Most shop owners check their conversion rate, bounce rate, and ad spend every single day. Almost no one, by contrast, regularly opens the analytics for their own search function. Yet that's exactly where the most expensive blind spot in the entire shop is often hiding — in that small search box in the top right corner. Customers who actively search want to buy; they already have a clear idea of what they need. When search lets them down, the revenue that was already practically in the bag simply disappears.
What makes this blind spot especially costly is that it doesn't cost you just any visitor — it costs you the person standing closest to checkout, while the marketing budget keeps running to send ever more visitors straight into that same leak.
Data from the Baymard Institute, which has systematically tested the search functions of major e-commerce sites for years, shows just how big this blind spot can get: on average, the "zero-results" search rate sits at 8 to 15 percent, and in industries like pharma, home improvement, or automotive it's often noticeably higher. Every single one of these zero-result searches cuts purchase likelihood by up to 88 percent. So anyone who isn't regularly checking is losing revenue without even realizing it. Seven warning signs reliably reveal when exactly that's happening.
1. Your search frequently shows "no results found"
The most obvious — and most expensive — problem is the zero-results page. According to Baymard data, search performance falls below an acceptable level at 61 percent of the online shops studied, and roughly 15 percent of all search queries fall into query types the system simply doesn't understand. For the customer, this feels like asking a salesperson for help in a store and getting nothing but a shrug in response. In that moment, they don't just abandon the search — in many cases, they abandon the entire shop.
Try it yourself: type a product name into your own search with a small typo, or phrase a query the way a customer would, not the way it appears on the product spec sheet. If an empty results page shows up more than occasionally, that's not an isolated incident — it's a system hitting its limits.
2. Typos and synonyms get ignored
People type faster on their phones than they can proofread, and they rarely search using the exact wording found in the product catalog. This is exactly where one of the most underestimated problems in site search lies: Baymard found that on 69 percent of the shops tested, autocomplete fails users on even slightly misspelled search terms — meaning customers end up searching right past your own assortment without realizing it. The problem is even more pronounced with synonyms: on roughly 60 percent of the shops studied, searching for a common synonym returns no results at all, even though the matching product has been in stock the whole time.
A customer typing "hoodie" instead of "sweatshirt," "laptop" instead of "notebook," or spelling a brand name slightly differently should never see an empty page. If that's exactly what's happening on your site, you're not losing customers because the product is missing — you're losing them because search doesn't recognize it.
3. Autocomplete is missing or poorly implemented
Suggestions that appear as you type sound like a nice convenience feature. In reality, they're one of the most powerful revenue levers a search function can offer at all. According to Baymard, well-built autocomplete extends the average search query from 1.7 to 3.3 words — and each additional word in a query signals clearer purchase intent, raising the likelihood of a purchase by roughly 15 percent. Altogether, Baymard reports revenue gains of up to 24 percent from functioning autocomplete alone.
The catch: autocomplete is now present on roughly 80 percent of all e-commerce sites, but only 19 percent actually implement it to established best practices — with product images, category hints, correct spelling correction, and sensibly ordered suggestions. Autocomplete that just lists random terms pulled from the database barely helps the customer and simply gets ignored.
4. Customers type the same search multiple times, with variations
One especially telling pattern in search logs: the same customer searches multiple times within seconds using slightly different terms. In Baymard's usability tests, users typically tried three to five different phrasings before abandoning a page entirely when no matching results appeared. These search iterations are essentially a loud cry for help from the customer — they're actively looking for the product but can't find it right away.
Take a look through your own search logs: do sequences like "men's shoes," then "mens shoes," then "shoes for men" show up suspiciously often within the same session? That's not a minor detail — it's a clear signal that your search logic isn't sufficiently merging synonyms, word stems, and rephrasings.
5. Mobile search is its own chapter — and usually not a good one
On smartphones, every weakness of desktop search gets amplified, because there's less room for results, filters, and suggestions, and the on-screen keyboard often blocks half the visible area. Baymard's mobile UX benchmark reveals persistent weak points: overly complex filter panels, cluttered result cards, and controls that are barely tappable on a touchscreen. On 94 percent of the sites tested, users can't continue searching within the category they're currently browsing — a detail that sounds minor but causes constant frustration in daily use.
This friction shows up clearly in abandonment numbers: purchases on smartphones are abandoned even more often, at roughly 74 percent, compared to about 67 percent on desktop. Part of that gap traces directly back to a search experience that simply wasn't designed with the small screen in mind.
6. Filters and facets confuse more than they help
A search function doesn't end with the results list — filters are what turn a rough set of matches into a precise result. When important filter options like price, size, color, brand, or availability are missing, or hidden so well that users never find them, customers are left scrolling through a far-too-long, unsorted list. Search usability research attributes a large share of all search abandonment specifically to missing or poorly designed filter options — ahead even of problems with the relevance of the results themselves.
A simple practical test reveals whether this applies to you: search for a broad term like "jacket" or "screw" and count how many clicks it takes before you see a set of filters relevant to you. If it takes more than a glance, chances are high that many customers give up before you do.
7. You're not analyzing your search data at all
Perhaps the biggest problem, precisely because it's the most invisible: many shops collect no reliable metrics on their own search function whatsoever. Yet analysis from Nosto shows that 69 percent of online shoppers head straight to search as soon as they land on a page, while at the same time 80 percent leave the shop entirely after a poor search experience — a share reportedly responsible for around 39 percent of total bounce rate. Globally, abandoned searches are estimated to cost over two trillion US dollars in lost revenue annually, with roughly 234 billion dollars of that in the US alone, according to a survey by Google Cloud and Harris Poll.
At the same time, this exact user group is the most valuable in the entire shop: according to eConsultancy data, customers who actively use search convert roughly 1.8 times better than the average visitor, and according to an analysis by Hello Retail, they can account for up to 31 percent of total revenue — often while making up only 20 to 30 percent of total site visits. A shop that doesn't know these numbers can neither recognize nor deliberately capture its biggest revenue potential. Without reporting on zero-result rate, click-through rate within results, and search-to-purchase conversion, marketing, category management, and IT are essentially flying blind past one of the most important touchpoints in the entire customer journey.
Getting started often takes just a handful of metrics that can be set up in most shop and analytics systems without much effort: the weekly zero-result rate, a ranked list of the ten most frequent unsuccessful search queries, the click-through rate within the results list, and the conversion rate of searchers compared to non-searchers. This narrow set of numbers is usually enough to identify the biggest leaks in the search funnel before investing in a larger technical solution.
What these seven signs have in common
What stands out across all seven points is that they almost never have anything to do with missing products. The item being searched for is, in most cases, sitting right there in stock or in the catalog — search simply doesn't find it, doesn't display it convincingly enough, or leaves the customer running into dead ends unnecessarily often along the way. That's also the good news: unlike a genuine assortment problem, a weak search function can be fixed with manageable effort — through better synonym management, working spell correction, thoughtful filters, mobile-optimized handling, and above all, reporting that shows exactly where in the search process customers are actually dropping off.
If you recognize these seven signs in your own shop, you don't need to replace your entire search system right away. The first and most effective step is simpler: honestly review your own search logs, list out the most frequent zero-result queries, and click through your own search as a customer would. In most cases, it quickly becomes clear at which of these seven points revenue is currently disappearing silently.