Traffic dropped: what do you check first, and in what order?

Check a traffic drop in this order: is the data real, when did it start, where did it land, then technical, security, update and demand causes. Google does not number its checks, so this order is our reasoning: confirm the data, locate the loss, then test causes that have their own report before those that need judgment.

• Google says an impressions logging error ran from 13 May 2025 until 27 April 2026 and affected impressions, CTR and average position, but not clicks (Google, Search Console Help, read 30 October 2026).
• Google's debugging guide tells you to select Last 16 months in the Date filter to check that a drop is not one that happens every year (Google, Search Central, read 30 October 2026).
• After a core update, Google recommends waiting at least a full week once it completes, then comparing that week with the week before it started rolling out (Google, Search Central, read 30 October 2026).
• Penlid's anomaly check measures across 27 April 2026 using 28 whole days on each side, and never uses the newest day in a file (Penlid, Search Console anomaly check, read 30 October 2026).

Why check a traffic drop in this order?

Check in this order because a data error would make every later check wasted work, and later checks need the dates and location that earlier ones supply. The sequence is our reasoning, not Google's: its guide to debugging traffic drops is organized by cause and does not say which cause to check first.

Is the drop real, or is it the data?

Treat the drop as confirmed only when it holds on complete days, nothing on Google's data anomalies page covers your dates and metric, and no system annotation flags a reporting issue. Search Console shows complete days by default, and Google says the newest data can be preliminary and may change in the next few hours; see why the last few days change.

Google's debugging guide also says the drop might be related to a change in the data processing or a logging error and points to the data anomalies page, which records known issues from the last 3 to 16 months. Read what each entry affects: the impressions error that ran from 13 May 2025 to 27 April 2026 did not affect clicks, so it does not explain a fall in clicks. Our guide to Search Console data anomalies covers the other entries. System annotations, which Search Console adds for data processing and reporting issues, do not show in comparison mode.

Penlid's free Search Console anomaly check does the date matching: it lists every documented anomaly for the report you pick that overlaps your export's dates and measures whether your own file carries the impressions logging error, in your browser.

When did the drop start, and does it happen every year?

Set the Date filter to Last 16 months, then compare periods, to find the first day of the fall. Google says this helps you make sure the drop is not one that happens every year because of a festivity or a trend.

The Compare tab offers the last 3 months against the previous period and against the same 3 months a year earlier. That year-over-year view needs 3 + 12 = 15 months of history, inside the 16 months the Date filter covers; a 12-month comparison would need 12 + 12 = 24. For two date ranges, Google says to consider weekly or monthly granularity, because aggregating by week or month neutralizes the day-of-the-week effect.

Write down the first date on which the loss is visible: every later check asks what changed on or just before it. For older data, see keeping Search Console data past 16 months.

Where did the drop land?

Find where the loss sits by splitting it by search type, then by page, query, country, device and search appearance, because the location narrows the causes. Google says to analyze search types such as web, images, video and news separately.

Sorting the Pages tab by Clicks Difference puts the biggest losers first. A drop across the whole site points to the Page indexing report, and a drop on a group of pages to the URL Inspection tool. If impressions held steady while clicks fell, Google says your page title and snippet may not be the best they could be, or other sites may have had a more appealing rich result; see impressions up and clicks down.

To tell one decaying page from a site-wide move, see content decay versus a site-wide dip. Penlid's content decay check subtracts the site-wide move over the same window, as the methodology describes.

Is crawling, indexing, security, a manual action or a site move behind it?

Any of them can be, so check the Page indexing and Crawl stats reports for changes that match the start date, then Security issues and Manual actions. Google says errors that stop crawling, indexing or serving can be site-wide, such as a website that is down, or page-wide, such as a misplaced noindex tag, with a slower decline.

Security issues covers hacked content, malware or unwanted software, and social engineering. Manual actions lists actions by a human reviewer and shows a green check mark when there are none. Google says most reconsideration reviews can take several days or weeks. After URLs change in a site move, it says ranking may fluctuate while Google recrawls and reindexes, and a medium-sized site can take a few weeks for Google to notice the change, a larger one longer.

Did a Google ranking update overlap the drop?

Compare the drop's start date with the ranking updates on the Google Search Status Dashboard, which lists each with a start date and a duration; an overlap is a lead, not a finding.

For a core update, Google says to confirm the rollout has finished, wait at least a full week after it completes, then compare that week with the week before it started rolling out. Google also makes smaller core updates continually and does not announce them, so an empty dashboard does not prove that nothing changed.

Overlap is timing evidence, not proof: Google says other events can also contribute to changes in performance. See measuring what a Google update did to your site.

Could seasonality or lower demand explain it?

Yes, if the same weeks fell in earlier years or searches for your topic fell everywhere. Filter to one high-traffic query at a time and check it on Google Trends, which Google says helps you understand whether the drop was only for your website or throughout the web; separating seasonality from a real SEO change shows the method.

What should you do before and after changing anything?

Before changing anything, size the drop; after changing something, wait before judging it. In its core update guidance, Google says a small drop in position (its example is position 2 to 4) needs no drastic action and recommends avoiding changes to content that is already performing well; for a large drop (position 4 to 29) it says to do a deeper assessment.

After a large drop in position, Google's debugging guide says some changes take effect in a few days and others could take several months, that you will likely want to wait a few weeks before analyzing your site in Search Console again, and that there is no guarantee a change will have a noticeable impact.

If a client is asking, say which checks you ran, in order, and what each showed. See monthly SEO report contents, proving SEO results to a client and forecasting without overpromising.