Was it the update? Separate rankings from seasonality and data errors

Before blaming a Google update for a traffic change, check measurement, then seasonality, then the update timeline. A five-step method and a short report for leadership.

Organic traffic falls 18% in a fortnight. A Google update was confirmed in the same window. The obvious conclusion is that the update caused it.

The obvious conclusion is often wrong. Before anyone rewrites pages or cuts a budget, a marketing team should run three checks: measurement, seasonality and only then the update itself.

This article sets out a repeatable way to do that in under a day.

Why this matters

Misreading a traffic change is expensive in both directions.

Blame an update that did not affect you, and you may rewrite pages that were working. Blame seasonality when you were hit by an update, and you lose months.

Google’s own core updates guidance warns against quick fixes. Taking a structured approach is the practical version of that advice.

Step 1: Confirm the data is real

Start with the possibility that nothing changed except the measurement.

Google publishes a running list of Search Console data anomalies. It is worth checking before every investigation.

The list shows how often this happens. Recent entries include:

• A logging error that under-reported impressions from 13 May 2025 to 27 April 2026. Clicks were unaffected, and the fix could appear as a decrease in reported impressions.
• FAQ rich results no longer appearing from 7 May 2026, which changes how some results display.
• A Discover logging issue on 21 May 2026, the same day the May 2026 core update began.
• Missing data in bulk exports for 28 February and 1 March 2026.

Also check your own systems. A changed analytics tag, consent banner, site migration or redirect can move numbers on its own.

Quick test: Do clicks and impressions in Search Console move in the same direction as sessions in your analytics tool? If only one source moved, suspect measurement first.

Step 2: Rule out seasonality

Many businesses have strong seasonal patterns that are easy to forget.

Compare the same weeks in the previous year, not just the previous fortnight. A drop that matches last year’s shape is probably demand, not ranking.

Google Trends can help check whether interest in your main topics changed. Use it carefully.

Google says Trends data is normalised to a 0 to 100 scale relative to total searches, is based on a sample, and may include privacy-related noise for low-volume terms. It also says Trends does not include searches made inside AI Mode or AI Overviews.

Google describes Trends as one data point among others. Treat it as a cross-check, not proof.

Quick test: Did impressions fall but average position stay broadly the same? That pattern usually points to lower demand rather than lower rankings.

Step 3: Check the update timeline

Only now look at updates.

Google’s Search Status Dashboard lists confirmed ranking updates with start and end dates. Recent examples include the May 2026 core update, from 21 May to 2 June, and the September 2026 spam update, from 24 September to 8 October.

Line up those dates against your daily data. A real update effect usually starts during the rollout window, not before it.

Google’s guidance suggests waiting at least a full week after a rollout finishes, then comparing with a similar period before it began.

Quick test: Did average position fall for many queries at once, beginning inside the rollout window? That pattern is more consistent with an update.

Step 4: Narrow it down

If the evidence points to a ranking change, find where it happened.

Break the data down by page type, topic, country, device and search type. Google’s guidance notes that web, image, video and news results can behave differently.

A drop concentrated in one section suggests a specific content issue. A drop across the whole site suggests something broader.

Google also distinguishes small movements, such as from position 2 to 4, from large ones, such as 4 to 29. Small changes do not usually justify drastic action.

Step 5: Write it down

Summarise the investigation in five lines for leadership:

1. What changed, and by how much.
2. Whether the data was affected by a known reporting issue.
3. Whether the change matches a seasonal pattern.
4. Whether timing matches a confirmed update.
5. What you will do next, and when you will review it.

This turns a worrying chart into a decision.

An illustrative example

The following example is hypothetical, and the numbers are illustrative.

A UK accountancy firm sees organic sessions fall by 20% in late January. The firm’s outside agency suggests a Google update is to blame.

The team checks measurement and finds nothing unusual. It then compares the previous January and finds a similar fall after the self-assessment tax deadline on 31 January.

Average position is stable. Impressions fell for tax-return queries only.

The conclusion is seasonality. The team keeps its content plan and reviews again in March.

Common mistakes

Comparing the wrong periods. Week-on-week comparisons around holidays or deadlines mislead.

Reacting mid-rollout. Rankings can move during an update. Analysing before it finishes risks chasing noise.

Treating one metric as the truth. Impressions, clicks, position and sessions each tell part of the story.

Ignoring the website itself. Releases, page removals and technical changes often coincide with updates by chance.

Keep a change log

The best defence is a simple shared log. Record site releases, tracking changes, campaigns, price changes and confirmed Google updates by date.

When traffic moves, the log turns a debate into a quick lookup. It also helps the next person who inherits the dashboard.