Why a Website Audit Score Can Change Between Scans
Website intelligenceWhy a Website Audit Score Can Change Between Scans
A changed score does not always mean the website improved or declined. Before comparing scans, check whether they measured the same page under comparable conditions and whether the underlying data was available in both reports.
Check that both scans describe the same page
A URL can redirect, change its canonical destination or serve different content depending on the path. Compare the requested and final URL, the response status and the page type before treating the scores as a before-and-after test.
Make sure one scan is not a mobile landing page while the other is a desktop page, or a staging version while the other is the public site. A different target can produce a different set of findings.
Understand measurement variation
Lighthouse-based scores can vary with browser, network, server and page conditions. Response timing can also change temporarily. A small movement in a score should be interpreted alongside the measured checks rather than treated as a definitive result.
SiteBoost displays the evidence and source available for the scan. When a result is unverified or unavailable, note that limitation instead of assuming it is a new website problem.
- Compare the same URL and device context.
- Review source and status of the result.
- Look at the individual checks behind the score.
- Repeat a test when conditions appear unusual.
Use repeat scans to confirm a pattern
When a change is important, repeat the check and look for a consistent result. Record the deployment or content change that happened between scans so the team can connect the evidence to a real action.
Do not chase tiny score fluctuations when no meaningful user or technical issue is confirmed.
