How do I figure out what pages to write?

1. Mine what people already ask you

This is the part everyone skips and shouldn’t. You are sitting on the best keyword data there is — it’s just not in a tool. Pull it from where it lives: the questions that come up on every sales call, the recurring support tickets, the objections that show up right before someone signs (or doesn’t), the things your front desk explains ten times a week. Each of those is a page. They’re high-intent because a real buyer asked them, and they’re phrased the way real buyers phrase them — which is often not how the industry phrases them.

If you do nothing else, write down every question a prospect or customer asked you this month. That list is usually twenty to forty items and most of them deserve a page.

2. Widen with competitor coverage

Look at the two or three sites already ranking for the searches you want. Map their topics — not their words, their topics. Anything every one of them covers is table stakes: if you don’t have a page on it, you’re conceding that query before you start. Treat their coverage as the floor you have to clear, then look for what none of them did well — the question they answered thinly, the comparison they skipped, the local angle they ignored. That gap is where you actually win, because matching them only gets you to parity.

3. Layer in keyword research

Now bring the tools in — not first, third. Keyword research is for the things the other two methods miss: demand you didn’t know existed, the exact phrasing people search (often different from what they say out loud), related questions that cluster around the ones you already have, seasonal patterns. It’s a multiplier on a list you already built, not the list itself. A keyword tool will happily hand you a thousand “opportunities”; most of them aren’t, and only someone who knows the business can tell which.

4. Organise against the buyer journey

A pile of good page ideas isn’t a plan until it’s sorted by where the buyer is. Four rough stages:

• Problem-aware — they know something’s wrong, not what to do. (“Why does my X keep doing Y.”)
• Solution-aware — they know the category of fix and are learning about it. (“What’s involved in a Z.”)
• Comparing — they’re choosing between options or providers. (“X vs. Y,” “how much does Z cost.”)
• Ready — they’re picking who to call. (“Z in [neighbourhood],” “emergency Z near me.”)

Sorting your list this way does two things: it shows you which stages you’ve over-covered and which you’ve ignored, and it tells you what each page’s job is — to inform, to compare, or to convert. A page that doesn’t know which stage it’s serving usually serves none of them.

Tools tell you what people search. Only you know which of those searches is a customer.

The output, and the shortcut

Run those four steps and what you’ve got is a topical map — the page-by-page plan a real cluster gets built from. If you’d rather not run it yourself, that’s a large part of what we do: the authority sites build starts with this exact research, and a paid SEO audit delivers the map and the gap analysis on its own if you want to see the thinking before committing to the build. Either way, the sequence is the same — your customers’ questions first, competitors’ coverage as the floor, keyword data as the multiplier, the buyer journey as the sort order.