SEO results do not all arrive at once. A page gets indexed in days, picks up its first impressions in weeks, settles into a stable position in months and brings in leads long after that. The question "how long does SEO take to work" therefore has at least four different answers, and almost every article on the subject gives only one. It is also why SEO goals should be written with a date next to every number.
The short answer, and why it is not enough
Three to six months for the first readable movements, six to twelve for an effect on traffic, more than a year on category queries. Search Engine Land gives the same range: three to six months for measurable results, six to twelve in competitive sectors.
A Morningscore survey of 75 SEO professionals, published in February 2026, found that 67.2% see the first result within two to four months and 82% within six. The sample is international: the questionnaire has no breakdown by market, so read the figure as an industry indication, not as a measurement of the UK SERP.
You will find this range on almost every first-page result for this search. On Google UK the AI Overview opens with "3 to 6 months", citing Search Engine Land and SEO.com, and the organic results from Shopify, Wildcat Digital and Echelon Media say three to six months, around six months and four to twelve months (DataForSEO, Google UK, September 2026). It is correct and of little use, because it does not say time to what.
Five things we all call "results"
When a business owner asks how long it takes, they mean leads. When a consultant answers "three months", they are thinking of rankings. These are two different clocks, and the gap between them is why almost every SEO engagement breaks down in the fourth month. You can shorten it, starting with the changes that pay off in weeks rather than months.
| Stage | What you measure | Where you see it | Time we observe on the projects we manage |
|---|---|---|---|
| Crawling and indexing | URL known and indexed | Search Console, URL Inspection | 1 day to 4 weeks |
| First impressions | the page appears in the SERP | Search Console, Performance report | 1-6 weeks after indexing |
| Stable position | the fluctuation settles | rank tracker | 3-9 months |
| Clicks | organic traffic | Search Console, GA4 | 6-12 months |
| Leads and revenue | leads, quotes, orders | CRM | 9-18 months |
The row that causes the most confusion is the third. An improving position is a real fact, but it is not yet a commercial result: below position ten traffic is statistically close to zero, so a jump from 60 to 25 does not change a line of revenue even though it is half the work. If you want the full map of what to look at, we put it in our guide to SEO KPIs.
The number that changed the answer in the last two years
In 2017 Ahrefs published a study that ended up all over the web: 5.7% of new pages reached the top 10 within a year. In 2025 the same study was redone on two million URLs created in October 2023 and 1.3 million keywords. The result was 1.74%.
- 72.9% of top 10 pages are more than three years old, against 59% in 2017.
- The page in first position is on average five years old, against two in 2017.
- Of the pages that reach the top 10, 40.82% get there within the first month.
The last point is the most useful and the least quoted: the distribution is not linear. Either a page moves early, or it rarely moves later. If by the third month a page does not have a single impression on any relevant query, the problem is not time, and it is the moment to tell apart the three cases that all get called "SEO isn't working".
It is worth saying what the study measures: keywords from the US market and English-language pages. UK queries compete with much of the same English-language content, so the finding travels better than most, but it is still not a UK measurement. The structure of the effect, the incumbency advantage of older pages, is the same one we see in the ranking factors that matter today.
Eight months on an Italian site, measured in Search Console
This is the part none of the first-page results has. We took an Italian dental site we manage, filtered for queries from Italy only and split the period into three windows around the work.
| Period | Days | Impressions | Clicks | Clicks per day | Average position | CTR |
|---|---|---|---|---|---|---|
| 1 May - 12 Sep 2025 | 135 | 17,421 | 177 | 1.31 | 47.5 | 1.02% |
| 13 Sep - 31 Dec 2025 | 110 | 9,406 | 156 | 1.42 | 18.1 | 1.66% |
| 1 Jan - 30 Apr 2026 | 120 | 13,955 | 213 | 1.78 | 16.9 | 1.53% |
Average position goes from 47.5 to 16.9, an improvement of 64%. Clicks per day go from 1.31 to 1.78, up 36%. In the middle window impressions fall by a third while position improves: the long-tail impressions beyond page five, the ones that inflate the report and bring nothing, have disappeared.
These numbers say three things worth keeping separate. First: between the moment position moves and the moment clicks follow, months pass, not weeks, and that delay weighs most when deciding whether SEO is worth it. Second: CTR rises from 1.02% to 1.53% and then stops, because above a certain threshold it depends on the title and snippet more than on position. Third, the one that usually goes unwritten: we cannot attribute the whole movement to a single cause, because in the same period Google released algorithm updates that affected everyone.
For comparison, over the same twelve months an Italian outdoor equipment e-commerce site we manage, with a mature domain and a catalogue already indexed, recorded 583,661 impressions and 13,704 clicks from Italy, with an average position of 6.6 and a CTR of 2.35%. That is the end point of a long journey, and it is the picture business owners have in mind when they ask how long it takes.
On category keywords you count in years
The case we describe in our guide to SEO for manufacturing has public numbers and a stated duration. Macropix, a Milan-based maker of custom LED screens, was in position 88 for "monitor pubblicitario" (advertising display) in 2020 and absent for "ledwall", a query with 4,400 monthly searches in Italy. In 2025 it is in position 2 and position 1 respectively.
The figure that matters in that case is time: five years of continuous work on the whole site structure. In July 2026 the domain holds 25.55% share of voice on the ledwall cluster, ahead of Amazon. Anyone promising that kind of position in six months is selling something else.
Why a new position does not show up straight away
There is a technical explanation that is rarely discussed. Google patent US 8,924,380 B1, granted on 30 December 2014 and titled "Changing a rank of a document by applying a rank transition function", describes a transition function: the system calculates a document's new position and then, for a period, applies an intermediate one that changes over time instead of assigning the final one immediately. In the patent, the aim is to make it hard to tell which change produced which effect, to discourage people who manipulate results.
Be careful how you use this. A filed patent is not confirmation that the system is live today, and Google has never said publicly that it uses it. It remains the most consistent explanation for something anyone working with a rank tracker sees every week: positions bounce around for weeks before they settle. If you need to know how to read those fluctuations, we explained it when we covered how a SERP is put together.
The Google quote everyone cites that is not in the documentation
Practically every article on this subject quotes Google saying "four months to a year". Google UK's own AI Overview for the query repeats it (DataForSEO, Google UK, September 2026). The line is Maile Ohye's and comes from a 2017 Google Search Central video on how to hire an SEO consultant.
In August 2026 we went looking for it on the Search Central page that replaced that content. It is no longer there. The current page gives no time estimate: it suggests asking the consultant what results they expect and over what timeframe, and stops there. Nine years on, the industry's most quoted source on SEO timescales is a sentence Google stopped writing.
What shortens the timeline and what does not
Working on a domain that already has a history shortens it, because it starts with indexed pages and a higher crawl frequency. Fixing indexing before writing shortens it: if half the site is not indexed, every month of new content is another month of waiting. Starting with queries that already get impressions between positions 8 and 20, where a small push gets you over the threshold, shortens it too, and it is the first thing we look at in an SEO audit.
Publishing more does not shorten it. In the Morningscore survey 52.1% of professionals name link building as the most time-consuming activity, followed by content writing at 21.1%: they are the two slowest levers, and that is also why agencies cut them first when the budget is low. Changing titles every week does not shorten it either, because every change puts the page back under evaluation.
Then there is a variable that international averages hide, and it is central to the reasoning on marketing for small businesses: how many real competitors are on the first page. In many B2B niches, where smaller firms still have thin websites or none at all, the direct competitors on page one number three or four, not thirty, and the timeline shortens considerably. It is exactly the situation we describe in our guide to B2B SEO.
How to tell in month three whether it is working
Month three is when almost everyone loses patience, and it is also when the data starts to say something. What to look at, in order.
- Number of indexed URLs out of all the pages you want to rank. If it is not rising, nothing else counts.
- Number of distinct queries for which the site gets at least one impression. This is the indicator that moves first, before rankings and well before clicks.
- How many queries have moved from the 21-50 band to the 11-20 band. This movement brings no traffic. It does tell you the direction is right.
- CTR on the pages that were already on page one. If it rises, the work on titles and on-page SEO is working regardless of rankings.
All four can be read in Google Search Console, free, without waiting for the agency report. If none of these four numbers has moved after three months, the conversation you need to have is not about time. We wrote about it more bluntly in how to tell if your SEO agency is working.
Frequently asked questions
On brand queries and very specific queries, yes: a new page can be indexed in days and appear straight away. On commercial queries, no. In the 2025 Ahrefs study only 1.74% of published pages reach the top 10 within a year, and of the pages that get there 40.82% do so within the first month: either a page moves straight away, or it needs far more than thirty days.
Usually yes, for two reasons. First, the Local Pack depends heavily on the Google Business Profile listing, reviews and proximity, signals that move in weeks. Second, there are few competitors on a local query. On the projects we manage, the first movements in the Local Pack typically come within two or three months.
Because below position ten traffic is close to zero. On an Italian site we manage, average position went from 47.5 to 16.9 in eight months, a 64% improvement, while daily clicks grew by 36%, from 1.31 to 1.78. That is normal: position moves first, clicks come later and in smaller proportion.
Six months on results, three months on method. By month three you cannot demand rankings, but you can expect growth in indexed URLs, in the number of queries with at least one impression and in the share of queries that moved from the 21-50 band to 11-20. If none of these three numbers moves, the problem is not time.
They have changed what you measure more than how long you wait. A page can be cited in a generated answer before it gets meaningful clicks, so average position and clicks in Search Console tell only part of what is happening. Anyone assessing an SEO project in 2026 should add monitoring of citations in AI answers to the classic KPIs.
One thing the industry rarely says: in SEO the clock resets every time you change direction. A site that in two years has changed its URL structure three times, its content strategy twice and its agency once carries six months of work repeated four times, and counts it as two years. Looking at the SEO projects we have run through several cycles, the variable that best explains results is not the monthly budget: it is how many times nobody changed their mind about the SEO strategy.
