Google algorithm update checker

Dates are US Pacific time, as on Google's dashboard and in Search Console. Each name links to Google's own page for that update. The note is Google's first post about it. From Google's Search Status Dashboard, read Nov 13, 2026, 9:34 PM PT.

1. 1

   Daily numbers, fair for every weekday

   The checker reads the daily clicks and impressions in your export. Most sites get fewer visits on some days of the week, so each day is divided by its weekday's usual share. A window that starts on a Monday and one that starts on a Saturday then compare fairly.

2. 2

   28 days before vs 28 days after

   For each confirmed update, it compares the average day in the 28 days before the rollout started with the 28 days after it ended. The rollout days in between are left out. Both windows stop before they reach another rollout or the edge of your export, and shrink to the same length. Under 7 days, there is no verdict.

3. 3

   Close rollouts are checked together

   When two rollouts overlap, or have fewer than 7 days between them, neither gets a clean week on each side. So they are checked as one, from before the first starts to after the last ends, and the result names all of them. It's named after the one whose own rollout days saw the biggest move.

4. 4

   Your site's normal swings

   Then it makes the same comparison at every other start date in your export, with the same window length and the same gap in the middle. Start dates whose gap overlaps a confirmed rollout, or the week after this update, are left out. What is left is how much your site moves in ordinary weeks. Each swing is measured from your site's usual change, so steady growth or decline isn't counted as a swing.

5. 5

   A percentile, not a feeling

   The update's change, measured from your usual change, is ranked against those normal swings. "Bigger than 99%" means only 1 in 100 ordinary stretches moved as far. At 99% or more, the result says "outside your normal swing". It needs at least 140 normal swings to rank against, so Search Console's 3-month export is too short.

6. 6

   Clicks first, then impressions

   Clicks are judged when either window has at least 50 of them per 28 days. With fewer, the same test runs on impressions (at least 100 per 28 days), and the result says so. With fewer still, there is no verdict. Impressions aren't judged across a day when a Google reporting change moved them for almost every site (see Sources).

7. 7

   When the change started

   The biggest one-week step inside the windows dates the change. If it came before the rollout started, or more than 7 days after it ended, and is clearly bigger than any step during the rollout, the result says the timing doesn't match.

8. 8

   Big changes with no update nearby

   It also compares every 2 weeks with the 2 weeks before (at the end of your export, as little as 1 week with 1 week). Changes of 25% or more that start at least 7 days from every confirmed rollout are listed. Those point to something else: often a change on your own site.

1. 1

   A match means the timing lines up

   The change came around a confirmed update and was bigger than your usual swings. That makes the update a likely suspect. It isn't proof that the update caused it.

2. 2

   Other things move traffic in the same weeks

   Seasons, news, a site change, lost links or a new competitor can all move clicks. Check the list of big changes with no update nearby, and think about what changed on your side.

3. 3

   Some matches are chance

   "Outside your normal swing" means bigger than 99% of ordinary stretches. Those stretches overlap, so they hold less than their count suggests. On made-up sites with nothing but noise, up to 1 in 20 updates still land there by chance. A 16-month export holds about 7 updates, so up to 3 in 10 sites where nothing happened still get at least one match. The result says so next to a match, and it leads with a match only when the change is 10% or more from your usual change. A core update match, with a drop you can see on the chart, means more than a match on a small spam update.

4. 4

   Within your normal swing isn't always nothing

   Single pages or searches can rise or fall a lot while the site total stays steady. Open your top pages and queries in Search Console and compare the same dates.

5. 5

   Not a penalty check

   A manual action is a person at Google acting on your site, not an algorithm update. Google tells you about those in Search Console's Manual actions report.

• “In general, most sites don't need to worry about core updates and may not even realize one has happened.”

  Google Search's core updates and your website

• “We recommend waiting at least a full week after a core update completes before analyzing your site in Search Console. After a week, try comparing this week with a week before the core update started rolling out.”

  Google Search's core updates and your websiteThe checker waits for the same full week and uses 28-day windows instead of one week, which evens out day-to-day noise.

• “Small drop in position (dropping from position 2 to 4): There's no need to take drastic action (in fact, we recommend avoiding making changes to content that's already performing well). Large drop in position (dropping from position 4 to 29): Do a deeper assessment.”

  Google Search's core updates and your website

• “Avoid doing "quick fix" changes (like removing some page element because you heard it was bad for SEO).”

  Google Search's core updates and your website

• “Some changes can take effect in a few days, but it could take several months for our systems to learn and confirm that the site as a whole is now producing helpful, reliable, people-first content in the long term.”

  Google Search's core updates and your website

• “Sites that see a change after a spam update should review our spam policies to ensure they are complying with those.”

  Google Search spam updates and your site

• “The Performance report tracks daily data according to local time in California.”

  Search Console Help: About Search Console dataThat's why the table and the checker use US Pacific dates, the same days Google's dashboard shows.

• Update list: Google Search Status Dashboard feed (about the last year of incidents, read live and kept for up to 6 hours), plus the dashboard's Ranking history back to November 3, 2021, copied on October 31, 2026. Google's list of ranking updates points to the same history. Only incidents Google files under Ranking are listed.
• Your numbers: the daily table (Dates.csv) in your own Search Console export, read in your browser.
• Days when impressions moved for almost every site: Google ending 100-result pages in mid-September 2025, and the impressions logging error on Google's Data anomalies page (May 13, 2025 to April 27, 2026). Clicks didn't move.
• Google's guidance, checked October 31, 2026: Google Search's core updates and your website, Google Search spam updates and your site, Search Console Help: About Search Console data, Search Console Help: Manual actions report.
• The percentiles are built from your own history, so they describe your site only. They are not a forecast, and a match in timing is not proof of cause.

3 days free, no card.