A zero-result search is one of the clearest signals of lost revenue you have. A shopper told you exactly what they wanted, in their own words, and your site replied with nothing. Most of them leave. The good news is that zero-result searches are measurable, diagnosable and — with a systematic approach — largely fixable.
This article lays out how to find them, work out why they happen, and put fixes in place that keep working as your catalogue and customers change.
Why zero results cost more than they look
When a search returns nothing, the shopper hits a dead end at the exact moment of highest intent. Someone using your search box is usually further down the funnel than someone browsing categories. In our experience, searchers convert at a meaningfully higher rate than non-searchers — which is precisely why a blank result page is so expensive. We make the broader case in Site Search Is Your Highest-Converting Channel.
The damage is also hidden. Most analytics dashboards report overall search volume and conversion, not the specific queries that failed. The failures sit in the logs, quietly losing money.
Step 1: Make zero results visible
You cannot fix what you cannot see. Start by getting the data into one place:
- Pull a zero-result report from your search platform or analytics. Look at the last 30–90 days.
- Sort by frequency. A handful of high-volume failing queries usually account for most of the lost sessions. Fix those first.
- Tag each query by type so you can choose the right fix (see the next section).
- Track the rate over time. Zero-result rate as a percentage of total searches is the single number to watch. A healthy figure is usually low single digits; double digits means real money is leaking.
Set up a weekly or monthly review so this becomes a habit, not a one-off audit.
Step 2: Diagnose why each query fails
Zero results are a symptom, not a cause. In practice, failing queries fall into a few buckets:
1. Vocabulary mismatch
The shopper used a word your catalogue does not. “Trainers” versus “sneakers”, “settee” versus “sofa”, a brand nickname, a regional term. The product exists; the words do not line up.
2. Typos and misspellings
“Hoddie”, “addidas”, “moisturizr”. Without fuzzy matching, these die instantly.
3. Over-specific or natural-language queries
“Black waterproof running jacket size medium” may be too many constraints at once, or phrased as a sentence your keyword engine cannot parse.
4. Genuinely missing products
Sometimes the shopper wants something you do not stock. These are not failures of search — they are signals for merchandising and buying.
5. Data gaps
The product exists but has thin titles, no attributes, or missing translations, so search has nothing to match against.
Tagging your top failing queries against these five buckets turns a vague problem into a prioritised work list.
Step 3: Apply the right fix
Match the fix to the cause rather than reaching for one tool for everything.
For vocabulary mismatch:
- Build and maintain a synonym dictionary mapping shopper language to catalogue language.
- Add redirects for high-value branded or campaign terms.
- For scale, move to semantic search, which matches on meaning rather than exact words and removes much of the manual synonym burden. See Semantic Search for eCommerce.
For typos:
- Enable fuzzy matching and typo tolerance, tuned so it does not become so loose it returns irrelevant results.
- Add “did you mean” suggestions and autocorrect for common misspellings.
For over-specific queries:
- Implement graceful degradation: when an exact match fails, relax the least important constraint and show close matches rather than nothing.
- Use autocomplete and query suggestions to steer shoppers toward queries that will succeed.
For missing products:
- Show a helpful empty state — popular products, best sellers in the closest category, or a way to get notified — rather than a bare “no results”.
- Feed these queries to buying and merchandising as demand signals. A recurring search for a product you do not stock is a free market-research report.
For data gaps:
- Enrich product titles, attributes and descriptions. Search quality is downstream of data quality. This is foundational, and we cover it in eCommerce Data Foundations.
Step 4: Design the empty state to recover the sale
Even with great search, some queries will return nothing. The empty state is your last chance to keep the shopper. A strong one includes:
- A clear, human message — not a stark “0 results found”.
- Relevant suggestions: popular items, recently viewed, or category best sellers.
- A corrected or broadened query offered automatically.
- An easy route to help — chat or contact — for high-consideration purchases.
Treat the empty state as a conversion surface, not an error page.
Common pitfalls to avoid
- Tuning fuzzy matching too aggressively. Returning something irrelevant can be worse than returning nothing, because it erodes trust in your search.
- Fixing queries one at a time forever. Manual synonyms and redirects do not scale past a point. Know when to graduate to a semantic layer.
- Ignoring mobile. Typos and partial queries are far more common on phones; test the experience there specifically. See Close the Mobile Conversion Gap.
- Never re-checking. New products, seasonal demand and campaigns constantly create fresh zero-result queries. This is ongoing maintenance.
Measuring the impact
Tie your work to revenue so the effort is defensible:
- Zero-result rate — should fall steadily.
- Search conversion rate — searchers who go on to buy.
- Revenue per search session — the figure that gets attention in a board meeting.
Run before-and-after comparisons on the specific queries you fixed, and watch the aggregate trend over the following weeks.
The bottom line
Zero-result searches are among the most tractable problems in eCommerce: they are measurable, the causes are well understood, and the fixes range from quick synonym wins to a deeper move toward semantic search. Make them visible, diagnose by cause, fix systematically, and treat the empty state as a recovery opportunity.
If you want help auditing your search logs and building a plan to recover that lost revenue, our AI Search and Recommendations and Conversion Optimisation teams do exactly this. Get in touch and we will start with a look at your own zero-result data.