The most damaging thing in an SEO project is not a bad tactic. It is a mismatch between when results actually arrive and when someone expects to see them. Projects get cancelled in month four that would have worked in month eight, and projects get extended for two years that were never going to work at all.
This article sets out what a realistic first year looks like, month by month, what to measure at each stage, and what a report should contain so that the decision to continue or stop is made on evidence rather than on mood.
| Period | Work | Visible result |
|---|---|---|
| Month 1 | Audit, baseline, page map, priorities | Nothing in traffic. A plan and a diagnosis. |
| Month 2 | Technical fixes, first pages rewritten | Crawl activity rises. Occasional early impressions. |
| Month 3 | Publishing, internal linking, local presence | First movement in the commercial tail. |
| Months 4-6 | Sustained publishing and authority work | Tail queries entering the top ten; non-brand clicks rising. |
| Months 7-9 | Depth, consolidation, mid-competition targets | Mid-difficulty queries moving; enquiries becoming attributable. |
| Months 10-12 | Head terms, refresh cycle, gap filling | Head terms starting to move; compounding becomes visible. |
Almost every project has one. Enough money has been spent to feel significant, and the head terms have not moved. This is exactly what a healthy project looks like at month four - which is why the metrics have to be agreed at month zero, when nobody is anxious.
The sequence is not arbitrary. Specific, lower-volume queries have less competition, so a genuinely good new page can rank for them within weeks. Head terms are contested by every established competitor and depend on accumulated authority, which by definition takes time.
The practical consequence is that month three should be judged on whether long, specific queries are entering the top twenty — not on whether "seo agency valencia" moved. A project can be working perfectly while its headline query has not shifted a place.
This is also why the tracked query set should be grouped rather than flat. Head, commercial tail, geo-modified and informational queries behave differently and on different timescales; averaging them produces a number that describes nothing. Grouping them in rank tracking from the start makes the early signal visible instead of drowned.
A two-hundred-row ranking table is not transparency. It is unreadable, which means the reader evaluates the dominant colour — and the dominant colour can be engineered by choosing which queries to include.
A useful monthly report has five numbers and three actions:
Non-brand clicks vs last month and last year. Share of visibility vs three competitors. Movement by query group. Pages published and indexed. Enquiries attributed.
What was done this month, what is planned next month, and what is blocked and by whom.
The full detail for anyone who wants to verify. Available, not the main document.
Fix the tracked query set at the start of each quarter and do not change it mid-quarter. Adding easy keywords halfway through to improve how the report looks is the most widespread way of misleading with entirely true data.
Month 2 with nothing changed on the website. If no page has been edited, no fix implemented and no content published, you are paying for documents. This is the earliest reliable warning and the one most often ignored.
Month 3 with no new pages indexed. Either nothing was published, or what was published cannot get into the index. Both need an answer now rather than at month six.
Month 6 with no movement in any query group. Not the head terms — any group. If nothing has moved anywhere, the targets are probably wrong, and continuing without revisiting them is spending on a hypothesis that has already failed.
Month 9 with traffic but no enquiries. The problem has moved. It is now conversion or intent, not visibility, and more traffic will not fix it.
Any month where the report cannot be verified. If claims cannot be checked in an account you control, the reporting is the problem regardless of what it says.
This trajectory is why SEO is a poor purchase for a business that may not exist in eighteen months, and an excellent one for an established business with a stable proposition. The pages built in year one keep working in year four with modest maintenance, while every paid click costs the same as it did on day one — usually more, as competition rises.
The compounding is also why stopping and restarting is so expensive. Content decays slowly rather than instantly, so a six-month pause does not lose everything — but the momentum, the publishing rhythm and the accumulated context are all rebuilt from scratch.
By month seven the conversation shifts from visibility to revenue, and this is where most reporting quietly falls apart - not because anyone is dishonest, but because attributing a customer to a channel is genuinely hard for local businesses.
The specific difficulty is that a substantial share of local enquiries arrive by phone, by WhatsApp, by walking in, or through a referral from someone who found you on Google three months earlier. None of those appear in an analytics report, and the ones that do appear are the least representative.
Three practical measures close most of the gap without building a data warehouse.
Ask, and record the answer. "How did you find us?" asked at every enquiry and written down in one place is crude, imperfect and better than every technical alternative available to a business of this size. The consistent recording matters far more than the sophistication.
Watch non-brand clicks as the proxy. If attributable enquiries cannot be counted cleanly, non-brand organic clicks are the closest honest stand-in for acquisition, because brand searches came from other marketing and direct visits came from people who already knew you.
Compare periods, not sources. If non-brand clicks rose 60% over two quarters and total enquiries rose 30% in the same window with no other change in marketing, the connection is reasonable even without per-customer attribution. Weak evidence used honestly beats precise evidence about the wrong thing.
What to avoid is the opposite failure: crediting SEO with every enquiry that arrived during the engagement. A supplier who reports total enquiries as an SEO result, without separating brand from non-brand or accounting for other activity, is producing a number designed to be renewed rather than understood.
The second year is a different project and it is frequently mispriced, in both directions. Some businesses cut the budget on the assumption that the work is done; others keep paying a build-phase retainer for maintenance-phase work.
The honest shape is that the composition changes more than the total. Building new pages gives way to updating existing ones, which is cheaper per page and higher return per hour. Technical work drops to monitoring, unless something is rebuilt. Authority work continues at the same modest pace, because relationships do not compound if they stop. And a genuinely new line appears: defending what you have won, which nobody budgets for in year one and which becomes real as soon as competitors notice you.
A useful test at the twelve-month review is to ask what would happen if you stopped entirely. On a healthy project the honest answer is that traffic would hold for two or three months, then decay slowly over a year as content aged and competitors published. That decay curve is what the second-year budget is buying, and framing it that way makes it a rational decision rather than an act of faith.
Ending a project honestly is a legitimate outcome, and knowing the criteria in advance makes it a decision rather than a fight. Three that we consider valid:
The market is not there. If the total search volume for every query you could realistically win is smaller than the traffic you need, no amount of execution changes the arithmetic. This should have been caught in month one, but it is occasionally only clear later.
The competition is structurally unbeatable. National portals and marketplaces with incomparable resources occupying every commercial query is a real situation. The correct response is to route around it — long tail, local pack, paid — not to keep spending on the head.
The business cannot supply what the channel needs. If nobody internally can give two hours a month, approve content or supply sector accuracy, the output will stay generic regardless of the supplier. That is a resourcing decision, not an SEO failure.
What is not a valid reason to stop: month four with no head-term movement. That is the plan working as designed.
Baseline, grouped query set and an indexing log - in an account you own. Set up before the work starts, it turns every later review into a factual conversation.
Sign in to Semalt See Google SERPMovement in the commercial tail around month three, meaningful non-brand click growth around months four to six, head terms from month nine onwards.
Not in head terms. It is a warning if nothing has moved in any query group, or if nothing has changed on the website.
For most local service businesses, break-even in year two and clear returns in year three. Budget for that horizon or choose a faster channel.
Five numbers and three actions, with the detail available as an annex. Anything longer is not being read, which is often the point.
You can - decay is slow. What you lose is momentum and accumulated context, which is more expensive to rebuild than the pause saves.
Most SEO relationships fail on expectations rather than on execution. The work follows a predictable curve — activity, then indexing, then impressions, then tail positions, then commercial queries — and almost every argument comes from someone judging an early stage by a late-stage measure.
Fix that at the start. Agree what will be measured at month three, month six and month twelve, write it down, and hold the measurement in an account you own so the numbers are not produced by the party being evaluated. It takes ten minutes and it is free.
If you want a realistic timeline for your specific situation — including an honest view on whether the market supports the numbers you need — get in touch. The initial audit is free and delivered within 48 hours.
Free analytics for Google Search, SERP rankings and AI answers - plus fast indexing for your new pages.
Sign in to Semalt semalt.comGet in touch for a free audit
