Skip to content
SEO Madmanby Adam Hafez

How long Google takes to crawl, index and update results

All 23 typical and slowest times Gary Illyes showed at Search Central Live Barcelona, as recapped by ROAST, plus caveats and Google's own docs.

Published · 5 min read

Written byAdam Hafez
An hourglass with sand running through it, standing on a wooden table

Key takeaways

  • Per the ROAST recap, Gary Illyes showed a typical and a slowest time for 23 processes across crawling, indexing and serving, based on Google's internal analysis.
  • The recap lists a typical site move at 1-3 months and core update recovery at 3-6 months, with slowest cases of 6 months to 1 year or more.
  • Five of the 23 slowest times include the word never, and three of those add quality in brackets, so a fast technical fix cannot rescue a page Google does not value.
  • Search Engine Journal notes that typical is undefined and that sample size and measurement period are not reported, so the figures are ranges, not deadlines.
  • Illyes cautioned that the processes are linked, so a delay in crawling carries into indexing and serving.

Google’s Gary Illyes showed a set of timing tables at Search Central Live Deep Dive Europe in Barcelona on 2 October 2026. They give a typical and a slowest time for 23 processes, from discovering a new URL to recovering from a core update. We have the tables only as ROAST’s John Campbell recapped them: they are not Google documentation, and Search Engine Journal reports the slides are not on Google’s events page. This is a reference copy of those tables, the caveats that come with them, and a comparison with what Google’s own pages say.

According to the recap, each process had a fastest, a typical and a slowest time, based on Google’s internal analysis. The recap prints only the typical and slowest columns. Search Engine Roundtable’s Barry Schwartz reprinted the same rows, and we checked every row against both pages.

How long does crawling take?

Six rows cover discovery, refresh, sitemaps, robots.txt and the two crawl settings.

ProcessTypicalSlowest
Discovery (new URL)~20 hoursWeeks to never
Refresh (known URL)~30 daysWeeks to never
Sitemap processing~24 hoursUp to 14 days, or never (quality)
robots.txt update~24 hours25 hours
Crawl capacity update4 hours to 1-2 weeks1-3 weeks (in recovery)
Crawl demand update~20 hoursWeeks to months

The recap adds one line under the table: crawl capacity can drop in seconds when Google backs off, for example if your server struggles.

How long does indexing take?

Ten rows cover rendering, annotations, indexing itself, removals, canonical changes, site moves, structured data and media.

ProcessTypicalSlowest
RenderingSeconds to render, hours in the queueDays to weeks
Meta annotations45-90 minutes1-4 days
Link annotationsMinutes to 1-3 weeksMonths
Indexing (end to end)~1.5 hoursMonths or never (quality)
Removal1-3 weeksMonths
Canonicalisation change1-3 weeksMonths (conflicting signals)
Site move1-3 months6 months to 1 year+
Structured data updatesHours to 1-2 weeksWeeks or never (quality)
ImagesHours to daysWeeks to months
VideosHours to daysWeeks to months (deep analysis)

The recap defines end to end as all the critical processes finishing successfully, and adds that a small site move can be done in a few weeks.

How long does serving take?

Seven rows cover how changes reach the results page and how long recovery takes.

ProcessTypicalSlowest
Removal in Search Console (owner)~2 hours24 hours
Snippet update1-2 daysSeveral weeks to months
Title update1-2 daysSeveral weeks to months
Text result image update1-2 weeksSeveral weeks to months
Manual action removal1-2 weeks4-6 weeks, or much longer for dormant sites
Core update change3-6 months to recover6 months to 1 year (next core update)
Spam update change1-2 weeks (continuous)Months (batch refreshes)

Below the table, the recap says core updates take 2-4 weeks to roll out and spam updates roll out in 1-2 days. The core update row measures recovery after a change, not the length of the rollout. For what Google said about spam updates at the same event, see our report on the spam update session.

What do the tables not tell us?

Typical is undefined. Search Engine Journal notes that neither the recap nor Google’s event pages define it, and that no sample size or measurement period is reported. We cannot tell whether typical means a median, a mode or a rough impression, or over which sites and which weeks.

The fastest times are missing. The recap says each process had one, and prints only the other two columns, as Search Engine Journal also points out.

Five slowest entries say never. Discovery and refresh say weeks to never. Sitemap processing, end-to-end indexing and structured data updates add quality in brackets. The recap reads this as Google deciding a page is not worth the work, so a fast technical fix does not help.

The processes are linked. Illyes cautioned, per the recap, that a page cannot be indexed until it has been crawled, so delays stack up. A slow stage early on shifts every later stage.

They are ranges, not deadlines. Search Engine Journal says the same of the reported ranges. Schwartz reports that Illyes told him the slides were an exercise to see if the audience can relate to the numbers pulled internally.

How do they match Google’s docs?

Google’s pages state a few of these timings, in looser words. Where they overlap, they do not contradict the tables.

  • Site moves. The site move guide says a small to medium website can take a few weeks for most pages to move, and larger sites take longer. The recap puts a typical move at 1-3 months and notes a small one can finish in a few weeks.
  • Core updates. Google’s core updates page says some changes can take effect in a few days, but it could take several months for its systems to confirm a site is producing helpful, reliable content. If a few months pass with no effect, it says that could mean waiting for the next core update. The recap’s 3-6 months, and 6 months to 1 year for the next core update, fit that language and add numbers the page does not give.
  • Canonical changes. The canonicalization troubleshooting guide says Google might hold pages in a duplicate cluster for up to two weeks after content fixes. The recap’s typical 1-3 weeks sits around that, and its slowest case, months with conflicting signals, goes beyond it.

The three Google pages we read give no timing for a title update or a manual action removal, so those rows have no documented counterpart here.

How should you read them?

This section is our reading, not Google’s guidance, and it rests on tables whose basis is unknown.

  • Use them to spot a stuck change, not to promise a date. A site move still unresolved at three months sits at the high end of typical, as Search Engine Journal reads it. At six months it is in the slowest range. Check the redirect map and the old URLs before blaming the clock.
  • Do not re-edit a title or snippet inside the typical window. One to two days is the typical figure, so a second change after a few hours cannot yet show the first.
  • Fix the crawl layer first. A broken sitemap or a robots.txt mistake delays every stage after it. Validate the file with the XML sitemap validator before you wait for the sitemap row.
  • Check canonical signals when a change drags. The slowest canonical case is conflicting signals, so run your URLs through the canonical chain resolver.
  • Date a recovery from the update, not from your fix. Compare your drop with the Google update checker and expect the 3-6 months the recap lists.
  • Read never as a quality prompt. If a page sits unindexed far past the typical time, the recap’s reading points at the page, not the plumbing. This is a technical SEO and indexing question, but also a content one.

The evidence

Period
Shown 2 Oct 2026
Sample
Not reported (Google internal analysis)

Method: This report runs no new measurement. It reproduces, row by row, the three tables ROAST's John Campbell published from Gary Illyes' session at Search Central Live Deep Dive Europe on 2 October 2026, after reading the live ROAST page and checking every row against the table Search Engine Roundtable reprinted. We did not see the slides: Search Engine Journal reports they are not on Google's events page. ROAST says each process had a fastest, a typical and a slowest time, based on Google's internal analysis, and that the recap carries only the typical and slowest columns. The comparison with Google's documentation uses only the wording of the site move, core update and canonicalization troubleshooting pages. The final section is our own reading and is labelled as such.

Typical time, processes given as one figure (hours)
Indexing, end to end~1.5 hours
Removal in Search Console (owner)~2 hours
Discovery of a new URL~20 hours
Crawl demand update~20 hours
Sitemap processing~24 hours
robots.txt update~24 hours

Sources

  1. 1.Google Search Central Live Deep Dive, Barcelona - Day 3 Recap - ROAST, October 2, 2026Primary
  2. 2.Google Shows How Long Crawling, Indexing & Recovery Can Take - Search Engine Journal, October 4, 2026
  3. 3.Google Search Data On Crawling, Indexing & Serving Timelines - Search Engine Roundtable, October 5, 2026
  4. 4.Site moves with URL changes - Google Search Central
  5. 5.Google Search Core Updates - Google Search Central
  6. 6.Troubleshoot canonicalization issues - Google Search Central

Frequently asked questions

Are these times official Google documentation?

No. They come from slides Gary Illyes showed at an event, as recapped by ROAST, and the slides are not on Google's events page. Google's own pages state only a few of these timings, in looser words such as a few weeks.

What does typical mean in the tables?

Nobody has said. The recap calls the times Google's internal analysis, and Search Engine Journal notes that neither the recap nor Google gives a definition of typical, a sample size or a measurement period.

Why do five slowest times say never?

The recap does not explain each one. Three add quality in brackets, which the recap reads as Google not judging the page worth the work. Treat never as a reason to review the page, not to wait longer.

Where can I read what Illyes said about the numbers himself?

Search Engine Roundtable reports that Illyes told it the slides were an exercise to see if the audience can relate to the numbers pulled internally. That is the only first-hand comment we found.

About the author

Adam Hafez
Adam Hafez

Founder, UpgradIQ FZC LLC

Adam Hafez works on technical SEO and search measurement: how pages get crawled, indexed, ranked and now quoted by answer engines. He founded UpgradIQ, which reads Google Search Console and GA4 to tie ranking movement back to the changes that caused it. He publishes what the data supports and states the limits of it.

  • Technical SEO
  • Search Console and GA4 measurement
  • Answer engine optimization
  • Structured data

The briefing

Only what actually changed in search, delivered in full by RSS, Atom or JSON feed.

Follow
Tools

Hreflang XML sitemap generator

Hreflang sitemap generator: turn language codes and URLs into sitemap XML with xhtml:link alternates, and check codes and URLs. Free, runs in your browser.