Most SEO strategy templates are either a project management board with no thinking in it, or a framework you have to sign up to download. This one is the whole document, on the page, free to copy.
A strategy is not a task list. It is the set of decisions that explain why those tasks and not others, and a template is only useful if it forces those decisions out of your head and onto paper.
What belongs in an SEO strategy, and what does not
A strategy answers four questions: where you are, where you are going, what you will do to get there, and how you will know whether it worked.
Everything else belongs in a different document. Task assignments and due dates go in your project tracker. Deliverables and payment terms go in the contract. The pitch goes in the proposal.
Mixing them is why so many strategy documents are forty pages that nobody reads. Keep the strategy to the decisions, and let the other documents carry the detail.
The full template
Copy everything below and replace the bracketed prompts. Aim for one page per section at most. If a section runs longer than that, you are probably writing a report rather than a decision.
1. Objective
[One sentence. The business outcome, not the SEO metric. “Grow qualified demo requests from organic search”, not “increase traffic”.]
[The target, with a date and a range. “From 40 to between 70 and 110 a month by 30 June.”]
2. Where we are now
[Current organic sessions, conversions and average position, each with the date you pulled them. These are your baseline and every later claim of progress is measured against them.]
[Indexed pages versus published pages. A gap here is a problem to solve before anything else.]
[Referring domains, and how many are genuine rather than directory noise.]
[The three pages that already earn the most impressions. These are usually where the fastest gains sit.]
3. The constraint
[One sentence naming the single thing most limiting growth right now: authority, content coverage, technical health, or site structure.]
[Evidence for that claim. Not an opinion, a number.]
[What happens if we work on everything except this. Being explicit here stops the strategy drifting back to whatever is easiest.]
4. Audience and intent
[Who we are trying to reach, and what they search when they have the problem we solve.]
[The intent types we will target: informational, commercial, transactional, or a stated mix.]
[Who we are not targeting. This line saves more time than any other in the document.]
5. Topic territory
[The three to five topic clusters we intend to own, with the reason we can credibly own each one.]
[Topics we are deliberately leaving alone, and why.]
6. The plan
[New content: how many pieces, in which clusters, at what cadence.]
[Existing content: which pages get improved and on what trigger. Pages sitting in positions 11 to 30 usually come first.]
[Technical: the specific fixes, in priority order.]
[Authority: how links will be earned, by whom, and at what rate. If this section is empty, say so plainly rather than leaving it blank.]
7. What we expect to happen
[A low, middle and high case, each with a date. Build these the way our SEO forecasting guide sets out, and list the assumptions underneath.]
[What we would see at the 30, 60 and 90 day marks if the plan is working. Early signals, not final results.]
8. How we will measure it
[The two or three numbers that decide whether this worked, drawn from your SEO KPIs.]
[The leading indicators we watch in between, such as impressions and average position.]
[Reporting cadence, and who reads the report.]
9. Risks and dependencies
[What could break this: a migration, an algorithm update, a developer queue, a budget review.]
[What we need from other people, and by when.]
10. Review date
[The date this document gets re-read and revised. A strategy with no review date becomes wallpaper within a quarter.]
How to fill the two hardest sections
Sections three and seven are where most strategies go soft, because both of them require you to commit to something that could be proved wrong.
The constraint is hard because naming one thing means deprioritising the others. The test is simple: if you doubled output on everything except your named constraint, would results move? If the honest answer is no, you have named it correctly.
A common and uncomfortable outcome is that the constraint is authority rather than content. When a site has very few referring domains, publishing faster does not fix it, and a strategy that says otherwise is avoiding the finding.
What we expect to happen is hard because it invites being wrong in public. Give a range rather than a single number, and write the assumptions directly underneath so the forecast can be argued with on its inputs.
For section six, the existing-content line is usually the highest value and the most skipped. Pages already sitting on page two or three need much less work than new pages to produce the same gain, which is what striking distance keywords are about.
A filled example
A blank template is easy to fill badly. Here is the same document completed for an invented B2B software company, kept short on purpose, so you can see the level of specificity each section needs.
1. Objective. Grow qualified trial signups from organic search. From 22 a month to between 45 and 80 by 30 June 2027.
2. Where we are now. 3,400 organic sessions and 22 trial signups a month, average position 34, pulled 18 October 2026. 61 pages published, 44 indexed. 19 referring domains, of which roughly 8 look genuine. Highest impression pages are the pricing page, the integrations list and one comparison post.
3. The constraint. Indexing, then authority. 17 of 61 pages are not indexed, so nearly a third of what we have written cannot rank at all. If we publish another 20 pages without fixing that, we should expect roughly a third of those to disappear too.
4. Audience and intent. Operations managers at 50 to 500 person companies, searching for ways to replace a manual process. Targeting commercial and comparison intent first, informational second. Not targeting students, job seekers or developers looking for an API.
5. Topic territory. Three clusters: comparisons against named competitors, the manual process we replace, and integrations with tools our buyers already run. We can credibly own these because we have customer data on all three. Leaving general productivity advice alone, since we have nothing distinctive to say about it.
6. The plan. New content: six pieces a month, four in comparisons and two in integrations. Existing content: fix the 17 unindexed pages first, then improve the nine pages sitting between positions 11 and 30. Technical: resolve the indexing causes, then internal linking from the three strongest pages into each cluster. Authority: two founder interviews and three integration partner listings per quarter, owned by the founder, not outsourced.
7. What we expect to happen. Low case 30 signups, middle 55, high 80, all by 30 June. Assumes the indexing gap closes by December, that comparison pages reach an average of position 12, and that our measured 0.9% visit to trial rate holds. At 30 days we expect indexed pages up, not traffic. At 60 days, impressions rising on comparison queries. At 90 days, first position gains.
8. How we will measure it. Trial signups from organic, and indexed pages against published pages. Leading indicators: impressions and distinct query count, reviewed monthly by the founder and the marketing lead.
9. Risks and dependencies. A planned site redesign in Q2 could reset rankings if URLs change. We need developer time for the indexing fixes, estimated two days, by mid November.
10. Review date. 20 January 2027.
Notice how much of that is numbers with dates on them. The sections that resist specificity, usually three and seven, are the ones carrying the actual decisions.
Mistakes that make a strategy useless
- No baseline with a date. Without it, every later report is an argument about whether things improved.
- Traffic as the objective. Traffic is a leading indicator of a business outcome, not the outcome. A strategy aimed at traffic will cheerfully chase visitors who never convert.
- No named constraint. A plan that improves everything a little is usually a plan that changes nothing.
- A keyword list instead of topics. Four hundred keywords in a spreadsheet is research output, not a strategy. Clusters you intend to own are the strategic unit.
- An empty authority section. Leaving it blank does not make the problem optional, it just makes the document quiet about the hardest part.
- No review date. Conditions change within a quarter, and a document nobody revisits stops describing reality long before anyone notices.
How far ahead should the strategy run
Six to twelve months is the usual window, and the reason is practical rather than conventional. Below six months there is not enough time for ranking changes to show, so you end up judging the plan before it can have worked.
Beyond twelve months the assumptions decay. Search results change, competitors publish, and features that take clicks today may not exist next year.
A reasonable compromise is a twelve-month direction with a detailed first quarter, reviewed every three months. The direction gives the work coherence and the quarterly review keeps the detail honest.
Keeping it alive after it is written
The difference between a strategy that works and one that gathers dust is whether anything in your routine refers back to it.
Tie the monthly report to section eight, so progress is always reported in the terms the strategy promised. If the report measures something the strategy never mentioned, one of the two documents is wrong.
At each review, score the previous period against section seven. Over two or three cycles this tells you whether your own estimates run optimistic or conservative, which makes every later forecast better. Then update the baseline in section two and date it again.
Frequently asked questions
What should an SEO strategy document include?
It should answer where you are, where you are going, what you will do, and how you will know it worked. In practice that means an objective tied to a business outcome, a dated baseline, a named constraint, your audience and topic territory, the plan itself, an expected outcome with assumptions, the measures, the risks, and a review date. Task lists and payment terms belong in other documents.
What is the difference between an SEO strategy and an SEO plan?
The strategy holds the decisions and the reasoning: which topics, which audience, which constraint you are attacking and why. The plan is the sequence of work that follows from those decisions. A plan without a strategy is a list of tasks with no test for whether they are the right ones.
How long should an SEO strategy cover?
Six to twelve months works best. Shorter than that and rankings have not had time to move, so you judge the plan before it could have worked. Longer and the assumptions decay as results pages and competitors change, which is why a twelve-month direction with a detailed first quarter and quarterly reviews is a practical compromise.
Should the strategy include a traffic forecast?
Yes, but as a range with its assumptions written underneath rather than a single number. A low, middle and high case is more useful than a point estimate, because the gap between them tells the reader how much uncertainty they are funding. If the low case does not justify the work, the decision does not depend on the forecast being right.
How often should an SEO strategy be reviewed?
Quarterly is enough for most sites, and the review date should be written into the document itself. At each review, update the baseline with fresh dated numbers and score the previous period against what you predicted. A strategy with no scheduled review stops matching reality long before anyone notices.
What if the strategy shows that content is not the problem?
Then write that down and change the plan. On sites with very few referring domains, the limit is usually authority rather than content volume, and publishing more does not move it. A strategy that identifies the real constraint and says so is more valuable than one that recommends comfortable work.