"How long does SEO take" gets answered with a range — three to six months, six to twelve — that is useless in both directions. It cannot be falsified, so it cannot tell anyone whether their programme is on track or quietly failing.
A better answer describes the sequence, because each stage produces a signal you can check before the one after it arrives.
What happens first?
Crawling and indexing, within days to a few weeks. New pages get discovered and indexed, and that is a binary you can verify rather than wait on. If pages published six weeks ago are still not indexed, something is wrong now, and no amount of patience fixes it.
This is the earliest failure point and the cheapest to detect. Check it directly rather than inferring it from traffic.
Every stage has a signal that arrives before revenue does. A programme that cannot show the current stage's signal is not early — it is stuck, and waiting will not move it.
Then what?
Impressions, usually inside four to eight weeks of pages being indexed. The pages start appearing in results, typically at poor positions. This looks like nothing on a traffic chart and it is the first real evidence of progress.
Watch the number of distinct queries producing impressions rather than the impression count itself. Breadth grows before depth: you appear for more things before you appear higher for any of them.
When do positions move?
Two to five months for competitive terms, faster for long-tail. Positions improve in clusters — a group of related pages moves together as topical coverage accumulates, rather than each page climbing independently.
This is also when the work compounds or fails to. A site building genuine coverage of a subject sees the whole cluster lift. A site publishing scattered pieces across unrelated topics sees individual pages move and no pattern behind them.
And traffic and revenue?
Meaningful traffic from four to nine months in most competitive categories; revenue after that, plus your sales cycle. A ninety-day sales cycle means revenue attributable to month-four traffic lands in month seven, which is worth saying out loud before anyone is surprised by it.
| Stage | Typical timing | Signal to check |
|---|---|---|
| Indexed | Days to 3 weeks | URL inspection in Search Console |
| Appearing in results | 4-8 weeks | Distinct queries with impressions |
| Positions improving | 2-5 months | Average position on filtered queries |
| Traffic | 4-9 months | Clicks from non-branded queries |
| Revenue | Traffic plus sales cycle | Conversions by landing page group |
What makes it slower?
A new domain with no history. A competitive category where the incumbents have years of accumulated links. A site with technical problems consuming the first two months. And publishing volume below what the category expects — coverage compounds, and a page a month compounds slowly.
What makes it faster is narrower than people hope: an established domain, a niche where the existing results are weak, and content targeting questions nobody has answered well.
Judging month three on traffic. Traffic is a lagging indicator of work done in month one. The right question at month three is whether the leading indicators moved, and that has a checkable answer.
How do you tell early that it is not working?
Three checks, each available before traffic would have arrived:
- Are new pages being indexed within a few weeks? If not, the problem is technical and nothing downstream will happen.
- Is the number of queries producing impressions growing month over month? If it is flat at month three, the content is not matching demand.
- Are positions improving in clusters rather than individually? Scattered movement suggests scattered topic selection.
Any of those failing at month three is actionable now. All three passing means the programme is on track even though the traffic chart is still flat, which is the situation this sequence exists to explain.
Check the second one today. It is one filter in Search Console, and it answers more than the traffic line will for another two months.
Three months in and wondering whether it is working?
We will check your leading indicators against where they should be by now and tell you plainly what they show.
Get in Touch →Does this apply to your site?
Reading about it is one thing. Point the scan at your own site and see whether this applies to you, and what it is worth fixing.
Free and unlimited. No account, no card, and you get every finding rather than a teaser.