.
  • Recommend Products: don’t be shy to suggest similar or discounted products. Put them in bundles or packages if you have to, but keep in mind that while you don’t want to be annoying, you definitely have to be a salesman. Define what and when to recommend that your customers will be grateful for.
  • Run Search Campaigns: besides your everyday sales and marketing campaigns, come up with creative ways to offer upsell or cross-sell right on your search result pages.
  • Track Data & Measure Results: once you have set the rules, start to watch them work. If they don’t, try other ones. Always base your decisions on the data that you see on your dashboards, and not gut feelings. Data-based decisions are the ones that will help you scale your business.

If you listen to the eCommerce Search Best Practices and base your decisions on the data about what your users like and what they don’t, you will increase key metrics like conversion, average order value, retention, and more.

Learn more about Search Promotions, On-site Ads, Intent Clarification, Custom Landing Page features if you want to boost your sales with product promotion.

Paige TyrrellHead of Marketing – Prefixbox

Paige is the Head of Marketing at Prefixbox, a leading eCommerce site search solution. She’s an American who’s been living in Budapest since 2017 and loves giving #alwayslearning sessions to help people optimize their online stores.

11 Features To Look For When Choosing An Enterprise eCommerce Search Provider

11 Features to Look for When Choosing an Enterprise eCommerce Search Provider

Site search is easily the most important function of any online store.

Apart from determining a huge part of the user experience, providing data about shoppers, it also increases your online revenue.

However, with so many providers and functionalities, it can be difficult to choose the best enterprise eCommerce search provider.

This guide will explain:

  • The most important capabilities in a site search engine
  • List the pros and cons of building vs. buying
  • How to choose the best eCommerce search provider

Ready to check it out?

Enterprise eCommerce Search Provider must-have features


Chapter 1

The Foundations of Great E-Commerce Site Search Engine

Chapter 2

Must-Have Capabilities for Your Site Search Tool

Chapter 3

In-house search engine development or SaaS vendor solution?

Chapter 1

The Foundations of Great Enterprise eCommerce Search Engine

Providing Relevant Results

It may seem obvious, and even an easy task, that a search engine should provide results relevant to an executed query. This is because we are accustomed to Google, which we use daily, and which is top-notch in relevancy. Google has universally set the standard of what a search function should accomplish. 

But keep in mind that Google uses an extremely complicated, ever-expanding and changing algorithm to be able to do that, backed by hundreds of developers constantly fine-tuning it – and it took them literal decades to arrive at today’s model.

Take a moment to think about what “relevant search results” actually means. For example, let’s say a shopper enters the query “cooking book” on Amazon.

The results they get are not simply based on matching keywords in the title or description of the product, because there are other important factors. Like shipping logistics and/or availability: they might get the best cooking book as a result, but how is it relevant, if the product does not ship to where they are, so they can’t actually buy it?

What if there is a typo in the query? The shopper is still going to expect relevant results, and will only get frustrated with a “no results” page and the fact that they have to repeat the search.

Then there is the question of how do you weigh the importance of product attributes – hint: if the keywords appear in a title, it should be more relevant than if it’s only included in tags, but shipping to the shopper’s location is even more important, and so on.

As you can see, relevancy is a much more complicated question, especially if you have hundreds or thousands of products in your online store.

Reliability, Uptime and Speed

On the technical side, you also have to make sure that your search engine actually works: shoppers can run queries anytime, from any part of the world, and they get the results almost instantly. 

The average Google search, again, is a standard that we are used to, and most times it takes only a fraction of a second to complete. Even if you can’t compete with them, you should use an enterprise eCommerce search engine provider and hosting solution that makes it possible for your shoppers to get to the result page as soon as possible.

Chapter 2

Must-Have Capabilities for Your Enterprise eCommerce Search Engine

1. Autocomplete

One of the most basic functions that shoppers expect is Autocomplete: this predicts shopper queries once a shopper focuses in the search box, or from the first keystroke. It suggests keywords, products, and categories that yield relevant results.

Predicting shopper intent  is only one aspect: what is even more useful is that it slightly guides shoppers down their path-to-purchase by suggesting searches that won’t end in zero result pages. 

There is not much more to this, except our advice: if a site search engine lacks Autocomplete, stay far away from it.

And autocomplete should also be…

2. Typo-tolerant

There aren’t many studies out there with numbers on the average amount of spelling mistakes, even fewer focusing on search engines because search engine providers don’t usually publish this kind of data. But we can take away some interesting points from a 2006 article by W. John Wilbur, Won Kim, and Natalie Xie titled Spelling correction in the pubmed search engine.

According to their analysis, there  is a “misspelling rate” of 26% for words on an academic site. It seems possible that the misspelling rate on a public site could be even higher.

Nordlie (1999) observes that two thirds of initial requests are unsuccessful in meeting their objective and an NPD survey (2000) finds that 77% of the time an initially unsuccessful search is modified and tried again on the same site.”

A typo-rate somewhere between 1 in 4 and 3 in 4 might seem extremely high, but with keyboards, fast typing, the prominence of typo-tolerant search engines and spell checking, it certainly can be possible, because we don’t really pay attention to spelling anymore.

If your site search engine can’t tolerate typos – as in, recognize them and recommend another search, or even better, show results for the correctly spelled keyword instead of returning no results – you run the risk of driving your customer away. 

3. Managing synonyms

Another important aspect: synonyms. In many cases, your shoppers are going to refer to products or categories, even certain attributes by a different name than you do. 

For example: if they are looking for a laptop, they may write “notebook” in the search field, luminosity and brightness might get confused with LED lights, someone who wants to buy an instrument panel may search for a control board.

And if your search engine can’t understand this, shoppers are going to end up on zero result pages, even if you do actually sell the product they want to buy.

Which, of course, is bad, because your products effectively remain hidden from shoppers with strong purchase intent, and you lose money on every one of them.

We recommend you choose a solution with built-in Synonym Management

You might think it would be easy to find a solution that includes this, maybe it’s just a matter of integrating a synonym database – but actually customizing a search engine this way is very tedious work and it’s not at all fast.

Which brings us to…

4. Customization

Out-of-the-box solutions are convenient, cheap, easy and quick to implement, but rarely satisfy all needs. They aren’t all flexible in terms of customization: usually you can’t expand on the existing functionality or performance.

Meanwhile, open-source search solutions can be customized as you want, but they are also resource-demanding – implementation, development, maintenance can take a lot of time and energy.

But what is a good, middle-ground solution?

The best middle ground is usually a search-as-a-service solution, where implementation is quick, there is great room for customization, and as with SaaS services in general, maintaining the system is relatively low-cost.

Before committing to any kind of site search solution, think about your unique business processes and the needs of your shoppers. 

Can the search provider meet both of these needs?

5. Personalization

Personalizing what your users see can be a great way of not only increasing your conversion rate, but also your average order value.

A few examples how you can personalize the result page based on the past behavior of your shopper:

Show them products they viewed (and didn’t buy) earlier, similar products to those, or products matching their previous searches. This way you may guide them to additional purchases, as they may see results they didn’t stumble upon before.

Overall, personalizing the result pages based on behavior can decrease site abandonment, increase engagement and conversion rates, and improve the user experience and customer journey of your webshop.

6. Dynamic filtering

Instead of providing a static results page when the shopper executes a query, you can improve user experience and increase your conversion rate by leveraging dynamic filtering.

What this means: the search engine displays relevant results and filtering options based on the initial query. The key here is that none of the filtering options will lead shoppers to a 0-results page. 

For a ‘real life’ example, let’s say you have an online book store. When a user enters “sci-fi novel”, they are presented by a number or results, and also a few checkboxes appear where they can filter by the most relevant authors like Isaac Asimov, Frank Herbert, Iain M. Banks etc. But they won’t see J. R. R. Tolikien’s name there, as he never wrote science fiction.

This also means that someone who searches for ‘romance novel’ won’t see the same filter options as those authors didn’t write romances. 

The search engine does this by analyzing the attributes of the results and applying relevant filters, which is why this is called faceted filtering

As the shopper starts checking boxes to filter results, the irrelevant filter options disappear. 

This not only guides shoppers to the product they want to buy, but it gives them a sense of control and improves their overall experience on your online store.

7. Natural Language Processing

Natural Language Processing (NLP) is a very complex topic as it encompasses both machine learning and AI, so we are going to concentrate on what it actually does when integrated into a search engine: it enables the search engine to better understand queries by, well, processing natural language.

NLP enables the search function to provide more relevant results, offer suggestions via Autocomplete, and learn from user behaviour based on the wording and language used. 

It also makes your search engine more language-independent, as NLP is not strictly tied to a specific language.

It is a resource-intensive feature to develop, and some solutions simply sell their advanced synonym management as NLP, so keep an eye on the validity of these claims. We can’t add much to the advice Shopify gives you:

When you don’t dig into the specifics [of the NLP feature], you might only be getting a demonstration of basic functionality like synonyms with a fancier label. That’s a problem if you’re expecting an automated system with sophisticated language parsing.

8. Ranking

Make sure that you have the option to customize certain ranking factors. While relevancy is the most important, if you want to be able to provide better customer experience, it might also make sense to highlight popular products on a result page for example. 

Ranking algorithms usually incorporate relevancy and popularity scores. However, search providers have dashboards to let their customers alter the boosting parameters to sell more of what they want.
These ranking algorithms should also be refreshed at least daily to ensure results remain accurate and shoppers get the best experience possible.

9. Analytics

Your on-site search engine is one of the most valuable resources of customer data. You can learn invaluable information about them including: how they search for procuts, what wording and language they use, what they’re searching for that you don’t currently sell/have in stock, if they are successful with their searches or need help, and much more.

You have to be able to view this data in an easy way. So you should be looking for a search solution, that

  • Actually can track and store this information.
  • Has the KPIs that are the most important for you (like click rates, “no results” queries, popular product, popular searches, average number of products/purchase,  average order value, etc.)
  • Has a user-friendly dashboard. 

If you have all of these, you can refine your customer journey, improve marketing messaging  and to achieve greater conversion rates!


10. Merchandising

Merchandising is a way to effectively promote the products you want to sell. These can be new products, products with a lot of inventory, or products on sale. 

You can customize the placement of these by adding banners, editing the products on the search results page, adding redirect rules, and disambiguation tiles.

11. Indexing

Indexing is the process by which the search engine creates a database with all the available data on your products and provides results based on this database.

With indexing the key is to choose a solution with a frequency that suits your business model. Many times the frequency with which the tool indexes products and attributes depends on how much you pay for the service, the tier you are in etc.

If you offer a few products and update them rarely, lower frequency would be fine – but if you add new products daily or weekly, a solution which indexes new data monthly will simply cause you harm.

Shoppers on your site expect to find products instantly – this includes new products. Most online revenue is generated by shoppers who use the search function. If your search doesn’t have access to an updated product database, it can’t direct shoppers to the products they want to buy.  

Which means you miss out on money.

Chapter 3

In-house search engine development or SaaS vendor solution?

When it comes to implementing site search in your webshop, you have three basic options to choose from:

  • You can select an out-of-the-box solution.
  • You can decide to develop your own site search engine.
  • You can opt for a SaaS solution.

We previously mentioned the problem with the first option, out-of-the-box solutions. They can be fast, easy and cheap to implement, but in the long term, they can be temporary solutions at best, as they are not at all flexible – it’s unlikely they will fit all of your needs.

Therefore,  the rest of this article will focus on in-house development and SaaS search solutions: we’ll provide you the pros and cons of both options, so you can make an informed decision yourself.

In-house Search Development with Elasticsearch or Solr

Elasticsearch and Solr are currently the most popular open-source search engines that you can work with. 

Sidenote: don’t try to develop something from the ground up, even if you have unlimited budget. The timeline of a project like this in itself will stretch out so much that you are way better off with an open-source engine as a basis.

The timeline for in-house development

Setting up an open-source search solution (1-4 weeks)

With Elastic or Solr you are getting a basic site search solution, for which the setup will take something between a week and a month. 

However, you should know that the time frame depends on your in-house capacity: the size of the dev team, their experience with search, how much time they have to focus on the project, and , the number of products and requests your online store receives. This means that if you have a small dev team (or one focused on other products) or a lot of products and shoppers, it can take longer.

Advanced Data Processing (2-8 weeks)

Your search engine must have all the data about your products – all the attributes, values should be standardized and processed, together with sentences from product descriptions, relevant keywords have to be extracted. 

This will provide the base for your engine to work, and continuously updated in the future. The time it will take to do it the frist time will depend on the expertise of your developer(s), but usually it will be around 2-8 weeks.

Configuring the Basic Features (4-8 weeks)

Synonyms, default facets, tokenizers, default stemmers and the rest of the core features get configured in this phase that will take somewhere between 1-2 months.

Configuring the Advanced Features (20 weeks)

NLP-based suggestions and autocomplete, custom filters, contextual dynamic facets and some of the more complex features: these can take up to 20 weeks to configure.

Custom Ranking (1-4 weeks)

As we have mentioned, customizing the result pages to fit your business model and marketing efforts is an important function. 

Determining and setting the basic factors for this should take 1 to 4 weeks, depending on complexity. 

(And a few days each month for updating these values.)

Custom Search Functionality (3 weeks, at least)

Depending on what kind of extra functions you and your online store require, it will take at least 3 weeks to set these up. 

Think: search assistants, voice search, personalization of the results pages etc. 

The more you want to customize, the longer it will take – this could possibly be extended by many months.

Post-Live Launch Bug Fixes and Algorithm Improvements (3-4 months)

This is the fine-tuning phase after everything is set up – which is, of course, something that will go on as long as your site is live. The first testing, bug fixing and improving phase should last at least 3-4 months after launching your search engine.

To summarize: even with a well-staffed and experienced team of in-house developers, building your custom search engine based on an open-source solution will take you at least 1 – 1.5 years and that’s for a basic-version without analytics. If you can’t afford that timeline (and the budget it implies), then a SaaS solution of an enterprise eCommerce search provider could be more advantageous for you.

Advantages of In-House Search Engine Development


Developing your on-site search engine with your own team means that you have complete control over the process and what to include, how to build it. This allows you to tailor everything to your business needs.

Flexibility in Terms of Making Priority Bug Fixes

If you have an in-house team, priorities can be set as you want: they can get to fixing critical issues right away and you don’t have to wait on a third party, on customer service and so on.

Results Can Be Leveraged Beyond Search

Your custom-built search engine can be useful in developing other important site functions which your developers can create based on this work.

Disadvantages of In-House Search Engine Development

Team Recruitment 

If core members leave your team, crucial knowledge can be lost with them, even if all the documentation and written know-how remains. This cannot replace their experience and the fact that they have been working together as a team, and new team members always have to be onboarded, introduced to customs and policies ang it generally could take quite a long time for them to get up to speed.

Team Management 

Your search team should be an independent one from your core dev team – otherwise multiple projects running parallel will take time and capacity from each other – consequently both basic development and building/maintaining your search engine will have less progress made in the same time.

It Takes Years to Fully Develop an Enterprise eCommerce Search

No matter what you do, basic development itself will likely take 1-1.5 years – and after that, adding extra functionality, customization, bug fixes, maintenance and updates will take years until your engine is fully formed. It’s essential that your team always stays up-to-date with e-commerce search best practices and continuously implements them.

Manual Infrastructure Setup and Maintenance

With no third-party partner or SaaS provider, you have to purchase, set up and maintain all the infrastructure needed for both development (tools and devices the team will use) and operating the search engine. In terms of resources, this means additional costs and time.

Continuous Performance-improvement, Monitoring and Optimization

You need to continually develop new features – the needs are going to be changing based on customer requests and competition, changing habits, alterations in your business model and a lot of other factors.

Lack of Up-to-Date Technology

Search engine development is a very specialized field, and most of the times you will not have relevant, fresh technology at your disposal to use – you have to do everything yourself, which will usually take months to accomplish.

No Support

With an in-house team, you can’t rely on outside expertise: if you hit problems, you will have to figure it out yourself or find the solution on dev forums – both of which can be time-consuming and frustrating.

SaaS Vendor Enterprise eCommerce Search Provider

If you don’t want to hire and maintain a full development team, and still want a high-quality,flexible solution, your best option is to use a SaaS provider.

But you are going to have to be careful about which one you choose (see the comprehensive list above for the most important factors to look at when choosing a provider)-, because search-as-a-service is a highly specialized field where every vendor, every solution has their specific strengths and specializations.

Quick integration

When using a search provider, some offer integration within weeks, which means you’re up and running in no time.

Some SaaS site search providers have integration times that take months, though. 

So if you’re looking to get started quickly, check the timeline of each individual provider, along with their integration process, to see which best fits your needs. 

Proven ROI before commitment

You not only get better search in a shorter amount of time, but also get the opportunity to commit to a service that has history: case studies, testimonials to back up its effectiveness. The core features are going to be constantly updated based on data not only from your webshop, but from all the shops and clients the vendor has, meaning best practices could be implemented with a much greater effect.

Detailed Analytics

You are going to get a dashboard and likely options to customize the analytics and reports you get, meaning you will be able to implement any useful data in your sitewide marketing efforts and business model from day one.

Dedicated Customer Support

A team of experts will be at your disposal as a part of the service, and you don’t have to pay their salaries or for the tools and devices they use. You can also be sure that they are experienced in search technology and will likely have the answers you are looking for.

Fully managed search

You and your team will not be the ones who have to make updates, develop features, manage scaling or hosting: you can reach out to the SaaS vendor and simply tell them your needs.

Freedom in Scaling Along With Your Business

SaaS solutions are traditionally scalable: as your online store and business grows, you can simply upgrade the service to fit your growing needs, from handling a larger amount of products to including more specialized, custom functions and options in the engine.

Predictable Cost

The expenses will be easy to integrate into your cash flow, and you are going to know exactly how much you are going to pay each month for the service – and if you need custom solutions, the price for developing and integrating them will also be set.

Disadvantages of Using a SaaS Vendor

No Real Control Over Development

As just one of the clients, you will have some say in what features have to be developed, what bugs are going to be fixed – but that say will be minimal. Their development team is going to focus on what the majority of clients and the market are demanding – which might not necessarily align with your personal needs

Important note: this is not the case with all vendors. Business models of SaaS can differ – notably, at Prefixbox we are developing features based on aggregated customer priorities. But this is not an industry standard.

Operational Features to Consider

Free Trial

Before committing to any solution, you should be able to try it out beforehand. Always look for a free trial or A/B testing opportunity with an enterprise eCommerce search provider, because you are about to make a long-term decision. Also, make sure that your analytics is set up correctly, so you can track any changes in user behavior, conversion rates, exit rates, and all other important KPIs after launching the free trial.

On-going Optimization Support 

Don’t count on a moment happening when you can kick back and say, “well, we are done with site search”. As one of the most important elements on the site, intimately tied to user behavior and revenue generation, continuous optimization, fine-tuning, A/B testing and bug fixing will be a part of your routine. So make sure the solution you chose has the team to support these efforts and are available to help with whatever you need (in a quick manner!).

Analytics Dashboard 

A monthly PDF in your inbox will not be enough when it comes to site search analytics. Make it a point to have access to a live dashboard where you can not only see the most important data and changes of KPIs.

Data Centers, Uptime

Uptime is a critical factor: always check the location of your provider’s data centers along with their SLA to find out their policy on uptime and redundancy. 

The best option is to work with a provider who has their data center in the same time zone as your business, as this will help with potential latency issues.

What is their SLA?

Service Level Agreement 

Many important factors can be found in an SLA, for example, the availability they vow to provide. 99% is the standard benchmark, but keep in mind that this is not all that good a percentage – if your site search works 99% of the time, it means that it is offline 7+ hours a month.

Monthly Cost

Monthly invoicing instead of a year-long commitment gives you both security and flexibility: you can expect fixed costs while also having the freedom to opt-out of the service if necessary.

Occasionally, yearly invoicing comes with a discount – so be sure to check that out.

The Decision is Yours

Choosing an enterprise eCommerce search provider solution should not be a fast decision. There are a lot of factors to consider to give shoppers the best possible search experience – and also support any ongoing marketing efforts and your business model, to achieve the highest possible ROI.

If you have doubts, start with the comprehensive list above, and if you still have questions, don’t hesitate to contact us, we are always happy to help with your decision.

Do your research before making a long-term commitment, and keep in mind that what you go with will fundamentally decide the customer journey and shopping experience in your store!

Paige TyrrellHead of Marketing – Prefixbox

Paige is the Head of Marketing at Prefixbox, a leading eCommerce site search solution. She’s an American who’s been living in Budapest since 2017 and loves giving #alwayslearning sessions to help people optimize their online stores.

12 Best Machine Learning Strategies for E-commerce Businesses

12 Best Machine Learning Strategies for E-commerce Businesses

In today’s world of E-commerce, it’s impossible to run a successful online shop without leveraging Machine Learning.

Check out this article to find out how to incorporate artificial intelligence into your
e-Business in these easy ways:

  • For price optimization
  • For search result optimization
  • For product recommendation
  • And many more

Let’s get started!


Today, when global E-commerce sales amount to $4.88 trillion, it’s an innovator’s game.

According to Gartner, up to 80% of customer interactions are managed by AI today. 

Introducing AI into sales, marketing, even inventory management is both a natural next step in the evolution of commerce and an absolute must for those who want to keep up and make cost-effective decisions.

And in this article, we are going to show you how you could effectively be leveraging AI in your E-commerce business.

Let’s kick off with the answer to….

How machine learning works –
AI Use cases in E-commerce

If we want to define machine learning in the simplest terms, it is the study and creation of algorithms that can learn from given sets of data and extrapolate from those to predict trends and outcomes.

This is a necessary function of any algorithm-based system that has to deal with large, frequently changing data sets – from natural language searches (where wording, typos, and new phrases are common) to facial recognition (where the system basically never meets the same input twice, but still has to determine the key features). Machine learning is used in email filtering, weather forecasts, diagnosing disease and predicting socioeconomic trends.

E-commerce is a field in dire need of machine learning. Even a small webshop can create millions of relevant data points daily that a single human, or even a team of experts could never hope to fully comprehend and utilize.

But, a machine can do it. 

It will not only monitor these data sets, changes and trends in them, but make connections. 

An AI based on machine learning will reason. Deduce, based on experience.

For example, it can help with…

1. Pricing Optimization

Price is, unsurprisingly, one of the most important factors when considering a purchase. In fact, price is one of the major drivers for at least 47% of customers according to BigCommerce.

If your prices are high – or just higher than your competitors’ – your cart abandonment rate can increase, and even more so if you don’t set shipping prices correctly. If your store offers international shipping, you also have to display prices based on the location of the customer.

There are many other factors that may influence your prices – supply and demand, which promotions are running, what time of the day or year it is and so on.

Via machine learning, these factors could be assessed in a fraction of a second and your site will display dynamic pricing: showing the customer the most up-to-date price for them based on the aforementioned factors. 

This ensures prices are instantly understood (no “how much is that in my currency”, “does that include shipping” etc.), with no surprises, and no need for you to manually change them.

2. Segmenting, Personalization and Targeting Customers

In traditional brick-and-mortar stores, segmenting was done solely by salesmen, who approached the customers. By observing and asking a few questions, they could gather more than enough information about demographics, their needs and doubts, and their overall mood based on non-verbal communication.

This made it possible to immediately address any concerns about the products, matching the words and expressions used by the customer, being able to upsell them on the spot, and reinforcing their purchase intent.

When the customer has to deal with an online experience, all of this has to be replaced, and scaled. 

Segmentation has to be based on behavior that can be measured. 

And on-site search is a great place to start.

When searching, shoppers use natural language, which gives clues about their background and native language. Based on the results they click on, an algorithm can determine what they are looking for and offer more relevant results, which encourages them to make a purchase. 

And based on what other customers in the store previously bought, in conjunction with this particular search, cross- and upsell offers could be made with great confidence. Part of the demographic information is readily available as soon as someone signs up – most times you will know their age, location and whatever data you ask for. 

You can also monitor their behavior on your site – what content do they read, what materials do the download, how often do they return and how often do they make a purchase.  

You can also keep track of the kind of emails they open and when are they most likely to make a purchase during a day, a week or year.

There is an incredible amount of data generated from all these interactions – and machine learning is what makes it possible to assess behaviors to ascertain patterns. 
These patterns help you make sense of data sets and make more effective marketing decisions. For example, you can create ultra-targeted campaigns with the most relevant messages, and increase your conversion rate by offering customers exactly what they are looking for.

3. Search Results Optimization

Providing search results based on keywords is just the very first, most basic step in on-site search. In order to provide shoppers with the best experience possible, your search should go much deeper than that. 

Given a large enough data set, you could determine which results are better for people in certain locations, how to optimize E-commerce search result filters, and which products best match their needs based on their previous behaviour.

By analyzing the data and figuring out what items go together, you can also recommend similar products, and even cross-sell items that are frequently bought together by your users. 

Machine learning is able to pinpoint the trends and patterns needed to automatically determine these.

Leveraging this leads to higher click-through rates on your results pages, higher conversion rates, and a higher average order value.

It’s worth noting that filtering results in real-time based on user inputs, via faceted search, paired with AI-based recommendations is currently the single most effective method of providing relevant results.

But let’s stay for a minute with recommendations…

4. Product Recommendations

If you’re curious about how effective automatic product recommendations are, look no further than Netflix. According to McKinsey, 75% of what people watch on the streaming platform is suggested to them via an algorithm that analyzes user behavior. (The same is true for 35% of purchases at Amazon.)

You might think that determining which products go well together is easy – for example, simply track what people watch after they finish Futurama, and if it’s Rick and Morty, recommend that, and that’s it.

But you should also take into account the demographics – people in non-English speaking countries might go for movies and series in their native language with more enthusiasm. The preferences of young and older audiences will differ, as will the taste of those living in East Coast metropolitan areas versus those in rural European towns.

And this doesn’t even account for personal taste of genres, styles, lengths, eras etc. 

To be able to offer ultra-targeted products to your customer, you have to work with a very large number of variables, each weighed differently.

Learn more about the most important related search features to increase conversion rates & revenue by 6%.

It is virtually impossible to do this without a machine learning algorithm – even if you have just a few hundred products.

As you can see, knowing your customers is a bit like predicting the future.

To prove the importance of providing the shopper with the right product suggestions throughout the customer journey, check how significantly Rossmann’s conversion rate and average order value improved by utilizing AI-backed recommendations:

5. Predictions About Your Customers

Machine learning can tell you a lot of things about the people who visit your site and make a purchase – even things like how likely they are to buy from you again or what they might be interested in. 

Check out what machine learning can predict.

Customer Lifetime Value Prediction

To fine-tune your communication and messages, it’s useful to know how much money  a customer is likely to spend in your store over a given time period. 

If you can estimate a lifetime value based on behavior, you can also make your marketing more cost-effective and targeted. 

You can also identify customers who are the most valuable and deserve special attention.

Predicting if a customer will make a purchase

Imagine this: you have an online store for office supplies. You have a customer who orders the same amount of ink cartridges from you about every 6 months.

Let’s say 5 months pass, and this customer logs into your site, but doesn’t order anything. Logically, this could mean that they are checking the prices in order to budget the next order, or they might be comparing your prices to your competitors’.

As a store owner, you most likely will never learn that they’ve logged in and done this.

However, the AI will notice it. 

And it can deduce that this is the right time to provide a little incentive to increase average order value – as the customer is a returning, but hesitant one, maybe it’s time to fire up a workflow that provides them with a personal discount for their next order, thanking them for their loyalty.

Predicting customer return (and purchases)

If, based on their behavior, a customer is likely to return to your store later, then an entirely different marketing message may resonate better with them.

Algorithms can predict, on the long term, whether this is the case and can initiate longer workflows with messages that target loyalty and reinforce branding.

Customer churn prediction

Retaining existing customers is one of the most important tasks if you want to keep your marketing budget from bleeding. Acquiring new customers is much most cost-intensive.

A machine learning algorithm can determine which shoppers are the most likely to abandon your site – based on things behaviour such as: returning to your store less frequently, smaller purchases, etc.

When the AI realizes this is the case, it can initiate workflows targeted at retention, giving shoppers incentives to stay.

Client size prediction

Based on average order value and purchase frequency, along with other information like number of employees, and company type, an algorithm can estimate the size of the client for you.

This gives you great insight into deciphering which potential customers to pay special attention to. You can make more personalized offers and give them deals that are more cost-efficient in the long term.

As you can see, machine learning really tells your future by predicting what your customers will or won’t do.

Another forward-looking application of machine learning that deserves its own subsection is…

6. Site Search Autocomplete

A truly useful autocomplete has to learn, instead of scouring through different product attributes and description.

It has to understand the natural language of the users instead of the often technical and machine-like phrases of database entries.

E-commerce search best practices dictate that online stores use an artificial intelligence-based autocomplete as it streamlines the shopping journey and provides shoppers with a sense of comfort as this is what they’ve come to expect. 

So, natural language processing and machine learning have to understand the kind of language and phrases your customers use, how often they use them, if the results are satisfactory for those phrases, and even common typos along with the correct spelling.

Learn more about the most important search autocomplete features to decrease zero-result pages by 31% and increase cart value by 19%.

This way, every user can feel that the search engine is truly there to help and pays attention to the smallest details, all without being annoying.

Speaking of details…


7. A/B tests using AI

A/B testing is a powerful tool in online marketing, but it can also be tricky.

Let’s say you want to run an A/B test for a product page.

Firstly: what do you change? The display of pricing? The location of your CTAs? The background color?

You see, if you change more than one thing, you can’t really be sure what causes the next positive or negative change. 

But, if you change only one thing, the change may be so small it can’t even be registered.

And what KPIs should you track? 

Yes, conversion and purchase rate are the most obvious numbers to track. 

But time spent on page, number of clicks, and return rate are also important. 

Machine learning and AI makes this testing process easier:

  • Based on historical data, it decides which elements should be tested and automatically creates variants.
  • It can dynamically change page elements based on test results. For example, displaying pages differently for different demographics or locations.

It can find the optimal versions much faster, because it can take all the variables into account and find the connections between even small changes.

8. Chatbots for Automated Customer Support

With customer support, there rarely is an optimal choice. If you want to solve all issues with manpower, your support team will be huge and expensive – and not at all efficient, as many times they will be dealing with things that could be solved by redirecting the customer to an FAQ page.

On the other hand, you can’t completely automate support, because a lot of issues will require human assistance, and your customers will quickly get annoyed if they can’t get it.

On the other hand, you can’t completely automate support, because a lot of issues will require human assistance, and your customers will quickly get annoyed if they can’t get it.

The solution to this problem is often implementing a chatbot based on machine learning. 

These chatbots are able to maintain a conversation with the customer. Not only by using previously defined answers, but also via AI – they’re able to learn about natural language from every conversation.

Of course the chatbot will require time to learn, to get to know the products and services as well as the customers and their way of communication. 

It will never pass a Turing test, but it could, in time, identify opportunities for upsell, create customized coupons, and open tickets for humans in customer support to handle.

Keep in mind, however…

A professional custom chatbot can cost anywhere between $30,000 and $300,000, depending on the desired functionality.

Speaking about money well spent…

Artificial Intelligence recently opened another avenue of sales for E-commerce stores by combining online shopping with chatbot assistance and messaging apps. This new means of shopping called conversational commerce, can not only enhance the customer experience, but improve conversion rates, and increase online sales overall. Check the video for more details:

9. Inventory Management

You know how a smart fridge reminds you you’re out of milk and puts it on your grocery list for you?

ML does that for E-commerce, times a billion.

Well, a trillion actually, as around 7% of the US GDP (more than $1.1 trillion) is tied up in cash in inventory along with accounts receivable and accounts payable.

Inventory management and logistics is nothing to sneeze at in terms of complexity. 

If you run a successful online store, you have to monitor your stocks, reorder items, predict trends in demand, coordinate contractors, deal with manufacturers, suppliers, mailing services, and manage your revenues accordingly.

This is exactly the kind of job machine learning was made for. 

By monitoring all your inventory, and even predicting future trends in supply, demand and even cash flow, you can be sure that with AI, you won’t be the airline that overbooks flights or flies empty.

You will be the airline that plans their next year schedule with absolute confidence.

10. Omnichannel Marketing Boosting with ML

Let’s get the obvious out of the way: omnichannel marketing brings you higher retention and conversion rates and boosts your revenue. But only if you use the available channels wisely.

Of course, you could just hire a marketing team, where there are dedicated people for social, email marketing, content creation, and tell everyone to give it 100%.

But is there a better way to do it?

Actually, yes.

Analyzing the data your customers create when interacting with your activities on those channels is majorly important. 

Based on customer behavior – ads that perform well, content that’s frequently read, email open rates – machine learning algorithms can analyze your messages and display them in a way that ensures every customer gets the perfect one.

Online real estate became more important than television and radio partly because of how results can be tracked and campaigns can be optimized. 

ML is the natural next step in this evolution.

11. Image Processing and Recognition

Image recognition can be a great tool for an online store with thousands of products in their inventory. 

Ideally, it enables a customer to just upload a photo they snapped of a given product at home or at the store. The system then processes it on the store’s servers, and instantly displays a response with availability, current price, shipping info so the shopper knows where to buy it.

More often than not, this leads to a purchase as it’s the most convenient shopping method., for example, was able to increase sales by 15% after implementing their visual search feature.

Another possible use of image processing is to give ultra-targeted recommendations. For example, processing a photo of the customer and then applying clothes available in store, so they can see how different styles would look on them.

12. Fraud Protection

Fraud is inseparable from commerce as a whole, and E-commerce is especially vulnerable. 

Online shopping offers huge opportunities for those who want to take advantage of automated systems, which is why it is important to have algorithms in place that can detect fraudulent activity.

Integrating a CAPTCHA is not enough. You also have to monitor behavior and look at how certain people use your site. 

Integrating a CAPTCHA is not enough. You also have to monitor behavior and look at how certain people use your site. 

With machine learning, you can identify repetitive patterns that don’t match with human behaviour; such as: filling in forms too quickly, opening dozens of pages for split seconds, entering multiple different sets of information rapidly at checkout, etc.


Introducing artificial intelligence in E-commerce processes is not something futuristic or even a highly complex task. There are applications and services readily available for most functions that have been developed by people over the past decade.

It’s not always cheap, but it’s worth the cost.

  • It will help you understand your customers and audience better
  • Increase your sales and average order value
  • Cut unnecessary work  
  • Give you deep insight that no human ever could

You don’t have to hand over everything at once to the machine of course. Start small. Implement a smart site search solution, start using recommendations based on machine learning – and get to the rest when you are ready.

But get moving, because the analogue methods are phasing out fast.

Balazs Vekony profile picture - Prefixbox
Balazs VekonyOnline Marketing Manager – Prefixbox

Balazs is an Online Marketing Manager at Prefixbox, a leading AI search solution provider for E-commerce sites. He is a marketing enthusiast from Budapest, who is interested in new technologies and solutions and believes in the power of search.

Algolia vs ElasticSearch: How To Choose The Best Site Search For Your Business [+2 Alternatives]

Algolia vs ElasticSearch: How To Choose The Best Site Search For Your Business

If you run an online business or a webshop, providing a seamless search experience for your customers is probably one of your main concerns.

However, choosing the right provider can be tricky as there are a lot of factors to take into account.

To help you make the right decision, we put this article together to:

  • Provide an overview of some of the most popular search solutions
  • Show different pricing models
  • Explain the pros and cons of different solutions

So, let’s get started!

Algolia vs Easticsearch site search comparison guide


Chapter 1

How do site search solutions work?

Chapter 2

Algolia vs Elasticsearch: what are the differences?

Chapter 3

What are some alternatives to Algolia and Elasticsearch?

Chapter 4

Site Search Solution Pricing Comparison

Chapter 5

Pros and Cons of the Site Search Solutions

Chapter 1

How do site search solutions work?

In essence, site search, as a whole, is the functionality that enables users of a website to search certain keywords and/or criteria and get relevant, ordered results from the site.

The functionality itself can greatly vary – on some sites, you will only find a search bar where you can write keywords, on other sites faceted search will guide you step by step, in real time, based on all relevant attributes.

The basic idea is the same as on Google, but the underlying mechanics can be vastly different. The way individual site search engines interpret individual keywords and phrases, whether they can understand typos, which product attributes are weighed as more or less important, the speed, the functionality of facets and search filters – all depend on how the site search solution .

Something important to note…

Most searches do not take place in real time. In general, a search engine will index the searchable content, create the possible results pages in advance, and display these when the search is performed, hence the great speed. 

However this also means that this index has to be updated regularly so that it takes changes into account and provides truly relevant results.

With other search solutions, such as with faceted search, the process is done in real time, based mainly on all the different attributes provided for the different results. This is of course a more resource-heavy, but also much more user-friendly solution than regularly updating the search index.

Now that we’ve gone over the basics or site search, let’s dive into a few of the site search solutions.

Shall we begin?

Chapter 2

Algolia vs Elasticsearch: what are the differences?

Algolia is mainly praised for its speed, easy implementation and excellent support. Typo handling and search analytics are some of the key features it offers.

In the case of Elasticsearch, the developer community values its powerful API, that it’s open-source and flexible.

What is Algolia?

Algolia is an on-site search engine that includes analytics. It enriches the search experience with Autocomplete,  filters, query suggestions and infinite scrolling.

For developers, it offers back-end API clients, front-end widgets, several integrations, a customizable relevance algorithm and extensive documentation. It is usually described as having a developer-friendly API and an extensive toolset for development.

Algolia is a more full-service search provider, but can still be built up further.

What is Elasticsearch?

Elasticsearch, by its own definition, is a “distributed, RESTful search and analytics engine”. The open-source solution offers various APIs and great search speed, and is widely favored by developers. It also has built-in analytics and is a fully managed Search-as-a-Service solution.

Elasticsearch is a basic search application, which developers can build a deep search on top of.

Hosting Services to Consider when Using Elasticsearch


Elastisearch is offered as a part of the Amazon Web Services as a fully managed search service, which means deployment is usually easy and fast, and security is provided by AWS. 

This makes it easier to develop and troubleshoot your own Elasticsearch based application in their environment and scale it as you need. 

Another strength of this offer is its cost-effectiveness: as with many SaaS solutions, you only have to pay for the resources that you actually use.

Elastic Cloud

You can also opt in for the Elastic Cloud managed service, which is compatible with not only Amazon Web Services, but also Google Cloud and Microsoft Azure.

The main advantages of this solution is its flexibility, as it is easy to configure and deploy (given you have experts for the job), it includes a lot of Elastic features like machine learning, Canvas, APM, workplace search, and it provides unified analytics. 

You also get backups, constant monitoring and ticket or SLA-based support as part of the managed service.

Chapter 3

What are some alternatives to Algolia and Elasticsearch?

In this section we are going to show you two options that you might opt for if Algolia or Elasticsearch is not suitable for you

  • Azure Search: a search solution by Microsoft
  • Prefixbox: a SaaS option with constant optimization and feature development with a focus on client needs.

What is Azure Search?

Azure Cognitive Search is a solution by Microsoft, and by its own definition, “the only cloud search service with built-in AI capabilities”. The aim of the service is to make search at scale easier and faster with AI.

It is based on the same natural language stack as Bing and Office, and includes the key features of auto-complete, geospatial search, filtering, and faceting capabilities, and also OCR, key phrase extraction, and named entity recognition.

What is Prefixbox?

Prefixbox is an eCommerce search provider with products enriched with AI capabilities. The solution is composed of different modules, which make it flexible and a suitable solution for medium to large eCommerce enterprises.

The modules are Autocomplete, Related Searches, Semantic Search Engine, and Digital Merchandising. These encompass features such as typo tolerance, advanced ranking, thorough search analytics, and of course faceted navigation for enriched user experience.

Other Algolia Competitors

In addition to the two solutions above, Algolia has the following alternatives in 2022:

  • Yext
  • Klevu
  • Loop54
  • Swiftype
  • Findalogic
  • FactFinder
  • Hawksearch
  • SLI Systems
  • Lucidworks Fusion
  • Celebros Site Search
  • BloomReach Experience
  • Optimizely Episerver Find
  • IBM Expert Personal Shopper
  • SearchSpring Relevancy Suite

6 Open Source Algolia Alternative

If you prefer free and open source site-search solutions, you can choose from:

  • MeiliSearch
  • Apache Solr
  • Typesense
  • Manticore Search
  • OpenSearch
  • ItemsAPI

Chapter 4

Site Search Solution Pricing Comparison

Every search solution has a different pricing model.

Here, we’re breaking it down based on what is available on each provider’s website.

Algolia pricing

Algolia offers three main pricing plans that serve different volumes of searches, defined as units, where 1 unit equals 1,000 search requests.

  • They have a Free plan, which serves up to 10,000 search requests.
  • Algolia Search is a $1.00/unit/month plan (20% off if you pay annually) that includes analytics. They also have a Premium package, a $1.50/unit/month plan (20% off if you pay annually) that includes advanced features like custom rules, merchandising, and customization.
  • Lastly, they offer Algolia Recommend, a flexible, hosted recommendation API with advanced programmatic control; this is a $0.60/1000 requests/month plan that includes related products & frequently bought together features.

On their site, you can also create a custom package based on your business’s search volumes and the specific features needed for your online store.

Elasticsearch pricing

There are four main plans you can choose from with the Managed Elastic Cloud service:

  • Standard, $16/month, with the core security features and basic functionality.
  • Gold, $19/month, custom plugins and support during business hours.
  • Platinum, $22/month, with advanced security features, workplace search, machine learning, full 24/7/365 support.
  • Enterprise, $30/month, which includes endpoint protection, detection, response and event collection.

All these plans are available as a free trial, and you can learn more about the detailed functionality included in each one if you visit their site.

Azure Search pricing

Azure has many plans for smaller and larger businesses too, with a Free plan also available, with limited features of course. The prices are given as an hourly rate and per unit in US dollars.

  • Free: you get a 50 MB storage, 3 indexes per service and not much more – this plan is basically a demo version so you can test the service and decide which paid plan is ideal for you.
  • Basic: with 2 GB storage and  maximum 15 indexes per service, it costs $0.101/hour
  • STANDARD S1: includes 25 GB storage (max. 300 GB per service) and 50 indexes per service for $0.336/hour. All standard plans have a scale out limit of 36 units per service.
  • STANDARD S2: Includes 100 GB storage (max. 1 TB per service) and 200 maximum indexes per service for $1.344/hour.
  • STANDARD S3: Includes 200 GB storage and 200 or 1000 indexes per service per partition in high density mode (max. 2 TB per service) for $2.688/hour.
  • STORAGE OPTIMISED L1: 1 TB (max. 12 TB per service) and 10 indexes per service for $3.839/hour.
  • STORAGE OPTIMISED L2: 2 TB (max. 12 TB per service) and 10 indexes per service. The pricing for this plan is $7.677/hour

As for image extraction, the basic and all standard plans are priced as follows:

  • 0-1M images – $1
  • 1M-5M images – $0.80
  • 5M+ images – $0.65

Prefixbox pricing

Prefixbox’s pricing is based on an eCommerce store’s number of searches per month and is billed annually.

Our search service includes:

  • Support by email or phone from a team of experts responsible for integrating the search solution, optimizing it, and the commercial TOS. 
  • High ROI proven during the first 30 days for sites with more than 300.000 visitors per month 
  • Managed synonym management prevents you from additional in-house work and provides better search results for your customers. 
  • Search Analytics at any time you want 
  • Product Ranking and relevancy optimization 
  • A/B testing with your current search solution 
  • Regular reviews and health-checks of your eCommerce search analytics  

If you aren’t sure how much you should be spending on a site search solution, you can use this calculator to see how much revenue an optimized search will generate. 

Or you can request a demo and talk to an expert about your search needs.

Chapter 5

Summary: Pros and Cons of the Site Search Solutions

Since we covered so much information, check out this quick list of pros and cons of the four site search solutions to get a simple overview of the core enterprise eCommerce search features and functionality.

Pros of Algolia

  • It’s fast and easy to implement.
  • It has excellent support.
  • The underlying technology is modern and easy to handle (for experts).
  • It can handle typos, which is one of the most important search engine functions.
  • It also provides detailed analytics.

Cons of Algolia

  • According to some developers, the documentation could be problematic and thus hard to work with because sometimes it is incomplete.
  • Algolia is one of the more expensive solutions, so it is usually not the best choice for small businesses. 
  • Scaling can be difficult.
  • Customizing the ranking with Algolia is not perfect, as it is based on prioritizing factors instead of weighing factors individually.

Pros of  Elasticsearch

  • It is generally considered easy-to-use.
  • It is well-scalable – easily managing Terabytes of data (given the technical conditions are in order.
  • It can handle complex keyword searches very efficiently.
  • Analytics is usually listed as one of the greatest pros as it provides a thorough overview of user behavior and search data for its clients.
  • With APIs based on REST, an HTTP interface and schema-free JSON documents, its relatively easy and fast to build different applications based on it for a variety of use-cases (including, in addition to on-site search, processing genetic data, managing and analyzing spatial information, modeling data behavior etc.).

Cons of Elasticsearch

  • In reviews, developers list a number of cons for working with Elasticsearch, such as occasional lack of documentation, debugging issues, and a steep learning curve.
  • Because of the pricing, it is not an ideal choice for small businesses.
  • Support is not always proactive and some developers noted that community support is more efficient than the one the company provides.
  • Switching between different versions and  configuring it can be tricky even for experienced developers. 

Pros of Azure Search

  • It can be seamlessly integrated with Azure Cloud, making it an ideal choice for clients who already use the Microsoft cloud service.
  • It provides a fully-managed service for loading, indexing and query content.
  • Cognitive search capabilities enrich not only the search experience, but also make it easier for the client to get in-depth analytics and data.

Cons of Azure Search

  • As a primarily enterprise-level product, it requires a larger budget.
  • It works well primarily with Microsoft products, can be tricky to integrate with other ecosystems.
  • Implementation requires experienced team members on the client side as for non-technical administrators this is a nearly impossible task.

Pros of Prefixbox

  • AI powered search provides a seamless search experience.
  • The faceted real-time search makes it easier for customers to find the products they are looking for on the site and improves their overall shopping experience.
  • Prefixbox is constantly releasing new features catered to eCommerce retailers.
  • Contains a full suite of products focused on increasing eCommerce revenue and conversion rates. 

Cons of Prefixbox

  • It’s not an ideal solution for SMBs where there are less than 100 000 searches performed on a website per month. (~300.000 visitors / month)
  • Does not include content search functionality.

A final few notes before you make your choice…

Pricing can often be complicated, so make sure that you correctly calculate prices based on your needs. 

Never make a final decision before consulting with the technical people on your team. Developers can encounter or even foresee many issues that others can completely miss.

Always look at the availability and proactiveness of support: site search solutions can be a bit complicated, and it can make a fundamental difference if you are able to get expert help when you need it.

And lastly: before making any final decision, always look up reviews from real clients who actually use these products: get first-hand information on their experience instead of reading general product reviews!

You can learn more about the most important eCommerce search engine features to increase conversion rate by 15% & revenue by 47%.

Paige TyrrellHead of Marketing – Prefixbox

Paige is the Head of Marketing at Prefixbox, a leading eCommerce site search solution. She’s an American who’s been living in Budapest since 2017 and loves giving #alwayslearning sessions to help people optimize their online stores.

eCommerce Optimization: 7 Simple Ways To Improve Search Data Infrastructure

eCommerce Optimization: 7 Simple Ways To Improve Search Data Infrastructure

The back-end of your search function is equally as important as the front-end.

eCommerce search data can give you insight into which of your products are popular, which search queries yield no results, and give you overall insight into how to best stock your store.

Infrastructure is also important – minimizing the downtime of your search function is critically important to your company’s success since search users convert the best on websites.

Frequently review search suggestion quality

It’s important to review your eCommerce search data to see if your search relevance is improving.

Frequently checking your search analytics will help you find problems when they arise, so you can fix them before they escalate.

Your site search solution should be continually improving – most solutions do not self-improve, so you will have to manually make changes by tracking user behaviour on your site.

However, there are some solutions that do improve automatically, such as Prefixbox.

These solutions eliminate the time consuming effort of monitoring and continually updating your search platform and thus are generally a great investment.

In any case, you should check for improvements in categories like click-through rate, revenue, and average clicked position, which are important metrics to track.

Make note of the most frequently searched terms

Stay up-to-date with the terms people are searching for on your website.

These popular searches give insight into which of your products are performing well, the seasonality of your products, and which new items your customers are looking for.

Note this: These popular searches give insight into which of your products are performing well, the seasonality of your products, and which new items your customers are looking for.

This eCommerce search data is especially helpful when you are re-stocking.

Use popular searches for SEO and PPC campaigns

Once you make note of the popular searches on your site, be sure to make the most of them!

The language your website visitors use on your search function is the same they use for online search, so your database of previously searched terms is a PPC gold mine.

You can use these keywords as a base for your SEO efforts along with your PPC campaigns.

By having insight into the direct phrasing your customers use, you can fine-tune campaigns to reach your exact audience.

This insight will help your campaigns perform better and boost the SEO of your website, enabling more people to find your products.

Find search queries that have poor results

Search queries that have poor results can provide valuable insight. These searches are terms with a click-through rate of 0.

This can happen when people search for products you don’t carry or when customers search for terms you don’t use to describe your products in product descriptions and end up on the 0-results page. If you need inspiration, check out similar product videos on YouTube.

  • If visitors frequently search for products you don’t carry, use this insight when considering what products to stock for the following year.
  • If you decide you don’t want to stock these products, you could provide information about them or recommend similar products, so you still provide value to customers searching for those items.
  • If your customers use different wording to find products on your site, consider switching to their terminology or creating synonym rules so both terms will direct the customer to the product. 

After you make these conversion rate optimization changes, be sure to keep monitoring your search data in your site to ensure the changes you made were beneficial. If not, continue testing options until you can perfect your product names and stock.

Check search data for keywords that become popular

Keep an eye out for search terms in Google Analytics that become more popular, so you can keep up with demand. Popular search terms often change – you might see regular seasonal changes or spontaneous cultural changes.

Terms that become more popular signal which products your potential customers want to buy. If you see a sudden increase in a search term, stock more of that product, so you can keep up with (the anticipated) demand.


Keep an eye on your top listed products

Having data on the click-through rate of your first 5 listed products per search term is important, since these products should be the most relevant to the search and therefore most clicked.

If you consistently see a high click-through rate on these products, great!

If not, try re-ranking your products to place the most popular ones first; this should increase the conversion rate on your site.

Gather mobile website data too

It’s important to track Google Analytics data from your mobile site, so you can see which features are the most important and frequently used by mobile users.

Site traffic patterns provide insight into how campaigns are performing and about user behaviour on your site, so you can see where to improve.

Be sure to track things like which browsers are used, how long visitors stay on your site, how they navigate, and and the shopping cart abandonment rate.

User experience is critical for eCommerce because it ensures that your customers can easily navigate through your site, find what they need and buy it. When you make it easy for your customers to buy from you, you’ll sell more frequently and improve customer lifetime value. That’s why you need to make sure your business offers the best customer experience for desktop and mobile users.

Have multiple data centres

Search plays a big role in directing customers to the products they want to buy, so it’s important your search solution doesn’t suffer from downtime.

Data centres sometimes fail and go offline, so in order to avoid down time, host all the critical parts of your web store in multiple data centres.

Consider using a SaaS search solution

SaaS, software as a service, solutions can save time and money for your eCOmmerce Business.

These solutions often work at a higher capacity and come at a lower price than on premise solutions.

Plus, they’re usually hosted in multiple data centres, which minimizes down time.

If you’d like to start collecting data on your eCommerce site, you can set up Google Analytics.

If your eCommerce search data collection and infrastructure are properly set-up, make sure your faceted filtersSERP, no result pages and mobile version of your site search are also optimized.

Balazs Vekony profile picture - Prefixbox
Balazs VekonyOnline Marketing Manager – Prefixbox

Balazs is an Online Marketing Manager at Prefixbox, a leading AI search solution provider for E-commerce sites. He is a marketing enthusiast from Budapest, who is interested in new technologies and solutions and believes in the power of search.

11 Must Have Conversion Boosting Search Features for Ecommerce Sites

11 Must Have Conversion Boosting Search Features for Ecommerce Sites

Site search is a powerful tool that can boost your online revenue and conversion rates while providing your shoppers with a better experience, but a lot of online stores still aren’t fully leveraging it.

Check out these 11 must have conversion boosting search features to find out how you can optimize your shop.

Site Search Conversion Boosting Features Aricle intro

Does Your Online Store Satisfy 85% Of Visitors?

People research what they want to buy before making a purchase. In fact, 87% of shoppers begin their product searches using online channels. Shoppers most often research products by executing a search from the search box to get directly to the product they’re thinking about buying. They’ll read a bit of information about it and check the same product on your competition’s site to compare prices.

If they can’t find the product on your site, it’s a sure bet they will be taking their business elsewhere.

Site search functions that are complete and clear keep visitors on your site instead of driving them to the competition.

Site search also maximizes revenue.

Did you know?

  • Site Search users convert to sales 4-6 times more often than non-search visitors.
  • Revenue from search function users makes up about 50% of a web shop’s total revenue.
  • A simple text-based search bar reduces site abandonment from 65% to 40%.
  • Semantic searches, which identify customer intent, reduce abandonment further from 40% to 2% and lead to an average of 16.2 page views per session.

Site search is one of the most powerful tools you can use to increase your sales, but most e-commerce sites are terrible at using it. An overview from a few years ago revealed how many businesses are failing site search basic training. Here are a few numbers:

  • 16% of ecommerce sites do not support searching by product name or model number.
  • 70% require users to use store jargon instead of applying any kind of synonym matching.
  • 60% of ecommerce sites do not use faceted search despite it being the foundation for filtering results properly.

If you’re making any of these mistakes, it’s likely that search is hurting your business.

How Much Does Bad Search Cost Your Business?

E-commerce search engines can boost conversion rates by an average of 15.8%. Optimized search drives an online revenue increase of 47%. What does that mean for your business? Let us think about just two numbers:

  • Your total revenue from sales on your website last month.
  • 1.5

Now, multiply your total revenue by 1.5. The answer is how much revenue your site could have generated last month if you had already been using an optimized search function.

ecommerce lost revenue calculation

We’ve compiled a list of the hottest 25 eCommerce site search best practices to help you reap the full benefits of search. Imagine putting all these into practice. Your user experience and conversions rates will skyrocket!

Consider These Must Have Site Search Features to Improve Your User Experience

By now, you should be convinced of the importance of a great site search function.

Your search directly guides shoppers to the products they want to buy – if your search can’t do that while meeting their expectations, you’re losing out on business.

Here are the search features you should include to boost your conversions and sales.

Site Search Box

A search box is one of the most basic site search features you can have on your website. Site search functionality is crucial to a positive user search experience.

Let us imagine a simple example: You sell island-themed shoes and clothes on your website. A visitor comes to your site hoping to buy a men’s extra-large Hawaiian shirt. After browsing your catalogue for a few pages, he is not closer to finding his size or design. Your visitor has two options:

  • Continue browsing slowly, while feeling more and more frustrated.
  • Use the search function to find exactly what he is looking for.

Which choice do you want him to make?

You want him to search, find the product, and make a purchase all while delivering a great experience.

However, just having site search isn’t enough; all site search isn’t the same.

There are some features you can use to make sure your visitors have a great user experience and easily find the products they want to buy.

Here are some of the features your site search box needs to include:

Typo-Tolerant Autocomplete

Your Autocomplete search has to be able to decipher typos. These occur in nearly 30% of searches and even more frequently on mobile. If your Autocomplete can’t, it’s going to lead shoppers to “no results found” pages – even if you sell the product they want to buy.


Ranking algorithms apply to product and keyword suggestions in the Autocomplete, as well as the products on the Search Engine Results Page.

These algorithms take into consideration both the relevancy of the results along with the popularity to provide the best results to shoppers. In some cases, search providers have a portal where you can edit these values to “boost” certain products you want to show.


Your business can use machine learning to personalize your visitors’ experiences. For example, personalization could capture a user’s previous search history to show more relevant results to them on their current visit. As your site search function remains in place, your users will receive better results.

6% of retail visits that involved AI suggestions led to a 4.5 times greater cart rate and a 500% increase in per visit spend.

Keyword, Product, and Category Suggestions

Your Autocomplete shouldn’t only suggests keywords. But instead, should recommend keywords, products, and categories.

This type of variation means that every shopper can have a great experience.

ecommerce search autocomplete extended with product and category suggestions

People who know the exact product they want, can quickly navigate to the product page. Shoppers who have a vague idea can browse through whole categories. And keyword suggestions can guide shoppers who have a semi-defined idea of the product they want.


Dynamic Filtering for Different Searches

Filters can’t be universally applied to searches. There is not much point in displaying shoe size filters to someone searching for bananas. That’s where Dynamic Filtering comes in.

This means that different (relevant) filters are applied to different searches. It also means that whenever someone starts using the filters, they update so that no combination will lead shoppers to a 0-results page.

A study done by Screen Pages showed that 20% of shoppers who use the search function further refine their search by using filters. This is why it’s important to have a full search-suite, not just a search box.

This is a subset of Faceted Search.

Faceted search, or guided navigation as it’s sometimes referred to, is a search method that utilizes the metadata attributed to a product in a store, providing visitors an opportunity to filter and refine their search queries when looking for specific products.

If you leverage faceted search, you’ll improve your shoppers’ experience by directing them quickly, and easily, to the products they want to buy.

Breadcrumb Trails

Breadcrumb trails keep people from getting lost on your site. They come in a few different versions:

  • History – People are very used to a history breadcrumb trail because it works the same as the back button on their browser. A simple trail shows the previous page, the one before that, and so on back to the beginning of their journey through your website.
  • Category – Trails broken down by category start at the broadest section of your website and then narrow down through different filters. For example, a clothing website could have Home –> Clothes –> Men –> Shirts as a trail through its pages.

SERP Product Result Visualization

Search Engine Result Pages have become standardized with the rise in online shopping. This means that shoppers tend to navigate these SERPs in a certain way and if yours doesn’t follow the best practices, you could confuse shoppers and cause them to abandon your store.

One of the most important things to remember is that all of the product thumbnails on your SERP should be similar – same size, same style, and the product visualized the same way.

standardized ecommerce search result page

Here are other things you should consider when designing your SERP:

Number of Products Per Page

SERP product visualization is customizable – you can determine the number of product results per page and also provide options to let your shoppers change the layout. Google shows ten per page for a reason: people don’t look any farther. We recommend you follow their example.

Keeping results numbers low goes back to faceted search and dynamic filtering. Helping customers find what they are looking for on the first page, in the first slot, is the goal of good search.

Keep in mind: you should display fewer results on mobile because of smaller screens.

Quick View and Add to Cart

Even better than just showing products to shoppers is helping them actually make a purchase.

add-to-cart and quick-view features on category pages

The add-to-cart feature lets shoppers immediately add items to their cart to purchase later. This is a big win for your business because it shortens the sales journey.

Dynamic Product Ranking

Google has been attempting to master search intent for years.

The big idea is your website must show visitors what they want to find. So, when they search, you want to display the best results to your users.

Dynamic product ranking filters your products according to their popularity for the search results. This ensures the customer sees the best search results every time.

Out of Stock Products

There are a few different ways to handle this situation:

  1. Remove them from the results
  2. Show the products anyway with a restock date
  3. Display related products

Zero Result Pages

Landing on a “no results found” page is one of the worst experiences a shopper can have. This tells them that you don’t have the product they want to buy – which means they’ll take their business elsewhere.

You will never completely eliminate them, but you should aim to reduce the frequency as much as possible. Since they’ll always occur, follow these best practices when designing your “no results found” pages.

Related Searches

Including Related Searches on your SERP helps your shoppers easily navigate to other keywords/products or refine their initial query. This feature suggests relevant alternative, or complementary keywords and products to shoppers.

Related Searches can appear as keyword suggestions above/below the SERP or as product suggestions below the SERP or on 0-result pages.

Deep Query Understanding

Understanding customer intent is difficult, but possible. The better a search solution can understand shopper intent, the better results (and user experience it can provide).

Search engines often leverage Natural Language Processing and AI to get better over time.

Some search providers also implement Synonym Management to further improve the quality of their results.

Frequent Data Refresh

Your website could change quickly when people are shopping. Stock might run low, a fad might suddenly cause one of your products to skyrocket in sales, or a global pandemic could lead to a big spike in demand for your personalized face masks.

If you don’t have frequent data refresh, your catalogue could easily become outdated and confuse shoppers.

Frequent data refreshing is also important if you leverage dynamic ranking (which you should) – this ensures product and keyword suggestions stay relevant.

Detailed Site Search Analytics

Google Analytics doesn’t provide enough insight when it comes to eCommerce site search – you need a solution with dedicated analytics.

This powerful tool lets you plan more product lines, identify weak products, see the language your customers use, and reduce 0-result search rate (among many other things!) and keep an eye on your customers’ shopping behaviour.

In general, site search analytics can be broken down to these three areas:

  • Search Engine Analytics tell you how many searches visitors conducted, what the results were, how many visitors clicked on a result, what rank the result was, and many more data points. Check out this guide to find more information on Search Engine Analytics.
  • Autocomplete Analytics reveal how users interact with the Autocomplete function.  Reports include data about suggestions usage, usage at position one, search box abandonment, conditional usage, and more. You can find more detailed information here.
  • Related Search Analytics provide data about popular products, popular searches, 0-result searches, engagement, cart actions, and search term details can help you tweak your search engine results pages to show your customers more of what you want them to see. More information is available here.

The data in these reports can help you decide which products to stock more of, how to name products, and information to include in marketing campaigns. This are just some of the ways you can use Site Search Analytics to boost your business!

Key Points

Let us recap a few key points to keep them fresh in your mind:

What is on-site search?

On-site is more than a simple query box. It encompasses Autocomplete, Related Searches, and Semantic Search Engines. It’s the best way you can enable shoppers to make a purchase on your site. And for you, it is a window into the mind of your customers. You can use it to understand their shopping desires and you can use its data to plan new product strategies and refine your business model.

How can search functionality be improved?

If you don’t have a site search function – get one! If you already have one, check that it has the features in this guide. And always keep an eye on your site’s analytics to find out the specific improvements you can make.

What features should an e-commerce website have?

The basic features your search should include are: typo-tolerant Autocomplete with keyword, product, and category suggestions, dynamic filtering, a cohesive SERP, Related Searches, effective 0-results pages, and detailed search-specific analytics.

Now that you’ve seen how much site search can impact your revenue, we hope you’re ready to optimize yours. If you implement these 11 conversion posting features, we’re sure you’ll see an increase in your conversion rate and revenue in no time.

Paige TyrrellHead of Marketing – Prefixbox

Paige is the Head of Marketing at Prefixbox, a leading eCommerce site search solution. She’s an American who’s been living in Budapest since 2017 and loves giving #alwayslearning sessions to help people optimize their online stores.