Skip to article content

Website traffic has dropped: how to check SEO and measurement

If your website records fewer visits, check whether fewer people are arriving or whether the way visits are counted has changed. Learn what to review in Google and your website’s analytics tool.

Name the metric that has fallen

The monthly report shows fewer visits and you wonder whether your search visibility has declined. First establish which metric fell, in which tool and across which part of the website.

Analytics visits, Google Search clicks and received enquiries describe different things. A fall in one does not establish that the others declined by the same proportion.

If you do not have access to these reports, the team managing your website can help you check them. Start by noting when you noticed the decline and any recent changes you know about.

Keep the report view with its dates, filters and comparison. If someone sends a screenshot, retain the metric name too. This avoids investigating a different decline from the one actually observed.

Search Console and Analytics count different things

Search Console shows how often your website appears on Google and how many clicks it receives. Analytics records what people do within your website. Each tool observes a different part of the visit, so their figures need not match.

Align dates, pages and segments, then consider measurement differences. Google explains their combined use in its guide to Search Console and Analytics for SEO.

Check report time zones too, and compare search traffic with search traffic rather than all sessions. A campaign ending may reduce total visits without reducing organic clicks. Measurement changes and an actual traffic decline can also happen at the same time.

Use each source for the question it can address. A difference between tools calls for examining collection and interpretation; it does not by itself establish that one is wrong.

Bring the signals into an investigation worksheet

What to review before proposing changes
AreaRecordQuestion to investigate
Observed declineTool, metric, dates and filters.What exactly are we comparing?
ScopeAffected pages, channel, device and country.Is the change widespread or concentrated?
SearchClicks, times your website appears in results and searches that show it.Has Google visibility or interaction changed?
MeasurementTagging, consent or configuration changes.Is activity being recorded in the same way?
Website and contextWebsite updates, errors, moves to another system and trading calendar.What should be compared with the change date?

The worksheet directs checks. A release coinciding with a decline is a lead to investigate rather than evidence of causation.

An example: analytics falls without a matching fall in clicks

Fictional scenario without real figures: after a consent-management change, Analytics records fewer sessions while Search Console shows no equivalent click decline.

Start by reviewing the implementation and measurement reports. Establish what changed, when tags run and which segments are compared. The review must respect user preferences; it is not a reason to enable analytics without consent to recover higher totals.

This scenario does not prove that actual visits stayed unchanged or rule out other issues. It does identify a focused question to address before rewriting content or changing every page.

If search also falls, narrow down where and when

Establish whether the decline affects the entire site or particular pages and queries. Compare periods in their trading context: demand can differ between weeks. Google’s traffic-drop troubleshooting guide covers technical problems, demand changes and other factors worth distinguishing.

In Search Console’s performance report, compare periods and review the pages and queries with the largest changes. Falling impressions and clicks call for investigating visibility and demand. Steady impressions with fewer clicks also warrant reviewing how results appear. These are investigation leads, not automatic diagnoses.

Access errors or indexing problems call for reviewing the affected pages. If slowness is also reported, document which journeys are affected without automatically attributing the traffic decline to them.

Keep the initial reference before making several adjustments. Simultaneous changes make it harder to understand the situation and assess individual interventions later.

Choose the next check using what you know

A useful outcome is a more precise description: “These clicks fell on these pages from this date,” or “Session collection has changed and needs reviewing.” Causes may remain open, but the investigation has a clearer starting point.

Nodo91 can review search visibility and technical signals. If AI-generated answers are another concern, use a separate method for observing ChatGPT visibility. Bring both needs into your priority list.