> ## Knowledge Base Index > Fetch the complete knowledge base index at: https://support.seotesting.com/sitemap.xml > Use this file to discover available pages before exploring further. > Pure-Markdown content can be obtained by appending a '.md' suffix to the content URLs listed in the sitemap (without the trailing slash). # Interpreting your test results Running an SEO test is only useful if you can correctly read what the results are telling you. A lift in clicks might look positive but be within normal variance; a drop might look alarming but be explained by a broader site trend. This guide walks you through how to read SEOTesting's test results - including the normalised metrics, the AI summary, and what the charts are actually showing - so you can make confident decisions about whether a change worked and what to do next. ## How to view your test results: From the Launchpad, click 'Tests' on the left-hand toolbar. Here you will see all of your tests and their results listed. From here, you can get a really quick **AI summary** of any of your tests. To do this, click on any test and hit the 'AI Summary' button on the top right: ![](https://storage.crisp.chat/users/helpdesk/website/-/3/4/3/d/343dcfd35d951400/ai-summaries-test_102gvc8.png =1100xauto) You'll get back a full analysis that covers: * A clear verdict: successful, inconclusive, or failed - with an explanation of why * A metric-by-metric breakdown across clicks, impressions, queries, CTR, and position * Seasonality checks - if your test ran over Black Friday or Christmas, it'll flag that * Causation vs correlation analysis - the bit that's hardest to do yourself * Actionable SEO takeaways you can apply immediately ![](https://storage.crisp.chat/users/helpdesk/website/-/3/4/3/d/343dcfd35d951400/ai-test_16bc9w3.png =1006xauto) This can be downloaded as a formatted PDF by selecting the 'Download PDF' button at the bottom of the report. There's also a client-ready summary (around 150 words) you can copy with one click and drop straight into an email, a Slack message, or a LinkedIn post etc. Back on the main test results screen, you'll see a results table that presents the main performance metrics SEOTesting tracks and measures: * **Clicks Per Day**: the number of clicks the test pages/queries get daily. * **Site Clicks Per Day**: the number of clicks all pages on the site get from Google. * **Impressions Per Day**: the number of impressions the test pages/queries get daily. * **Average Position**: the average position of the query/pages. * **Click**-**Through Rate**: the CTR of the test query/pages. * **Queries Per Day**: the number of queries the test pages rank for per day. ![](https://storage.crisp.chat/users/helpdesk/website/-/3/4/3/d/343dcfd35d951400/screenshot-2026-04-16-101806_1nouzms.png =1003xauto) When Google Analytics 4 tracking is enabled, additional GA4 data appears below the results table, providing deeper insights into user behavior. The SEO test results page also shows a graph for each metric to visualize its performance over time, with a vertical line marking the test start date. You can switch on to view Clicks, Impressions, Position, CTR and Queries by clicking the relevant ones for you (located under the x-axis of the graph): ![](https://storage.crisp.chat/users/helpdesk/website/-/3/4/3/d/343dcfd35d951400/screenshot-2026-04-16-101908_e7lm7.png =847xauto) At the bottom of the page, you'll find statistical calculations that help determine the significance of your test results. This includes the p-value to interpret the statistical validity of your findings: ![](https://storage.crisp.chat/users/helpdesk/website/-/3/4/3/d/343dcfd35d951400/screenshot-2026-04-16-102002_1cqgig8.png =991xauto)\* **The p-value helps answer this question:** “If there was actually no real difference between the control pages and test pages, how likely is it that we would still see a result like this just by chance?” A low p-value suggests the difference between the control and test groups is unlikely to be random. A high p-value suggests the result could quite easily have happened by chance. In SEOTesting, a common way to think about it is: Low p-value = more confidence that the change may have had an effect. High p-value = less confidence; the difference could be noise or natural fluctuation.