How Long SEO Takes, And What Moves The Date

Every enquiry we get about search eventually arrives at the same question, usually in the third or fourth email: how long before I see results? It is a fair question. You are spending money every month and you would like to know when it stops being an expense and starts being an asset.

Anyone who answers with a single number is either guessing or selling. What can be described honestly is the shape of the curve, what changes at each stage, and the specific conditions that pull the date forward or push it back. That is what this piece does.

Why a single number is always wrong

SEO — search engine optimisation, the work of making a site the sort of thing search engines can find, understand and prefer — is not a service with a fixed production time. It is a competition. The finish line moves depending on who else is running.

Consider two businesses in East Delhi. One is a chartered accountancy practice with a three-year-old website that already ranks for its own name and a handful of long queries. The other is a new interior design studio with a site built last month on a fresh domain, going after "interior designers in Delhi" — a phrase where the first page is held by directories, aggregators and firms who have been publishing for a decade.

Same monthly budget. Same quality of work. The accountant may see meaningful movement inside a quarter. The design studio may spend a year building the foundation before the head terms shift at all, while picking up useful traffic on narrower phrases along the way. Quoting either one "three to six months" would be dishonest to the other.

What compounding actually means here

SEO gets described as compounding, which sounds like marketing language but describes something mechanical.

A page that ranks earns clicks. Some of those clicks come from people who link to it, mention it, or send it to someone else. Those signals help the page and, to some degree, the rest of the site. Meanwhile the page keeps earning while you write the next one. Each new page starts from a slightly stronger base than the last one did.

So the curve is not a straight line from zero. It is flat for longer than anyone finds comfortable, then bends. The flat part is not nothing happening — it is pages being crawled, indexed, evaluated and slowly re-evaluated. It just does not look like anything in a traffic graph.

This is also why stopping halfway is the most expensive decision available. Six months of work abandoned in month seven captures almost none of the bend. The asset was being built; it had not started paying.

Stage by stage: what actually changes

The first four to six weeks

Nothing you would call a result. What should be happening is diagnosis and repair. Which pages return a 404 (page not found) or a 500 (server error)? Is the site returning a 200 (fine) on the pages you care about? Is anything blocked in robots.txt by accident — a genuinely common problem when a site goes from staging to live and nobody removes the block?

Page speed work belongs here too, because it is one of the few areas with published thresholds. Google's Core Web Vitals define a good Largest Contentful Paint as 2.5 seconds or under, Interaction to Next Paint as 200 milliseconds or under, and Cumulative Layout Shift as 0.1 or under. Those are checkable numbers, not opinions. A photo taken straight off a phone is often four or five megabytes; a page carrying six of them uncompressed will fail the first threshold on an Indian mobile connection without any argument.

If the site itself is the problem — a builder platform you cannot edit properly, a theme that cannot be made fast — this stage may honestly conclude that the site needs rebuilding before optimisation is worth paying for. That is an uncomfortable finding but a cheaper one than a year of work on a broken base.

Roughly months two to four

The first observable movement usually appears in impressions rather than clicks. Google Search Console — the free tool that shows what your site appears for — starts logging your pages against queries you were previously invisible for. You are on page four instead of nowhere.

Owners often find this stage frustrating because impressions do not pay salaries. But it is the honest first signal, and its absence is meaningful. If you are three months in and impressions are flat across the board, something is wrong and you should be asking hard questions rather than waiting.

Local search often moves faster than the rest. A Google Business Profile that is complete, correctly categorised, has real photographs and is collecting genuine reviews can start showing in map results within weeks, because the ranking factors there lean heavily on proximity, category and profile completeness rather than on a decade of accumulated links.

Roughly months four to nine

This is where long-tail queries start converting. Not "CRM software" but "CRM for a distribution business in Delhi" — phrases with low volume, low competition and buyers on the other end of them. Enquiries begin arriving that you can trace to search.

This is also the stage where content production rate becomes the binding constraint. If you are publishing one substantial, genuinely useful page a month, you will move slower than a competitor publishing four. If you are publishing four thin pages a month that say nothing, you will move slower than the person publishing one good one.

Nine to eighteen months and beyond

Competitive head terms become realistic. Pages published early are now at their strongest. The site starts ranking for phrases you never explicitly targeted, because Google has enough signal about what the site is broadly about. Traffic growth begins to look less like a project and more like a property.

The conditions that move the date

Condition	Pulls the date forward	Pushes it back
Domain history	Site live for years, already indexed, no penalties	Brand new domain, or an expired one with spam history
Competition	Niche B2B, local service, specific product category	Insurance, loans, real estate, generic city-wide terms
Technical state	Fast, crawlable, sensible URL structure	Slow builder site, JavaScript-rendered content, duplicate pages
Content capacity	Someone internally who can answer questions properly	No subject-matter input; agency writing blind
Decision speed	Approvals in days	Recommendations sitting unapproved for six weeks
Geography	One city, one service area	Pan-India, multiple languages, many locations

The one nobody expects on that list is decision speed. A great deal of lost time in SEO is not Google's — it is a recommendation made in March that goes live in June because it needed sign-off from three people.

What does not speed it up

Some of these are things an agency could happily sell you and we would rather you did not buy them.

• Guaranteed rankings. Nobody controls Google's results. A guarantee is either meaningless small print or a promise to rank you for phrases nobody searches for.
• Bulk link packages. "500 backlinks for a fixed fee" buys you links from sites that exist only to sell links. Google has publicly documented link spam policies. Best case, the links are ignored and you wasted the money. Worst case, cleanup takes longer than the original build.
• Volume content with nothing in it. Thirty 500-word posts a month, generated and unedited, on topics your buyers do not care about. It fills a report. It does not move a curve.
• Paying to raise a third-party "authority score". Domain Authority and similar metrics are estimates invented by SEO tool companies. Google does not use them. Optimising for the estimate rather than the thing is a category error.
• Restarting the strategy every quarter. Changing direction in month four resets the clock on the only thing that was working: accumulation.

How to tell it is working before rankings move

You do not have to sit blind for six months. There are leading indicators, and you should be shown them without asking.

1. Indexed page count. Are the pages you published actually in Google's index? Search Console's Pages report tells you, and tells you why not.
2. Impressions by query. Not total impressions — the list of new queries you now appear for. Growth here precedes click growth by weeks or months.
3. Average position on target terms. Moving from position 48 to 22 produces no traffic and is genuine progress.
4. Core Web Vitals status. Pass or fail against the published thresholds. Binary and checkable.
5. Enquiry source. Ask every caller how they found you. Low-tech, and it catches results that analytics misses.

Where AI search changes the picture

Honestly: we do not know yet, and anyone claiming precision here is ahead of the evidence.

Search results increasingly include AI-generated summaries, and buyers increasingly ask questions of chat assistants instead of typing into a search box. What is observable is that these systems tend to draw on content that is clear, well-structured, and consistent about facts across the web — your name, address, services and claims saying the same thing everywhere. What is not established is a reliable timeline for influencing them, or a stable set of levers. Anyone selling a fixed programme with promised outcomes for visibility inside AI answers is selling certainty that does not exist. The sensible position is to do the work that helps either way and to watch closely.

What to do next

Before committing to a monthly retainer with anyone, get three things established: the current technical state of your site, an honest read on how competitive your target phrases actually are, and agreement on which leading indicators you will be shown every month. If a proposal skips straight to a price and a promise, that is the thing to question.

If you want that read on your own site, tell us what you sell and where your buyers are and we will tell you which of the conditions above apply to you — including if the answer is that your money is better spent on something other than search work this year. There is more on how the individual pieces fit together across the rest of our writing.

Questions we get asked