How Long Does SEO Take to Work, and How to Tell If It Is

Four months in. The monthly report says organic sessions are up, impressions are up, and there is a slide with a line going the right way. Your bank balance has not noticed. Nobody on the call is lying to you, nothing is visibly broken, and you have no way to tell whether the right move is patience or a different agency.

I read the pages ranking for this question before writing this. Every one of them hands you a month‑by‑month table — month one audit, month three content, month six results. Not one separates the organic traffic you earned from the organic traffic your brand already had. That distinction is the entire answer to whether your SEO is working, and it is missing from all of them.

How long does SEO take to work?

Three to six months for movement, and nine to twelve before it compounds into something worth the invoice. That is the position published on the rest of this site and it has not needed revising. Technical fixes are the exception and can move things within weeks, because there you are usually removing something that was actively hurting you rather than building something new.

The honest caveat is that those windows describe an established store with existing pages and some history. A store launched last month is answering a different question, which is whether Google has indexed it at all — covered separately in why a site is not showing up on Google.

Anyone promising page one in 30 days is describing terms you already ranked for. That is not a figure of speech, it is the specific mechanism, and the next two sections are about how it hides inside a normal‑looking report.

The report is up and the revenue is not

The symptom is precise enough to recognise: total organic sessions rising, impressions rising, average position improving, and orders flat. Every number on the report is real. The report is still telling you almost nothing.

It happens because “organic traffic” is two different businesses added together. Some of it arrives on searches containing your brand name — people who already knew who you were and used Google as a address bar. The rest arrives on searches for the product, the category or the problem, from people who had never heard of you. Only the second kind is something an SEO program created. The first kind is demand your ads, your email list, your packaging and your customers created, and it would show up whether or not anyone was doing SEO at all.

Why branded traffic makes a stalled program look healthy

Branded search is the most reliably growing line in most ecommerce accounts, and it is the line SEO has the least to do with. That combination is why a stalled program can produce a report that goes up and to the right for a year.

Branded queries also convert far better than anything else, because the person already decided. So a small rise in branded sessions moves revenue visibly, while a large rise in non‑branded sessions may move it barely at all in the first months — new visitors at the top of a category search are not buying on the first visit. Reported together, those two effects can cancel out into a flat line that nobody can explain, or add up into a rise that nobody can attribute.

Separate them and both problems disappear. Non‑branded impressions and clicks are the scoreboard for the work being paid for. Branded is a scoreboard for the brand, which is worth watching for its own sake, in its own row.

The ten‑minute check in Search Console

You can settle this yourself, today, without asking anyone. Open Google Search Console and work through these in order.

1. Performance → Search results. Set the date range to the last six months and turn on the comparison with the previous six. Make sure Impressions and Average position are both switched on, not just Clicks.
2. Add a query filter: Queries → Custom → does not contain → your brand name. Do it again for the obvious misspellings and for your name without spaces. What is left is the non‑branded business, which is the only view that answers the question.
3. Read impressions before clicks. Impressions move first. If non‑branded impressions are clearly up over six months, something is working even if the clicks have not arrived. If they are flat, nothing is being surfaced for anything new and no amount of waiting changes that.
4. Switch to the Pages tab with the same filter on. Filter the URL to your collection pages, then your product pages. Category and product templates are where ecommerce revenue lives. Growth that is entirely in blog URLs is real traffic and usually not buyers.
5. Open the Page indexing report. Look at how many URLs sit under “Discovered — currently not indexed” and “Crawled — currently not indexed”.

Know what the words mean before you draw a conclusion from them. Google’s definitions of impressions, position and clicks make average position a relative ranking averaged across every impression, which is why it moves for reasons that have nothing to do with revenue. And the page indexing report documentation is specific: “Discovered — currently not indexed” means the page was found but not crawled yet, typically because crawling it was expected to overload the site, while “Crawled — currently not indexed” means it was fetched and not added to the index. Those two states mean different things and the second is the more uncomfortable one.

That check tells you whether the pages you changed are moving. It will not tell you whether something structural is holding them back. SEO Karbon looks for exactly that from the outside, free and with no login, ranked by the revenue at risk. A stalled programme is sometimes just patience, and sometimes a crawl problem nobody went looking for.

What should move, and in what order

The order is fixed by mechanics, not by opinion, and it is the most useful thing to hold in your head. A page cannot get clicks before it gets impressions, cannot get impressions before it is indexed, and cannot be indexed before it is crawled. So progress shows up in the early rows long before it shows up in the last one.

Google’s SEO Starter Guide puts the front of that chain plainly: “Some changes might take effect in a few hours, others could take several months,” and it advises waiting a few weeks before judging whether a change helped. That is the official version of the same point — the early rows move on their own schedule and the last row is not a reliable early signal of anything.

How do I know if it is stalled rather than just slow?

Three tests, and between them they separate a program that needs more time from one that needs replacing. Run them at the end of a full quarter, not before.

• A meaningful share of the new pages is not indexed. That is not slowness. Work sitting in “Discovered” or “Crawled — currently not indexed” after weeks is a signal about crawl priority or page quality, and waiting another quarter will not resolve it on its own.
• Non‑branded impressions are flat over a full quarter. Impressions are the cheapest thing to move and the first thing that moves. Flat impressions after three months means nothing new is being surfaced at all, which is a stalled program rather than a slow one.
• Impressions are up, position is stuck outside the top twenty, and clicks are flat. This one is genuinely ambiguous, and it is the honest middle case. You are being surfaced and not chosen, which is usually a competitiveness problem rather than a time problem — and more months of the same work is the least likely thing to fix it.

There is a fourth pattern worth naming because it is the one people least expect: rankings improving genuinely and sharply while the revenue does not follow. Petticoat Lane is the account where that happened to me, and the case study says so in those words — rankings up sharply, and the organic revenue that has not followed yet — rather than showing the position chart and leaving you to assume it meant sales. Positions are a leading indicator, not a result, and they can improve on terms that were never going to convert. Ask to see the pairing, always.

What actually makes SEO faster?

Removing things is faster than adding them, which is the one lever that genuinely compresses the timeline. Everything below is ordered by how quickly it tends to show.

1. Fix what is actively suppressing you. A page blocked from indexing, a botched migration, duplicate collection URLs competing with each other. You are not building authority here, you are removing a brake, and brakes come off fast.
2. Improve pages that already have impressions. A page ranking at position 12 with impressions already accumulating is most of the way there. Rewriting its title and content is a far shorter path than a new page starting from nothing.
3. Point internal links at the pages that matter. Free, immediate, and the most consistently underused thing in an ecommerce catalog. The Shopify SEO checklist has the order worth doing this in.
4. Publish new content. Genuinely useful and genuinely slow. It belongs in the program, it does not belong at the front of it, and an agency that opens here has chosen the slowest available option first.

When is it reasonable to stop paying?

When a full quarter has produced no movement in non‑branded impressions and no improvement in indexation, and the monthly report has never once separated branded from non‑branded. That third condition matters as much as the first two, because an operator who does not measure the distinction cannot be steering toward it.

Before that point, the fair questions are narrower: what specifically is expected to move next quarter, which pages, and which of the stages in the table above it should show up in first. A good answer names pages and a stage. A bad one says SEO takes time, which is true, and is also what you would say if nothing were happening. If you are weighing that decision against a change of supplier, the criteria are in how to judge an ecommerce SEO agency, and the way I work on Shopify SEO starts from the same reporting split.

Questions people ask