Compare your analytics traffic against Search Console clicks for the same period. They will never match exactly, but they should move together. If one fell and the other didn't, the problem is in the counting — and every conclusion you draw before fixing that will be wrong.
Run this first
This is the cheapest check in the entire series and the one most often skipped. A measurement change produces a chart identical to a real decline, and firms spend months and real money chasing a problem that exists only in the reporting.
Do it before you diagnose anything else.
The comparison
Pull two numbers for the same date range:
- Analytics: organic search sessions or users.
- Search Console: total clicks.
They will not match. They count different things — Search Console counts clicks from Google only; analytics counts sessions from all search engines, and applies its own definitions. A gap of twenty or thirty per cent is unremarkable.
What matters is whether they move together. Plot both across twelve months.
Both fell. The decline is probably real. Continue diagnosing.
Search Console flat, analytics fell. The visits are still happening and something stopped counting them. Measurement problem.
Analytics flat, Search Console fell. Less common. Check whether Search Console is still verified and whether a property or domain change occurred.
What breaks counting
The tracking code stopped firing. A theme update, a plugin conflict, a migration, or a consent tool loading in a different order. Check the code is present on every page type — not just the homepage. Service pages and blog templates are where it goes missing.
A consent banner changed. Analytics that only fire after consent will report a fraction of actual traffic, and that fraction changes whenever the banner's design, defaults, or wording change. A banner redesign can look like a forty per cent traffic collapse.
Someone switched properties. A move between analytics properties or platforms resets history. Comparing across the boundary compares two different measurement systems.
Filters or bot handling changed. An internal-traffic filter added incorrectly can exclude a whole region. Changes in automated-traffic filtering shift the baseline.
Multiple tracking codes. Two installations double-count, which is a nastier version of the problem because the numbers look better rather than worse.
Cross-domain gaps. If part of the journey sits on another domain — a booking tool, a payment page, a vendor-hosted section — sessions can fragment.
The date question
If the drop happened on a specific day, the cause is almost certainly technical. Real search declines are gradual; they reflect ranking movements that unfold over weeks.
A cliff on a single date is a measurement change until proven otherwise. Find what shipped that day. Plugin updates, theme changes, DNS changes, consent tool changes, a new tag manager container.
What you can and can't conclude
Verified — Whether the two sources diverge. Both are yours and the comparison is direct.
Directional — That divergence indicates a measurement problem. Very likely, but not certain: a shift toward search engines other than Google would show a similar pattern for a real reason.
Pending — What broke. Identifying the specific cause means auditing the tracking implementation, the consent configuration, and the change history together. The comparison tells you that something is wrong, not what.
The habit worth starting
Keep a change log. Every plugin update, theme change, and configuration change, with a date, in a shared document rather than one person's memory.
Most of this post becomes a five-minute check if you have one, and an archaeology project if you don't.
Ruling out measurement is the first move on any traffic decline. The full map is at how do I tell if my website is losing me clients.