How Long Does GEO Take? What the Sources Say About Timing
There is no fixed timeline, and anyone who promises one is guessing: we found no independent study that measures a typical time to results. What can be sourced is how long each step takes. OpenAI and Perplexity say crawler access changes take up to about 24 hours to apply. Google says crawling can take a few days to a few weeks and some changes take months. A new page can be used by search-enabled answers once it is indexed, but changes to a model's own memory wait for a later release.
Key takeaways
- OpenAI says its systems can take about 24 hours to adjust after a robots.txt change, and Perplexity says up to 24 hours. That is the lag for access rules, not for being cited.
- Google says crawling can take "anywhere from a few days to a few weeks" and advises waiting a few weeks to assess the effect of a change.
- Models learn slowly: OpenAI's GPT-6 Astra has a knowledge cutoff of April 30, 2026 and was released September 3, 2026, about four months later.
- Only 1.74% of newly published pages reached Google's top 10 within a year in Ahrefs' 2025 study, a reminder that most new pages never reach the top 10 at all.
Is there a standard timeline for GEO?
No. Claims like "90 days" or "3 to 6 months" circulate widely, but we found no primary study or official source that measures them. Citepoint's own FAQ says it plainly: there's no fixed timeline, and anyone who promises one is guessing.
What you can do is break the work into steps and use the sourced timing for each.
How long does each step take?
| Step | What happens | What the sources say |
|---|---|---|
| Access | An AI crawler reads your updated robots.txt | OpenAI: it "can take ~24 hours from a site's robots.txt update for our systems to adjust." Perplexity: "it may take up to 24 hours for our systems to reflect changes." Anthropic's page states no figure |
| Discovery | Search engines find a new or changed page | Google: "Crawling can take anywhere from a few days to a few weeks." IndexNow says that without it, discovery "can take days to weeks" |
| Indexing | The page enters the index that answers draw on | Google says indexing and serving "isn't guaranteed" and gives no time |
| Retrieval | The engine finds your page for a question | Claude's documentation says web search lets it answer "with up-to-date information beyond its knowledge cutoff" |
| Selection | Your page and mentions compete with others | No sourced timeline |
| Training | A new model learns your facts | Months after the data cutoff, and only with a new release |
How long does Google take to notice a change?
Google's SEO Starter Guide says: "Some changes might take effect in a few hours, others could take several months. In general, you likely want to wait a few weeks to assess whether your work had beneficial effects." Its recrawl documentation adds that requesting a crawl "does not guarantee that inclusion in search results will happen instantly or even at all," and that requesting a recrawl multiple times for the same URL won't speed it up. This matters for GEO because Google says a page must be indexed and eligible to show with a snippet to be a supporting link in AI Overviews or AI Mode.
IndexNow can tell participating engines about changes. Its current list includes Bing, Yandex, Seznam, Naver, Yep, the Internet Archive and Amazonbot. Google is not on it, and IndexNow says each engine decides for itself whether to index. We found no source linking IndexNow to ChatGPT or Perplexity answers, so don't assume it speeds those up.
Why does training lag matter?
What a model remembers about your brand comes from data that is already months old when the model ships. OpenAI's GPT-6 Astra lists a knowledge cutoff of April 30, 2026, and its changelog shows a release on September 3, 2026, about four months later. Anthropic lists a reliable knowledge cutoff of February 2025 for Claude Haiku 4.5, which was released on October 15, 2025. A fix you publish today reaches a model's memory only with a later release.
That is why the retrieval path matters. Search-enabled answers can use pages published after the cutoff, so a new page can start counting once it is indexed and an engine chooses to cite it. Our homepage FAQ says the same, and adds that some questions move sooner than others.
What does the freshness research say?
Ahrefs analyzed 16.975 million cited URLs and found the average URL cited by AI assistants was 1,064 days old, against 1,432 days for organic Google results, which it calls 25.7% fresher. ChatGPT showed the strongest preference for newer content, and Google's AI Overviews cited content 16 days older than organic results. The authors add that "the average age of cited ages is still 2.9 years," so AI assistants still cite long-lived content, and that low-quality content updated every day "will not have a magic positive effect." Seer Interactive found, in a June 2025 analysis, that about 71% of ChatGPT's citations were to content from 2023 to 2025.
For you, this means new pages don't need to age before being cited, but established pages are real competition, and refreshing pages that already have authority is a legitimate lever.
What is the reality check from traditional SEO?
In an Ahrefs study of 1 million random URLs first seen in September 2023, only 1.74% ranked in Google's top 10 within a year. A second sample of 2 million non-empty English pages created in October 2023 gave 6.11%. Of the pages that did reach the top 10, 40.82% did so within a month. Ahrefs suggests that after about six months, if a page isn't in the top 10, you should look at updating it.
This is a Google ranking benchmark, not a GEO one. It still shows that most new pages don't reach the top 10, and Ahrefs found it takes longer to rank for high-volume keywords than for low-volume ones.
What do vendors claim?
Vendors publish faster numbers. Profound's customer stories claim that Ramp's visibility rose from 3.2% to 22.2% "in just 1 month," that Airbyte's ChatGPT visibility "tripled in one week," and that OpusClip exceeded its goal "within the first 30 days." These are single-brand stories picked by the vendor and measured with its own visibility metric, with no control group. The chart on Ramp's page, for example, covers December 1, 2024 to February 15, 2025. Treat them as anecdotes. On the agency side, Siege Media's page says owned content and freshness wins typically surface within the first quarter and citation share builds over 3 to 6 months, which is again one agency's claim.
How should you measure progress?
Because AI answers vary between runs (see why AI answers change), judge progress over weeks, not days.
- Check the leading indicators first: is the page indexed, and are AI crawlers fetching it in your server logs?
- Track the questions you targeted, per engine, and compare rolling periods such as the last 28 days.
- Watch citations and visibility separately. They can move independently.
- Start where you're close. Questions where you already rank or get named are likely to move first.
- Keep the question set fixed, so changes reflect your work.
What can you do to speed it up?
Allow the search crawlers, list new pages in your sitemap, use IndexNow for engines that support it, and request indexing in Search Console for key pages, noting the quota. Then work on what no crawler setting can fix: third-party mentions and pages that answer the question better than the ones engines cite today. For setup, see robots.txt for AI crawlers.
Frequently asked questions
Can GEO work in 30 days?
For some questions, possibly. A new page can count once it is indexed and an engine cites it. But vendor claims of one-week or 30-day gains are single-brand anecdotes, and we found no independent study of typical timing.
Why did a page rank on Google but never show up in AI answers?
Ranking is one input. Engines retrieve several pages, often through several related searches, and write the answer from what they find. A page that ranks but buries the answer, or a brand no one else mentions, can still be left out.
Do I need a new model release for AI to know about my brand?
Not for search-enabled answers, which can use pages published after a model's cutoff. For what the model remembers on its own, yes, a new release is needed, and releases arrive months after their data cutoff.
How long after I change robots.txt will AI crawlers respond?
OpenAI says about 24 hours for its search systems to adjust, and Perplexity says up to 24 hours. Anthropic's crawler page gives no figure. That is the delay for honoring access rules, not for appearing in answers.
Does IndexNow help me get into ChatGPT?
We found no source saying so. IndexNow notifies participating engines such as Bing, and Google isn't listed. Use it for faster discovery where it is supported, but don't assume it affects ChatGPT answers.
Sources
- Google Search Central: Ask Google to recrawl your URLs
- Google Search Central: SEO Starter Guide
- Google Search Central: AI features and your website
- OpenAI: Overview of OpenAI crawlers
- Perplexity: Perplexity crawlers (PerplexityBot and Perplexity-User)
- Claude Help Center: Does Anthropic crawl data from the web, and how can site owners block the crawler?
- IndexNow: Home
- IndexNow: FAQ
- IndexNow: Participating search engines
- OpenAI API docs: Models
- OpenAI API docs: Changelog
- Claude Platform docs: Models overview
- Anthropic: Introducing Claude Haiku 4.5 (October 15, 2025)
- Claude Platform docs: Web search tool
- Ahrefs: New study, AI assistants prefer to cite fresher content (17 million citations analyzed)
- Seer Interactive: Study, AI brand visibility and content recency
- Ahrefs: How long does it take to rank in Google, and how old are top ranking pages?
- Profound customer story: Ramp
- Profound customer story: Airbyte
- Profound customer story: OpusClip
- Siege Media: Generative Engine Optimization (GEO) services