Internal search and filters: the conversion drivers your online shop isn’t making the most of

Introduction

There is a category of visitors to your shop who convert two to six times better than others. They know what they want, they have a clear intention to buy, and they tell you so explicitly. These are the visitors who use your search bar.

And yet, in the majority of e-commerce shops, the internal search engine is the least well-maintained, least monitored and least invested-in component. When was the last time you looked at your internal search report in GA4?

This article gives you the keys to diagnose, optimise and transform your search and filtered navigation into real conversion drivers — with concrete actions you can implement today.

1 – Visitors who are searching are your best customers (and you’re not paying attention to them)

Before you optimise anything, you need data. The first step is to check that internal search tracking is enabled in GA4 — without this, no data is collected. The setup process takes less than ten minutes and doesn’t require a developer.

Once set up, here are the four metrics that really matter.

The 4 metrics to prioritise

The most frequently searched terms reveal what your customers want and what you may not have highlighted sufficiently in your catalogue or navigation. This is a goldmine for your merchandising decisions.

The ‘zero results’ rate is your direct indicator of frustration. Every search with no results is a potentially lost sale. The acceptable benchmark is below 5% — anything higher requires immediate attention.

The post-search exit rate measures how many visitors leave the shop after viewing search results. If this rate exceeds your overall exit rate, your results are not meeting your visitors’ intent. The search vs non-search conversion ratio is your key metric for justifying investment in the search engine. This figure, calculated in two minutes in GA4, is often the deciding factor.

Immediate warning signs

Three situations require immediate action: a ‘zero results’ rate exceeding 10%, a post-search exit rate higher than your overall exit rate, and recurring spelling mistakes that your search engine cannot handle — for example, ‘robbe’ instead of ‘robe’, or ‘jean slim homm’ instead of ‘jean slim homme’.

2 – The 5 search features that really boost conversion rates

You don’t need to do everything at once. Here are the five high-impact features, ranked by priority.

Predictive autocomplete

Displaying suggestions as soon as the first or second character is typed reduces cognitive effort and guides the visitor towards relevant results before they’ve even finished typing. Ideally, the dropdown should offer products, categories and popular searches — and display a thumbnail image for each suggested product. This last point, often missing from basic solutions, makes a real difference to the click-through rate to product pages.

Tolerance for typos

A high-performance search engine recognises ‘shoe’ and ‘shoes’, ‘nike’ and “NIKE”, ‘T-shirt’ and ‘tshirt’. This is the bare minimum — and yet it is often missing. PrestaShop’s native search can handle synonyms, but only if they are configured manually and maintained regularly. Every unmanaged typo is a potential loss of a visitor.

“Did you mean…” suggestions

For typos that aren’t automatically corrected, offering an alternative rather than a blank page makes all the difference. A blank results page with no suggestions is a dead end — the visitor leaves, and often for good.

Filters in search results

Being able to refine search results without starting from scratch is essential for large catalogues. This is one of the most common shortcomings — including on PrestaShop, where filtering of search results is not available natively. This functionality is covered by solutions such as Atomic Suite (202).

Customising search results

Highlighting bestsellers, in-stock items and new arrivals in your search results transforms your search engine into an active merchandising tool. The order of the results is not a neutral factor — it is a business decision.

3 – Faceted navigation: turning your catalogue into an exploration tool

Internal search caters to visitors who know exactly what they’re looking for. Faceted filters serve those who want to browse — and often, they need both.

But before you optimise your filters, ask yourself the right question: do you have enough of them? In most shops, the problem isn’t having too many filters — it’s having too few. A visitor looking for waterproof women’s hiking boots who can’t find a ‘waterproof’ filter will give up. You lose a sale not because of your product, but because of access to that product.

To identify the relevant filters for each category, start by analysing your internal search data: what attributes do your visitors type in? Supplement this with an analysis of usage patterns in your sector. Once the filters have been identified, it’s often a matter of enriching your catalogue — adding the missing attributes to your products.

What visual filters offer

Colour attributes displayed as swatches — coloured dots, rather than a drop-down list — significantly enhance the user experience for visual products. But beware of a technical point that is often overlooked: if you offer a colour filter, the product image must change to reflect the selected colour. This behaviour is not native to PrestaShop — it is supported by solutions such as Advanced Search 5 or Atomic Search (202). This also means that the retailer must have correctly entered the images for each colour variant in their catalogue.

‘In stock now’ filters prevent a common frustration: discovering that the featured product is out of stock. This is a filter you should always enable as soon as you start managing stock.

How to organise your filters

Prioritise filters based on how often they’re used, as measured in your analytics. Display a maximum of 5 to 7 filters in the visible section — the rest under ‘see more’. On desktop, filters in the sidebar should remain visible when scrolling (sticky position); on mobile, a ‘Filter’ button triggers an overlay — this is the only truly usable solution on small screens.

4 – The SEO pitfall of faceted navigation (and how to avoid it)

This is the hidden side of faceted navigation that many retailers only discover too late. Every combination of filters generates a unique URL — which can result in tens of thousands of indexable pages for a catalogue of just a few hundred products. The result: wasted crawl budget, duplicate content, and keyword cannibalisation.

Conversely, well-managed faceted navigation can become a massive SEO asset. The ASOS case is often cited as a benchmark: over 3.8 million organic keywords indexed, a significant proportion of which is down to careful management of the indexing of filtered pages.

The 3 technical solutions

The canonical tag: all filter variations point to the main category URL. Simple to implement, but does not allow you to capitalise on high-search-volume combinations.

The noindex parameter: prevents the indexing of URL combinations with low SEO value, whilst still allowing crawling. This approach is more nuanced and effective for large catalogues. The robots.txt file: directly blocks the crawling of URL parameters generated by filters. This is a radical solution, to be used only in cases where the volume of low-quality pages is critical.

The recommended approach on PrestaShop

PrestaShop does not offer any native functionality to manage this. The available solutions rely on specialised modules: Advanced Search 5, for example, allows you to create static pages (pseudo-categories) based on a faceted filter — enabling you to strategically index certain high-potential combinations. At 202, we have developed an equivalent feature integrated into Atomic Suite.

The logic remains the same regardless of the solution: identify which filter combinations correspond to a real search volume, create dedicated landing pages for those (e.g. ‘men’s running shoes size 42’), and block the indexing of the rest.

5 – The ‘No results’ page: turning a dead end into an opportunity

This is one of the most under-optimised pages in the entire e-commerce sector. Every unoptimised ‘no results’ page is a lost sale — and sends a strong negative signal about your shop’s user experience.

A good “no results” page does more than simply display “0 products found”. It actively offers a way out of the impasse: a suggested rephrasing (“you may have been looking for…”), a selection of popular products from the closest category, alternative categories to explore, and ideally a contact option or availability alert.

The technical reality with PrestaShop

There is no native solution in PrestaShop for customising this page. SaaS solutions such as Doofinder or Algolia generally include this feature. For merchants who do not have a SaaS search solution, the options are either a bespoke development — which is technically simple and quick — or, for 202 clients, the implementation of Atomic blocks configured to display on empty results pages.

Choosing a search engine for PrestaShop

PrestaShop’s native search function may suffice for very small catalogues — but its limitations are quickly reached. To allow for typos, you need to manually configure synonyms, which must be updated regularly. Autocomplete does not display product images in the drop-down menu. Results are limited to products, without showing relevant categories. It is not possible to sort the results. And search results cannot be filtered — a significant shortcoming as soon as the catalogue exceeds a few dozen items.

To take things further, here is the range of available solutions:

Elasticsearch is a high-performance open-source solution, but it requires dedicated hosting and significant technical configuration. It is best suited to teams with in-house development resources.

Algolia is often cited as the go-to SaaS solution for e-commerce search. It delivers on performance, but the solution requires a significant investment in configuration and comes at a high cost — it’s a bit like the Klaviyo of search. Suitable for medium to large catalogues with dedicated technical teams.

Doofinder is an e-commerce-first SaaS solution: quick setup, integrated analytics, and easy to get started with. Well-suited to merchants who want results without advanced technical expertise.

Motive is a direct alternative to Doofinder, to be considered depending on the available integrations and the context of each project.

Key selection criteria: catalogue size, budget, available in-house technical resources, and desired level of customisation.

Conclusion

Internal search and faceted navigation are probably the most underutilised CRO levers in your shop — and among the quickest to deliver results, because they precisely target your most qualified visitors.

Start with an audit: open the internal search report in GA4 and note your ‘zero results’ rate. If it’s over 10%, you have a priority action to tackle today.

Want to know what your customers are really searching for on your shop? The 202 e-commerce teams will audit your internal search and faceted navigation to identify the conversions you’re missing out on.