I read the first nine results for this question before writing anything.
Every one of them says three to six months. Almost none of them cites a source, and the few numbers that do get attributed contradict each other badly enough that you can find 22%, 5.7% and 95% all credited to the same company.
So the consensus answer to the most-asked question in this industry is a group of people quoting each other.
Here is what the data I could actually trace says, and it is a different shape entirely. Ranking is closer to fast or never than to slow and steady. Which means the question you want answered is not how long it takes, it is how long before you stop waiting, and that has an answer: about a month for a signal, about six for a verdict.
The numbers that break the three-to-six-month answer
Ahrefs re-ran their time-to-rank study in May 2025, and the method is stated, which is why I am using it.
They took a million URLs their crawler first saw in September 2023 and checked whether any reached the top ten within a year. 1.74% did. The other 98.26% did not. In 2017 the same figure was 5.7%, so it got roughly three times harder in eight years.
Filter it more kindly, to two million URLs created in October 2023 with actual English content on them, and you get 6.11%. Ahrefs say themselves that this is probably the fairer number to use, and I agree, because the first cut includes every junk URL a crawler trips over.
Now the number that changes the decision.
Of the pages that did reach the top ten, 40.82% got there within the first month.
Read that carefully, because it is conditional and the conditioning is the whole point. It is not that 41% of pages rank in a month. It is that among the small minority which ever ranked, something like four in ten arrived almost immediately. They also found that for higher-volume keywords, if a page was going to rank at all, month one was when it was most likely to happen.
Put the two together. Most pages never make it, and a large share of the ones that do make it fast.
That is a bimodal outcome, not a gradual one. And three to six months is precisely the window in the middle where, on this evidence, the least new information arrives.
Fast or never, and the five-year-old number one
There is an obvious objection to all of that, and the same study supplies it.
72.9% of pages sitting in Google’s top ten are more than three years old, up from 59% in 2017. The average number one result is five years old, up from two. So the top of the page is clearly dominated by things that have been there a long time, which sounds like ranking takes years.
Both findings are true and they are not in conflict. The resolution is what “old” means.
Those are old URLs, not old drafts. A five-year-old page that holds position one has usually been rewritten repeatedly, has accumulated links and a track record, and is being measured on history your new page does not have. Age is a proxy for incumbency.
Nothing ripens. A page does not slowly earn its way up by sitting there.
What happens over years is that the page gets improved, linked to, and kept current, and the incumbency compounds. If you publish something and leave it alone, the clock is not working for you. It is just passing.
Which also explains why the incumbents look unbeatable and mostly are, on the biggest terms. That is an argument about which terms to go after, not about how patient to be.
So what is three to six months actually describing
I want to be fair here, because the number is not invented and the people repeating it are not lying.
Three to six months is roughly how long it takes to do the work. Audit, fix the technical problems, publish a real body of content, earn a few links. That is a genuine delivery timeline and it is honest when described that way.
The sleight of hand is in letting a delivery timeline sound like a results timeline. “SEO takes three to six months” gets heard as “you will see results in three to six months”, and nobody corrects the hearing, because the correction is uncomfortable and the invoice is monthly.
It also happens to be slightly longer than the point at which most clients would otherwise start asking hard questions. I do not think that is usually deliberate. I do think it is why the number survived.
The checkpoints I would put in a contract
Not a promise of rankings, because anybody promising you a position or a date is selling you something they cannot deliver. These are dates to make a decision on.
Week two to four, is it indexed and is it getting impressions at all? Not ranking, not traffic. Impressions in Search Console for queries you intended to show up for. If a page has been live three weeks and has never been shown to anybody for any relevant query, you do not have a ranking problem, you have a crawling, indexing or relevance problem, and waiting another five months does not touch any of those.
Month one, did anything move on the easy terms? The long-tail, low-competition, specific-question terms. If nothing has moved there, that is early evidence, not proof, that the content is not differentiated enough to beat what is already answering those questions.
Month three, is the trend direction right on a widening set of queries? The count of queries you get any impressions for is the most useful early number in the whole account, and almost nobody reports it. It moves before clicks do.
Month six, decide. Ahrefs’ own advice on their data is that if a page is not in the top ten by about six months, it is time to change it rather than keep waiting, because the odds after that are poor without intervention. I would apply that to the engagement, not just the page: six months in, you should be arguing about which pages to rewrite, not about whether to be patient.
That last one is the one to actually hold people to, including on the pages you published earlier.
What this data cannot tell you
Three honest limits, and the third is the big one.
It is a vendor’s own crawler data. Ahrefs sells SEO software and this study is good advertising for taking SEO seriously. The method is published and the numbers moved against their own interest in places, which is why I trust it more than most, but it is not audited and it is one company’s index.
The 40.82% is conditional, as above. Treat it as a statement about how winners behave, not about your odds.
And top-ten placement is not the thing you actually wanted. In 2026 a page can hold position three and lose half its clicks to an answer printed above it, so a timeline to ranking is now a timeline to a leading indicator rather than to revenue. Any conversation about how long SEO takes that never mentions this is working from a 2019 model of the search results page.
Where I come out
If somebody gives you a number in months, ask what happens at each checkpoint and what they would change if it is missed. The answer tells you more than the number does.
And if your own pages have been sitting at nothing for six months, the patient thing and the correct thing have come apart. Go and rewrite them.
If you’d like to think through whether what you have been waiting on is worth waiting on, drop me a line on email, WhatsApp, phone or LinkedIn and we can have a quick chat, and my notes on SEO cover the work itself. I’m taking on a small number of contracts at the moment, so I’ll tell you straight whether it’s something I could help with.