The September 2026 spam update has ended at Google, and it was the slowest spam rollout of the year by a wide margin. The Search Status Dashboard now reads, “The rollout was complete as of October 8, 2026.” Search Engine Journal reported Thursday that the run began September 24 and is listed at 13 days 16 hours.
This updates our earlier coverage, including our Sept. 29 report on the first weekend spike and our Oct. 2 report on the second wave. Both warned that early volatility did not mean the rollout had ended. It had not.
The duration is the clearest fact available. Measured against the dashboard, the three earlier 2026 spam updates were far quicker: August took 2 days 16 hours, June took 2 days 1 hour, and March finished in 19 hours 30 minutes. The September run used nearly all of the two weeks Google allowed at launch. It was not a record, though: last year’s August spam update lasted about 27 days.
Google has not said what this update targeted or why it took longer. Search Engine Roundtable’s Barry Schwartz reported that Google described it on LinkedIn as “a normal spam update,” and he questioned that label given the length. Google’s documentation says only that spam updates improve its automated detection systems. For Google’s own account of why spam updates are arriving more often, see our report on its Barcelona explanation.
Schwartz also logged when volatility showed up: September 25 through 27, September 30, and October 4 through 6. These are his observations, and they assume Google shipped no other updates during the period. The September 30 date matches the second wave we covered. Treat all three as leads to check against your own data, not as proof of cause.
Impact data remains thin. Diib, an analytics vendor, studied 32,569 sites with connected Google Analytics and measured how much of each site’s visits came from unpaid Google search over September 24 through 26, set against the matching three days one week before. The median fell from 34 percent to 31 percent, and 10 percent of sites lost half or more of their share, against 5 percent in the earlier period. Diib called this “an early read, not a verdict,” and the data does not identify which sites Google treated as spam.
Rank trackers add noise. SEOmonitor said many trackers showed sudden ranking changes starting September 26 that searchers did not see, and it attributed that to incomplete results Google returned to automated searches. That is SEOmonitor’s explanation, not a Google statement. Its practical point stands: spam-related changes usually hold for several days and appear in impressions and clicks inside Search Console.
Here is how to use that. Set your Search Console comparison to September 24 through October 8, and compare it against an equal-length period ending September 23. Then view the window day by day instead of as one block, because a 15-day rollout blurs a single before-and-after average. A tracker drop that has no matching fall in impressions and clicks is probably tracker noise. A decline that persists for several days in Search Console, and clusters in specific pages or query groups, deserves a real investigation.
None of this establishes why any site moved. A dip inside the window may have other causes, including seasonality and changes you made yourself.
Wait for post-rollout data to accumulate before drawing conclusions. Teams that see a sustained Search Console decline should audit the affected pages against Google’s spam policies first, and treat recovery as a months-long process.
Reported by Search Engine Journal (Matt G. Southern, October 8, 2026) and Search Engine Roundtable (Barry Schwartz, October 8, 2026).
