The October 2023 Core Update: When Three Google Events Hit at Once

Three updates, one window, one wrong assumption

October 2023 is the canonical case study in why "I got hit by the core update" is so often the wrong place to start.

Most core update retrospectives open the same way: a core update is a broad reweighting of Google's ranking systems, sites move, here is who won and lost. That framing is fine for an update that rolls out cleanly on its own. The October 2023 core update did not. It landed in the middle of one of the most crowded volatility windows Google has ever shipped, and that crowding, not the reweighting itself, is the story worth remembering.

Three distinct Google events overlapped in a single two-week span. The core update ran October 5 to 19, 2023. A concurrent spam update ran October 4 to 20, covering almost the identical window. And the core update itself introduced a bug in Google Discover that was not fixed until October 31 and not disclosed until November 1. A single traffic drop in that period could have had three unrelated causes, each with a completely different fix: improve content quality, clean up spam signals, or do nothing at all and wait for a bug fix to reverse on its own.

That is the thesis of this post. Before you decide how to recover, you have to diagnose which signal actually moved. October 2023 is the update that proved why.

The timeline and the verbatim announcements

Google announced the update on October 5, 2023, on the Search Status Dashboard and echoed it through its Search Central and SearchLiaison channels on X.

Released the October 2023 core update. The rollout may take up to 2 weeks to complete.Google - Search Status Dashboard

The rollout was marked resolved on October 19, logged at 13 days and 23 hours. On a normal update that would close the file. Here it did not, because the same incident record later carried a second, more revealing note about Discover.

As of October 31, we fixed the bug. Some sites may see an increase in Discover-related traffic, as a result of this bug fix.Google - Search Status Dashboard

Google attributed that issue to "a bug with how our October 2023 core update was applied to Discover." Read the two quotes together and the practical problem becomes obvious. Google told you the core rollout finished on October 19, but a side effect of that same rollout kept moving Discover traffic for almost two more weeks. A site owner watching their dashboard on October 25 had no clean way to know whether a swing was the core update settling, the overlapping spam update, or the unfixed Discover bug.

This was the third core update of 2023, after the March 2023 core update and the August 2023 core update, and it arrived only weeks after the September 2023 Helpful Content Update. For many publishers it was the third or fourth consecutive hit in a four-month span.

Who moved, and why the labels are slippery

SISTRIX's UK and US visibility analysis named specific movers, and the pattern is instructive. The absolute-visibility winners were established mega-domains. The clearest losers clustered in two verticals: dictionaries and song lyrics.

Here is the catch, and it is the whole point of this update. SISTRIX itself noted that the spam update likely explains why losers outnumbered prominent winners, and that the "winners" were largely big sites backfilling the gaps left by demoted or removed pages. In other words, the visible winners may have gained from the spam update's removals as much as from the core reweighting. Because the core and spam updates overlapped almost completely, no analyst could cleanly attribute any single site's movement to the core update alone. Treat every named domain above as core-plus-spam combined, not pure core.

Diagnose the signal before you optimize

The transferable value of October 2023 is not a recovery playbook. It is a diagnostic discipline. When several Google events share a window, your first job is not to fix anything. It is to determine which signal actually moved, because each possible cause points to a different action.

1. Separate Search from Discover. Discover traffic and organic Search traffic are different reports for a reason. In October 2023 a Discover swing could have been the unfixed bug, which needed no action at all and reversed itself on October 31. Confusing it with a Search drop would have sent you optimizing content that was never the problem.
2. Separate content from spam and links. A core update reweights how quality is assessed; a spam update demotes sites tripping spam signals. The fixes are unrelated. Overlapping windows mean you check both report surfaces before concluding either.
3. Check the dates against your drop. The core update finished October 19, the spam update October 20, the Discover bug fix landed October 31. Aligning the exact day your metric moved against those dates is the cheapest, fastest piece of attribution evidence you have.
4. Resist the single-cause story. The instinct is to name one villain. October 2023 punished that instinct. A single drop genuinely could have had three causes, and "recovering" from the wrong one wastes weeks.

More concurrent variables, not fewer

This is a 2.5-year-old update with no live recovery intent. The Helpful Content system has since been folded into core, and 2024 and 2025 brought multiple larger core updates. There is no "recover from October 2023" project to run in 2026. So why keep the lesson?

Because the attribution problem has gotten worse, not better. October 2023 had three concurrent variables. A volatility window in 2026 can have more, because AI Overviews and AI Mode now sit on top of every classic update as additional moving parts. A traffic drop today might be a core reweighting, a spam update, a tracked Search Console bug, or a shift in how AI Overviews source and display answers above the blue links. The diagnostic question is identical, just with more surfaces to check.

• Surfaces multiplied: Search, Discover, AI Overviews, and AI Mode are now distinct exposure channels that can move independently in the same window.
• Overlap is normal: Google frequently runs core and spam updates in adjacent or overlapping windows, exactly as it did October 4 to 20, 2023.
• Bugs are real events: the Discover bug proved that a documented Google fault, not your content, can move your numbers for weeks past a "completed" date.

If you want the modern version of this discipline as a repeatable process, the step-by-step core update diagnosis framework formalizes the same separate-the-signals logic October 2023 forced on everyone.

The takeaway we hand to teams

When a client comes to us mid-volatility convinced the core update tanked them, we slow the conversation down before anyone touches the site. The October 2023 window is the example we point to, because it is the cleanest proof that the obvious cause is not always the real one.

• Diagnose first: confirm which report moved (Search vs Discover) and on which exact date before proposing any change.
• Map the calendar: line your drop up against every Google event in the window, not just the headline core update.
• Rule out non-content causes: a spam update, a links issue, or a documented Google bug each demands a different response than a content rewrite.
• Only then act: if and only if the evidence points to content quality do you treat it as a core update content problem.

October 2023 did not teach the industry a new ranking factor. It taught a posture. The volatile window is the moment to be a careful diagnostician, not a fast fixer, and that posture has only become more valuable as the number of concurrent variables keeps climbing.