How long does SEO take, and what you should see each month

Why there is a delay at all

The delay is not someone taking their time. Three separate clocks run before a new page earns anything, and they run in sequence.

First, the page has to be found and indexed. Google has to crawl it, decide it is worth keeping, and file it. For a small site that is usually days to a few weeks, and it can be longer if the site is new or the pages look like each other.

Second, it has to be tested. Google puts a new page somewhere in the results and watches what happens, and that position is usually far down at first. It moves on evidence, and the evidence takes weeks to accumulate.

Third, it has to earn trust it does not start with. A page on a site with history and links ranks sooner than the same page on a site nobody has ever linked to.

That third clock is why two businesses doing identical work see results months apart, and it is also the part nobody can shortcut for you.

The order things actually move in

This is the part worth memorizing, because every one of these happens before the phone rings, and each is visible if you know where to look. A plan that is working shows them in this order. A plan that is not shows none of them.

What to check in month one

Month one produces no results and it does produce evidence. The difference between a good first month and a bad one is entirely visible if you look at the right thing.

Check that the new pages exist and are indexed. Check that the profile changed: categories, services listed, hours, photos. Check that the technical fixes shipped, which usually means the site is faster and the broken links are gone.

What you should not check is rankings. Nothing has had time to move, and looking will only teach you to distrust a process that is working normally. The six screens in what to check in Google Analytics are the right instruments for this month, and most of them are baselines rather than readings.

• The month one question. Can I click on what was made? If the answer is a strategy document and a list of keywords, ask what gets published in month two.

Months two and three, where impressions come before clicks

This is the stretch that decides whether people stay with a provider, and it is also the stretch with the clearest leading indicator that almost nobody watches.

Open the queries report in Search Console and sort by impressions rather than clicks. Impressions are the times your site appeared in front of someone. They rise well before clicks do, because a page at position twenty-five appears often and gets clicked almost never.

A rising impression count on searches you want, with clicks still near zero, is exactly what a working plan looks like at this stage. It means Google is now showing you for those searches and has not yet put you high enough to be chosen. That is a position problem, and position is what the next two months of work moves.

The count of distinct queries matters as much as the volume. Appearing for sixty searches this month against twenty last month is the plan working, even if the traffic chart is flat.

• The months two and three question. Am I appearing for more searches than I was, and are they searches a customer would make? Yes to both is progress, whatever the traffic number says.

Months four to six, when it starts to pay

Positions climb, and the climb is not smooth. A page sits at fourteen for six weeks and then moves to six two weeks later, because ranking is a comparison and it changes when the comparison changes.

By month six a plan that is working looks specific. More pages than the home page earn entries. The query list is full of searches you would pay for. Some of those sit on page one. Map visibility holds across your service area rather than only at your own address. And the phone produces calls you can trace to a page rather than to your name.

This is also when the work changes shape, from building pages that did not exist to deepening the ones that nearly rank. The queries report starts naming what to write next, which is the loop described in what ongoing SEO actually means.

What makes it faster

Five things genuinely shorten the timeline, and knowing which you have tells you where in the three to six months you are likely to land.

• A site with history. An established domain with existing pages and links starts ahead of a site that launched last month. The same work on a new domain takes longer, and there is no way around it.
• Something obvious being broken. A noindex tag left on after a rebuild, a site nobody could crawl, a profile with the wrong category. Fixing a blocker produces movement in weeks, because the demand was already there.
• A neglected profile. If the map half has never been worked, it is the fastest available gain, and it is mostly free.
• Weak competition. Thin pages and stale profiles above you means the bar is low. Two thousand reviews and twenty years above you means it is not.
• Volume. Eight pages a month gets there sooner than two. This is the part you are actually choosing between when you compare plans, and it is most of what separates one price from another.

What makes it slower

And five that stretch it, none of which are a reason to stop.

A new domain, as above, with no history for Google to lean on.

A rebuild first. If the site has to be replaced before the work can stand on it, the first weeks go into the build and the clock on content starts later. That is a real cost of a bad starting point rather than a delay someone invented.

A competitive market, where the businesses above you are doing the same work and have been for years.

Thin publishing. Two posts a month on a site with nine services will take years to cover them, and the arithmetic is worth doing before you judge the pace.

A previous provider who did something harmful. Bought links and spun content take time to clean up before anything new can help, and this one is genuinely unfair, because you paid for it twice.

The season problem

Judging a trade by the wrong month is the most common way owners conclude the work failed.

A roofer in February and a holiday light installer in March are looking at the floor of their own year. A furnace company in September is looking at the calm before it. If you compare this month against last month, the season does the talking and the work is invisible underneath it.

Compare the same month a year apart, or compare impressions and query counts, which move with the work rather than the weather. And do the building in the quiet season on purpose, because pages published in the off season are ranked and ready when the phones start.

When it is fair to conclude it is not working

Patience has a limit and it should. Here is where I would draw it.

At three months, with no new pages published, no profile changes, and no technical fixes, the problem is not the timeline. Nothing was done, and more months will not fix that.

At six months, with the work genuinely done, no rise in impressions, no growth in the number of searches you appear for, and no map movement, something is wrong with the targeting rather than the pace. Usually the pages are aimed at searches nobody makes, or at searches you cannot win yet.

At twelve months, with impressions up, positions improved, and still no calls, the search work succeeded and the site is losing the visitor after arrival. That is a different problem entirely, and it is the one in what makes someone call.

None of those three conclusions requires you to understand search. They require a report that says what was done and a queries report you can read, which is why both belong in any arrangement you sign.

One clock that runs differently

AI answers do not follow the same schedule, and it cuts both ways.

Getting named by an assistant often turns on facts it can verify, so fixing your hours, your license record, and the pages that state what you charge can change an answer faster than the same work changes a ranking. I have seen the details in an answer follow a profile fix within weeks.

It is also less stable. Ask the same question twice and the names can differ, so a single answer proves very little and a monthly count across the same prompts proves something. That method is in AI visibility, and it is worth starting in month one precisely because it moves on a different clock from the rankings.