The 9-Step SEO Content Strategy Framework We Use With Every Client

Step 1 — Interview the people who talk to customers

Every content strategy we have ever built started with a conversation that had nothing to do with keywords. Before a single tool is opened, we sit down with sales, support and — where possible — customers themselves.

The reason is practical: keyword tools tell you what phrases have volume, but they cannot tell you which of those phrases signal a person who will buy. Your sales team already knows. They hear the objection that kills deals, the comparison every prospect raises, and the phrase buyers use that nobody in marketing would ever write.

The questions worth asking are boringly specific:

• What do prospects ask in the first ten minutes of a call?
• Which competitor comes up most, and what do buyers believe about them?
• What do people misunderstand about how the product works?
• Which objection, if answered earlier, would shorten the sales cycle?
• What does a customer say in the moment they decide to buy?

Record everything. Those transcripts become headline copy, FAQ sections and — more often than you would expect — entire articles that no keyword tool would have surfaced.

The most profitable article we wrote last year came from a support ticket, not a keyword tool. It targeted 90 searches a month and converted at 11%.

Step 2 — Build the keyword universe

Now open the tools. The goal at this stage is coverage, not precision: you want every term your market could plausibly search, from every source available.

We pull from seven places, in this order:

1. Your own Search Console data. Queries where you already rank between positions 8 and 30 with meaningful impressions are the cheapest wins on the internet.
2. Competitor organic keywords from Ahrefs or Semrush, filtered to pages that get real traffic.
3. Seed expansion from your product vocabulary, category names and use cases.
4. Autocomplete and People Also Ask for every priority seed.
5. Community sources — Reddit, industry forums, review sites, Slack communities — where people ask questions in their own words.
6. Your internal site search logs, which reveal what visitors could not find.
7. Sales and support language from step one.

Expect a list of several thousand rows. That is fine — nobody will ever look at most of it. The list exists so that the filtering in the next steps has something to filter.

Step 3 — Classify intent before you look at volume

Volume is seductive and mostly irrelevant until you know what the searcher wants. We classify every candidate term into four buckets, and the classification is made by looking at the results page — not by guessing from the phrase.

The rule that saves the most money: if the top ten results are all listicles, you cannot rank a product page there, no matter how good your product page is. Either publish the format the SERP rewards, or choose a different keyword.

Step 4 — Score winnability honestly

Keyword difficulty scores from tools are a crude proxy for backlink strength. They do not know your domain, your topical authority or your resources. We replace them with a five-point winnability score that combines four inputs:

• SERP strength — the median authority of pages currently ranking, and whether any of them are weak (forums, outdated posts, thin pages).
• Content-type match — whether we can credibly produce the format the SERP rewards.
• Topical authority — how much related content we already have or plan to publish in that cluster.
• Business value — proximity to revenue, scored by the sales team, not by marketing.

Terms scoring 4 or 5 go into year one. Terms scoring 1 or 2 go into a “later” tab, revisited each quarter as the domain gets stronger. This single step is what stops teams burning six months on head terms they were never going to win.

Step 5 — Cluster by SERP overlap, not by string similarity

Here is the mistake that produces cannibalisation in almost every content programme we audit: grouping keywords because they contain the same words.

“Content audit” and “content audit template” look like siblings. Check the SERPs and you will often find they share fewer than three results — because one wants a guide and the other wants a downloadable file. They are two pages. Meanwhile “how to check if my content is working” might share eight of ten results with “content audit” and belongs on the same page.

The method:

1. Pull the top ten URLs for every priority keyword.
2. Compare overlap between each pair of keywords.
3. Three or more shared URLs: same page. Fewer: separate pages.
4. Name the cluster after the intent it serves, not after the keyword.

Do this for 60 keywords and you will typically end up with 12 to 20 clusters — which is your content plan, already prioritised by how much search demand each cluster represents.

Step 6 — Design the architecture and the internal links

A cluster needs a shape. Ours is conventional because it works: one pillar page targeting the broad term, four to eight supporting articles targeting specific sub-questions, and deliberate links in both directions.

What gets forgotten is the third link type. Supporting articles should link out of the cluster to your commercial pages, otherwise you build a beautiful content hub that never touches revenue. Every informational article in our plans carries at least one contextual link to a service, product or comparison page.

Document this in the strategy as a table: URL, target keyword, cluster, links in, links out, CTA. That table is what makes the plan executable by somebody who was not in the room.

Step 7 — Sequence the calendar around dependencies

Order matters more than volume. Three rules govern our sequencing:

• Money pages first. If your service or category pages are not optimised, blog traffic has nowhere useful to go.
• Supporting articles before pillars. A pillar page ranks because the cluster around it exists. Publishing the pillar alone and waiting is a common, expensive mistake.
• Refresh before you write. If existing pages sit in positions 11–30, updating them will produce results faster and cheaper than anything new. See our content refresh playbook.

Then set a volume your team can genuinely sustain. Four excellent pieces a month beats twelve rushed ones, and consistency matters more than either — a quarter of silence undoes a quarter of momentum.

Step 8 — Write briefs that remove guesswork

The brief is where strategy becomes execution, and it is the deliverable clients tell us they value most. A brief that only contains a keyword and a word count is not a brief; it is a hope.

Ours contain: the target keyword and secondary terms; the intent and funnel stage; the reader (with the objection they arrive carrying); the angle and why it differs from the top-ranking pages; a full outline with H2s and H3s; the questions that must be answered; sources to cite; internal links to include; the CTA; the expected length based on the SERP; and the metric that will judge the piece.

Write that, and any competent writer — in-house, freelance or agency — can execute it. Skip it, and you will spend the revision round rebuilding the article instead of improving it.

Step 9 — Measure at 30, 60 and 90 days

Content strategies fail quietly. A piece is published, nobody checks it again, and six months later the team concludes “SEO does not work for us”.

Build the checkpoints into the plan:

Report on the pieces that failed as well as the ones that worked. A strategy with no failures is a strategy nobody is being honest about — and the failures are where the next quarter's improvements come from.

If you would rather have this run for you end to end, that is exactly what our SEO content strategy service does.

Frequently asked questions

Silvia Moreno

Co-founder & Content Strategist

Silvia spent eight years in newsrooms before moving into SEO. She builds the keyword maps and content strategies behind every Witty Content Writers engagement.