What this storefront test checks
Enter your public Shopify store domain to run four checks against its predictive search endpoint. The test looks for a reachable response, a misspelled catalog term, a descriptive query and a singular or plural variation. You can see the evidence returned by the endpoint directly below each check. No account, email address or Shopify installation is required to see the result.
The service reads a small public product sample to select a term. If that sample is unavailable, it uses a fallback term instead, which makes the findings less representative. A nonzero response shows that products were returned; it does not establish that those products match the shopper’s intention. Open the storefront and inspect the products before deciding what to change.
Read the score in context
Each of the four checks has equal weight. Four passes produce 100, three produce 75, two produce 50 and one produces 25. Scores of 80 or more indicate a strong sample, 50 or more a mixed sample, and lower scores a weak sample. These bands describe these checks only. They are not a benchmark against other stores or a measurement of conversion performance.
A blocked endpoint, password page, redirect or temporary network problem can prevent the scan. In that case there is no score or revenue scenario. The service reports its own evidence strings; a reported zero can also reflect a failed individual request. Result counts are capped by the predictive endpoint at eight. The plural check does not include a count, so none is added here.
Turn a sample into a useful next step
Optional monthly sessions, conversion rate and average order value create an illustrative lost-revenue scenario. The calculation uses the failed-check proportion as an assumed failure rate, a 43% search share and a 1.8× searcher conversion multiplier. For example, two failed checks mean an assumed 50% failure rate. This is not measured search analytics, an attributed loss or a sales forecast. Leave the fields empty to skip it.
Use the result to choose queries for a broader catalog review. Compare actual product relevance, spelling variants, languages, synonyms and mobile behavior. Then validate changes against real search sessions and orders. The full audit is the next step for reviewing your whole catalog; an optional email request shares this sample with Indexa for manual follow-up.