TapCub is live — analytics, insights and live chat, free on one platform
Glossary

What is search analytics? Search-engine data vs on-site search

Search analytics means two different things: how people find you on a search engine, and what visitors type into your own site search. Definitions, formulas, a worked example for each, and the mistakes that mix them up.

Su He Reviewed by Chloe Morgan Also called search data analysis Last updated Nov 17, 2026 Glossary

Quick answer

Search analytics is the analysis of data generated by searches, and it covers two separate things: people searching on a search engine and seeing your page (impressions, clicks, click-through rate, position), and visitors typing into your own site search (the terms they use).

Formula: CTR = clicks ÷ impressions

What is search analytics?

Search analytics is the analysis of data created by searches. The term is used for two different things, and mixing them up is the most common mistake: data about people searching on a search engine such as Google or Bing and seeing your page, and data about visitors typing into your own site's search box. Both involve search terms, but they answer opposite questions and live in different tools.

Most pages that rank for the term mix the two. Some define it as the study of searches made within a website's own search tool or through external engines, and then cover both in a few paragraphs. This page keeps them apart, because the fix for a problem in one has nothing to do with the other.

Two kinds of search analytics

Search-engine analyticsOn-site (site search) analytics
Who is searchingPeople on Google or Bing, before they know youVisitors already on your site
Question it answersDo we show up, and do people click?What are visitors looking for that they cannot find?
Main dataImpressions, clicks, click-through rate, average positionSearch terms, number of searches, what happens after the search
Where the data comes fromThe search engine's own report, such as Google Search ConsoleYour site, where the term appears in the results-page URL
Typical actionRewrite a title or description, improve a pageFix navigation, add or rename content, add a help article
Common limitDelayed by a day or more; only the search engine has the full list of queriesOnly visitors who use the box; the term is only as good as your search parameter

Search-engine metrics and formulas

Google defines four numbers in its Search Console performance report:

  • Impressions: the number of times someone saw a link to your site in Google results.
  • Clicks: the number of times someone clicked a link from Google to your site.
  • Click-through rate (CTR): clicks ÷ impressions.
  • Average position: a relative ranking where 1 is the top spot. For each query, the position counted is the topmost spot your page occupied, then averaged across queries.

The last definition has a consequence. An average position of 6.8 can hide a query at 4 and another at 22. Always read position next to the queries that make it up. For the term behind impressions, see impressions. For why queries that name your brand behave differently from the rest, see branded vs non-branded keywords.

What happens after the click is a separate matter. Slow or unstable pages lose visitors that search already won; that part belongs to page experience and real-user performance, not to search analytics.

On-site search is a list of what your visitors typed, which makes it the most direct statement of what they want and could not find by clicking. Three readings are worth the time:

  1. Top terms. The most common words show what the navigation fails to surface.
  2. Share of sessions that search. A high share means browsing alone is not working.
  3. Terms with no good result. These need a log from the search tool itself, since only it knows what it returned.

A worked example with both kinds

Sample data, search-engine side: a blog page earns 12,000 impressions and 360 clicks in a month.

  • CTR = 360 ÷ 12,000 = 3.0%.
  • It appears for three queries: 8,000 impressions at average position 4, 3,000 at position 9 and 1,000 at position 22. Weighted by impressions the average position is (8,000 × 4 + 3,000 × 9 + 1,000 × 22) ÷ 12,000 = 81,000 ÷ 12,000 = 6.75. The unweighted mean of 4, 9 and 22 is 11.7. The weighting matters: the headline "6.75" is close to the position of the query that brings most of the visibility.
  • Decision: if the page ranks in the top 10 but its CTR is below your site median, the title and description are the cheapest thing to change before any rewrite.

Sample data, site-search side: a site records 1,000 searches in a week across 8,000 sessions.

TermSearchesShare of searches
pricing14214.2%
refund969.6%
api717.1%
invoice646.4%
All other terms62762.7%

1,000 searches across 8,000 sessions means at most 12.5% of sessions used the box (fewer if some sessions searched twice). If "pricing" is the top term while a pricing link sits in the header, the link is not doing its job on the pages where people search. Two terms, pricing and refund, make up 23.8% of all searches, which is a clear brief for what to surface first.

Common mistakes

  • Treating the two kinds as one report. A fall in Google clicks says nothing about what visitors search for on your site.
  • Expecting keywords in an analytics tool. Browsers rarely pass the search keyword to a third party site, so the organic-keyword column of any analytics tool is thin. Full queries come from the search engine's own account.
  • Reading average position as a rank. It is an average over queries and days, not where you rank for a given phrase.
  • Reading zero data as zero traffic. If no search account is connected, there are no impressions to show. That is a missing source, not an empty one.
  • Mistaking site search for SEO. The words visitors type on your site rarely match the words they typed into Google, but they are an excellent source for help content and navigation.
  • Reading the last 1–3 days. Search engine data arrives late; compare complete periods.

Search analytics in TapCub

Use the two examples above as a checklist.

  1. Connect the search accounts. In Monitoring › Search performance, connect Google Search Console or Bing Webmaster. TapCub then shows clicks, impressions, click-through rate and average position by page, query and country. Search data is usually one to three days late, and the page states the date it is current through. Baidu offers no API for clicks and impressions, so there is nothing to show for it.
  2. Find the pages worth rewriting. Rank buckets group queries by average position, weighted by impressions. On the scatter plot each point is a page, sized by impressions. A page in the top 10 whose click-through rate is below the median is marked, since changing its title and description pays off first.
  3. Check the search terms on Sources. Open Web analytics › Sources and the Search terms tab, which splits terms into organic terms, paid terms (from utm_term) and on-site search terms. Organic terms exist only when the referrer carried the keyword, so that column is often thin by design.
  4. Read on-site search. The website script stores the on-site search term on the pageview when the landing query matches a parameter name. The defaults are s, q and search; edit them in Web analytics › Configuration › Web settings. A wrong parameter name leaves no search term. On-site terms are captured in full.
  5. Know what TapCub does not do. TapCub reports what the search account gives it; the position is an average synced from the search product, not a rank TapCub computes. It does not do keyword research or rank tracking, and it does not rewrite titles for you.

For the thin organic column, the help page why organic search terms look thin has the explanation.

Sources

  • Google Search Console Help, Performance report (Search), definitions of clicks, impressions, CTR and average position, checked 2026-11-17.
  • Elastic, What is search analytics?, definition covering website and search-engine searches, checked 2026-11-17.
  • Pirsch, Search analytics, definition covering site search and external engines, checked 2026-11-17.
  • TapCub documentation, Sources & channels and SEO & content decay (Monitoring › Search performance), checked 2026-11-17.

Was this explanation clear?

We use your answers to fix and expand this page.

Su He

Data Analyst

Explores traffic sources, conversion funnels, and retention, using clear metric definitions and concrete examples to explain what the numbers mean.

FAQ

Questions people ask

Short answers to the follow-up questions that come up most.

Still have a question?

Chat with the team behind TapCub — we usually reply within a few hours on business days.

What is search analytics?

Search analytics is the analysis of data created by searches. It means either data from search engines such as Google, with impressions, clicks, click-through rate and average position, or data from your own site search, with the terms visitors type.

What is the difference between search analytics and site search analytics?

Search analytics on a search engine tells you whether people find and click your pages before they know you. Site search analytics tells you what visitors already on your site look for and cannot find. They use different tools and lead to different fixes.

What does average position mean in search analytics?

In Google Search Console, average position is a relative ranking where 1 is the top spot. For each query it counts the topmost position your page held, then averages across queries, so an average of 6.8 can hide a query at 4 and another at 22.

Why are my organic search terms so few in analytics?

Browsers rarely pass the search keyword to a third-party site, so the organic-keyword column is thin in any analytics tool. Full queries come from the search engine's own account, such as Google Search Console.

See this number on your own site

Add one script tag for web and apps, and the first reports fill in within minutes.

See your data clearly. Find your growth.

Every click, backed by data. Install one line of code and see your first numbers in a minute.

No credit card · Free plans for analytics and chat · Cookieless analytics