What Google itself says about the timeline
Google Search Central's guidance on hiring an SEO puts the figure at four months to a year for the work to be implemented and for the potential benefit to appear. It is written by the company that owns the index, and it is the closest thing to an authoritative answer that exists.
Two things about that range are worth noticing. It is wide — three times wider at the top than the bottom — and it includes the time to do the work, not just the time to wait afterwards. A large part of any honest SEO timeline is production: pages written, structure fixed, information corrected. None of that is waiting.
The rest of this piece is about where in that range your business is likely to land, because that is the part nobody publishes.
Three clocks are running, and they do not tick together
Google's own documentation separates crawling, indexing and serving into distinct stages. That separation is the reason a change can be live on your site, present in the index, and still not showing up for anyone.
| Stage | What has to happen | Typical feel |
|---|---|---|
| Discovery and crawling | Google has to find the page and fetch it | Days to weeks on an established site; longer on a new one |
| Indexing | The page has to be judged worth storing and understood | Weeks; not guaranteed at all for thin pages |
| Ranking | The page has to beat what is already there | Months, and it is a competition rather than a queue |
The third stage is the one that breaks expectations. Crawling and indexing are broadly mechanical. Ranking is relative — you are not clearing a bar, you are displacing somebody who has been there longer. That is why identical work produces movement in six weeks in a quiet category and nothing in six months in a crowded one.
A new site is slower than a broken one you fix
The single biggest variable is what you are starting from, and it splits into two very different jobs.
Repairing a site that already has history is a different job, on a different clock, from building authority a domain has never had.
Repair work is fast because the demand and the history already exist and something is in the way. Pages that were never indexed, a setting that told crawlers to stay out, duplicate versions of the same page competing with each other, a redesign that quietly dropped half the content — these produce movement in weeks once removed, because you are not asking Google to reconsider you, you are letting it see you.
Building is slow. A new domain has no history, few pages and nothing pointing at it. Everything has to be produced and then earn its position against incumbents. Plan for the upper half of the range and put a channel with a faster feedback loop alongside it — paid search exists precisely to cover the gap while the organic programme matures.
Local visibility runs on a third clock again. Google's local ranking guidance weighs relevance, distance and prominence rather than domain authority, so a business profile with accurate details and a steady flow of reviews can register faster than the same business competing on national terms. If your customers are within driving distance, the local layer is usually the first thing that moves.
Repairing a site that already has history is a different job, on a different clock, from building authority a domain has never had.
What should be visibly different in the first ninety days?
Rankings are the last thing to move, which makes them a terrible early progress measure. These are the things that should have changed while you are waiting, and every one of them is checkable by the person paying for the work.
- More of your pages are indexed than were at the start, and you can see the count
- Impressions in Search Console are rising even where clicks have not yet
- You are appearing for a wider set of terms than before, including ones nobody targeted
- Pages that were broken, missing or duplicated have been fixed, and there is a list of which
- Average position for terms you care about has moved, even from twenty-eight to nineteen
If nothing is measurably different in the first ninety days, the problem is not patience. It is that nothing was actually done.
That last point is the most useful thing in this piece. The four-to-twelve-month range is real, and it is also the most convenient excuse in the industry. The distinction between slow progress and no progress is entirely visible in Search Console, which is free and which you should own rather than your agency.
How to tell slow progress from no progress
Ask for three things at the ninety-day mark. What was changed, in a list, with dates. What has moved in Search Console, in impressions and average position rather than a screenshot of one good keyword. And what is expected to move next, so the following period has a prediction attached to it.
An agency doing real work can answer all three quickly, because those are the artefacts of the work itself. An agency that cannot answer them is not being slow. The distinction between the two is set out at length in the briefing on reporting you cannot follow.
Be careful with rank-tracking reports as the primary evidence. Results vary by location, by device and by the person searching, so a tracked position is an estimate of an average. Search Console is first-party and reports what actually happened.
If nothing is measurably different in the first ninety days, the problem is not patience. It is that nothing was actually done.
When SEO is the wrong thing to buy first
SEO compounds, which is exactly why it is the wrong first purchase for a business that needs work inside the quarter. If payroll depends on enquiries arriving in the next eight weeks, a channel with a four-to-twelve-month horizon is a poor place for the last of the cash.
It is also the wrong first purchase when the site cannot convert the visitors it already has. Doubling the traffic to a page that turns nobody into an enquiry doubles nothing. Fix the page that receives the visit first; it improves every channel at once and it is measurable in days rather than quarters.
And it is the wrong purchase when what you actually need is to be found in the map results by someone half a suburb away with an urgent problem. That is a different discipline with a faster clock, and the citation and profile side of it is covered in the piece on local listings.
None of that is an argument against SEO. It is an argument for sequencing it honestly, so the business can still be trading when the compounding starts.
Common questions
Can SEO work faster than four months?
Sometimes, and almost always when the site is being repaired rather than built. Fixing pages Google could not index, or a site accidentally blocking crawlers, can produce movement in weeks because the demand was already there and the door was shut.
Does a brand-new domain take longer?
Generally yes. A new site has no history, no established pages and nothing pointing at it, so every signal has to be built from scratch. Expect the upper half of the range and plan a paid channel alongside it if you need work before then.
Is local SEO faster than national SEO?
Usually. Local results weigh relevance, distance and prominence, and a profile with accurate details and steady reviews can register faster than a domain competing nationally on authority. It is a different competition with a different clock.
How do I know it is working before rankings move?
Watch impressions and average position in Search Console, the number of pages actually indexed, and whether you are appearing for a wider set of terms than before. Those move first. Clicks and enquiries move last.
Do I have to keep paying once I rank?
Not at the same level, but positions are held rather than owned. Competitors keep publishing, your content ages, and sites change. Most businesses move from a build phase to a lighter maintenance phase rather than stopping outright.