SERP Gates: Validate Before You Build

Full transcript

1,011 words · script draft, under review.

Intro

I once spent weeks on a project in a niche I liked. Everything looked right: the idea, the products, the pages. Then I finally measured the demand properly. It was so small that most of the keywords didn't even register in the data.

I stopped working on it. The weeks were gone.

Nothing in that project was badly made. It just should never have been started. And the check that would have stopped it takes an afternoon.

In this episode, I'll show you the three gates I now run before building any page or project: a demand floor, a winnable results page, and the right page type. And what to do when something only passes two of them.

Why gates, not scores

So far in this module we've built a lot of measurements. Difficulty, weakness, volatility, device and location differences. All of them are useful. But a pile of measurements doesn't make a decision.

A gate does. A gate is a simple rule with a clear outcome: pass or fail. It's decided before you look at your favourite idea, so you can't bend it to fit.

Scores are for ranking options that are all viable. Gates are for deciding which options are viable at all. Gates come first. In the gap analysis module, starting in episode thirty-seven, everything we rank will already have passed them.

Gate one: demand

Gate one is demand. Is there enough search demand to justify the work?

My own rule of thumb is a floor of about five hundred searches a month for a project, before I build anything. Not for every single page, but for the core of what I'm planning. You might pick a different number. What matters is that you pick it before you look at the data.

And the number has to be honest. Four checks from module C.

Validate the word form. Singular and plural can have very different volumes, as we saw in episode twenty. Run a small validation task first, about nine cents, before you trust a big list.

Don't sum synonyms. When variants show exactly the same volume, they're one group in Google's data. Count it once.

Use buying phrases, not just head terms. A bare head term often measures people looking for information, not people who want what you're building. Episode twenty-five covered how to break it down.

And treat zero carefully. Zero means below the measurement floor, around ten a month. It's not proof of no demand, but it also doesn't count toward your floor.

Gate two: a winnable SERP

Gate two: can you realistically get onto page one?

This is where episodes thirty-one and thirty-two come in. Your difficulty band tells you how strong the results page is. Your weakness index tells you whether there's a crack. For the gate, I keep it simple. The core keywords of the project must be in the approachable or possible band, or have a clear weakness in the top five. If every core keyword is hard with no weak spot, the gate fails.

And check the results page as it looks today, with a live call, on the device and location your audience uses.

Gate three: the right page type

Gate three: can you build the type of page Google ranks?

Look at the top results the way we did in episode thirty. If the results are product pages from shops and you're planning a guide, you're building the wrong thing. If they're tools or calculators and you can't build one, that's a fail too.

This gate is the one people skip, because it feels obvious. But it's the most common reason a good page with real demand never ranks.

Worked example: three ideas

Let's run three made-up ideas for our example shop through the gates.

Idea one: a section about travel coffee gear. Demand after validation, synonyms counted once, buying phrases only: around seven hundred a month. Results page: possible, with forum threads in the top five. Page type: buying guides and product roundups, which we can build. Three passes. Go.

Idea two: a guide to coffee history. Lots of searches, but almost all informational, and the results are dominated by encyclopedias and big publishers. Demand passes on paper, but the buying phrases are tiny. Difficulty: hard, no weak spots. No-go.

Idea three: a grind size calculator. Demand: around three hundred a month, below the floor. Results page: weak, only a few simple tools. Page type: a tool, which we could build. Two out of three.

That third verdict matters. Not every fail is a no-go. When something misses the demand floor but is otherwise strong, I park it. It goes on a list with a date to check again. Demand changes, and a parked idea costs nothing.

Recording the decision

One more habit, and it's the one that pays off later. Write the decision down.

For every idea, I log the date, the result of each gate, the verdict, and where the data came from: which volume pull, which SERP snapshot. For parked ideas, I add a date to check again. Three months is usually enough.

Why bother? Because six months later, someone will ask why you never built that section. You'll have the answer in one line, with the data behind it. And when a parked idea comes back with higher demand, you can see exactly what changed.

This log also becomes evidence in the gap analysis report we'll write in episode forty-five.

Common pitfalls

Three pitfalls.

One: moving the goalposts. If you lower the floor because you like the idea, you don't have a gate anymore.

Two: passing a gate on one source. Demand from one tool, a SERP from one day. Cross-check the ones that are close.

Three: skipping the gates for small things. A single page is cheaper than a project, but the same three questions take five minutes. Ask them anyway.

Wrap-up

So: three gates, decided in advance. Demand above your floor, a winnable results page, and a page type you can build. Go, no-go, or park. The gate checklist is linked in the description. Next episode: competitor analysis from the SERP, finding out who really ranks across your keywords, and which of them are worth picking. See you there.