Chapter 4: Content Strategy and Marketing

Content production SOP, from idea to iteration

1. Source the topic from evidence, not from a brainstorm. Feed the calendar from four streams: the query map, competitor gaps, sales objections, and support tickets. A topic with no evidence behind it is a writing exercise.
2. Confirm the target query and intent. Search volume, difficulty, and — more importantly — what the SERP says the searcher wants. If the top ten is all product pages and you plan a tutorial, pick a different query or expect to lose.
3. Audit the top five. Note format, depth, subtopics covered, freshness, and whether any of them deserve their position. Your brief is built out of what they miss.
4. Write the brief before the draft. Target query, secondary queries, audience, angle, required sections, internal links, sources to cite, what not to say. A brief is a decision document; without it the writer makes your strategy decisions for you.
5. Draft with a real person’s experience in it. Whoever writes needs access to the practitioner — a call, a transcript, a screenshot of the actual dashboard. E-E-A-T is not a markup problem; it is whether the page contains something only a practitioner would know.
6. Edit for structure and retrieval. Opening answer first, one idea per section, headings that read like the questions people ask, short paragraphs, tables where comparison is the point.
7. Fact-check and attribute. Every number gets a source and a date. If a claim cannot be sourced, either soften it to a range or label it as field experience. This is the single highest-value editorial rule on a reference site.
8. Design and typeset for skimming. Add the diagram, the comparison table, the checklist. A wall of competent prose loses to a mediocre page with better structure.
9. Publish properly. Slug, schema, internal links from and to the page, mobile check, date shown.
10. Distribute once, deliberately. Newsletter, the two communities where the audience actually is, and direct outreach to anyone quoted or referenced. Publishing without distribution is a coin flip.
11. Review at one week, one month, one quarter. Week one tells you whether the page is indexed and associated with the query. Month one tells you whether it ranks. Quarter one tells you whether it deserves to exist.

Interview questions and answer frameworks

Q1: How do you make content that both ranks and genuinely helps the reader?

Framework: Treat them as the same goal, because the engine is now measuring the same things readers care about. Confirm real demand first — a page nobody searches for cannot help anyone. Answer the core question in the first paragraph, so both a skimmer and a retrieval system get the answer immediately. Cover the sub-questions the SERP implies but does not answer well, which is where information gain comes from. Add something that only you can add: a test result, a screenshot, a number you measured, a practitioner’s judgement call. Then verify with data — CTR, scroll depth, returning visitors — and be willing to rewrite when the data disagrees with the draft.

Q2: What is a topic cluster, and how would you build one?

Framework: A pillar page that covers a subject broadly, plus cluster pages that go deep on each sub-question, all interlinked so the engine reads them as one body of expertise. Build it in five moves: choose the subject you intend to own commercially; mine the full query set around it and cluster by SERP overlap rather than by keyword stem; write the pillar as the map, not the encyclopedia; write one cluster page per sub-intent with a single clear promise each; and link hub-to-spoke and spoke-to-hub so no page is an orphan. The test of a good cluster is that a reader who lands on any page can find the next thing they need in one click.

Q3: An old post is losing traffic. What do you do?

Framework: Diagnose the loss before rewriting. Check whether rankings fell for specific queries, whether demand itself fell, or whether impressions held while CTR collapsed — those are three different problems. If rankings fell, look for new competitors, a SERP format change, or a freshness signal you are failing. If CTR fell, the title and snippet are the problem, not the body. Then decide between three actions: refresh (update facts, add the missing sections, improve structure), consolidate (merge into a stronger page and redirect), or retire. Give the refresh a defined review window and a success criterion agreed in advance, so the decision at the end is not a matter of taste.

Q4: How do you keep quality consistent when you are publishing at volume?

Framework: Standardise the parts that should not vary and leave the parts that must. Standardise the brief template, the intro pattern, the source-and-date rule, the schema, and the internal-link requirement. Leave the angle, the examples and the judgement to the writer. Pair that with a two-tier review — a structural pass against the brief and a factual pass against sources — and a small library of three or four exemplar pages that show what “good” looks like here. Volume without a standard produces pages that are individually fine and collectively worthless.