Weak-SERP Hunting: Spot Winnable Keywords

Full transcript

1,059 words · script draft, under review.

Intro

You search for something in your niche. And on page one, in position two, there's a forum thread from years ago. Three replies. One of them is wrong.

That's not a strong results page. That's Google saying: this is the best I've got.

Those moments are worth more than any difficulty score. A weak results page is an open door, and most people walk past it because they only look at volume and a KD number.

In this episode, I'll show you how to hunt for weak SERPs on purpose: the four signals I look for, how to turn them into a simple weakness score, and the traps that make a SERP look weak when it isn't.

Strength versus weakness

In the last episode we built a difficulty score. It asks: how strong is this results page overall?

Today's question is different. Where is it weak? A results page can have seven strong pages and still have three cracks. And you only need one crack in the top five to have a real chance.

So the difficulty score and the weakness score answer different questions, and you want both. Low difficulty and several weak spots: that's the hunt list. High difficulty with one weak spot: maybe. High difficulty, no weak spots: move on.

The four weak signals

Here are the four signals.

Signal one: forums and user-generated content. Forum threads, question-and-answer sites, comment sections. When they rank for a question that deserves a proper answer, it often means no good dedicated page exists. You detect this with the domain-type list from episode twenty-nine.

Signal two: outdated pages. A year in the title that's several years old. A date in the snippet. Products that aren't sold anymore. Prices that don't make sense. In the API result, look at the title, the snippet, and any date field the result carries.

Signal three: poor title match. We used this in episode thirty-one as part of difficulty. Here it works at the level of a single result. A ranking page whose title doesn't contain the core words of the search is usually ranking on general strength, not because it answers the question.

Signal four: thin pages. A short page, a stub, a category page with a list of links and nothing else. This one needs one extra step: you fetch the ranking page and measure how much real content it has. Do that only for the top five, and only for keywords that already passed the cheaper signals.

Turning signals into a score

Now make it a number. For each result in the top ten, mark it weak if it shows at least one of the four signals.

Then weight by position. A weak result in positions one to three is worth three points. Positions four to six, two points. Seven to ten, one point. Why? Because a crack near the top means Google is already sending real traffic to a page you could beat. A weak page at position ten is nice, but it doesn't move much.

Add up the points, and you have a weakness index for that keyword. It's simple on purpose. You can explain every point in it, and you can check it by looking at the SERP yourself.

Worked example

Let's run it on our example shop. Made-up numbers, for illustration.

We take two hundred coffee equipment keywords, pull the top ten for each, and cache the results. At depth ten, that's well under a dollar.

The cheap signals run first: domain type, dates in titles, and title match. Say thirty-five keywords have at least one weak result in the top five. Only those get the extra step, where we fetch the top five pages and measure content.

After that, say fifteen keywords have a weakness index of six or more. One example: "burr grinder for travel". Position two is an old forum thread. Position four is a general grinder roundup that mentions travel in one sentence. That's three plus two plus some smaller points further down.

Then we put the difficulty score from episode thirty-one next to it. Low difficulty and high weakness goes to the top of the list.

Where to look first

Where do you find keywords worth hunting in the first place? Three places work well for me.

Your own Search Console. Queries where you already get impressions, but no page on your site is really about them. Google is already testing you there. If the SERP is also weak, that's a very short path.

Keyword ideas from the API, which we pulled in episode twenty-three. Especially the long, specific phrases. Big companies rarely build pages for them, so the SERPs are often filled with near misses.

And questions from "People also ask" boxes in SERPs you already have. Each question is a search of its own, and many of them have never been answered by a dedicated page.

Run the cheap signals on all of them. Only pay attention to the ones that score.

When weak isn't weak

Now the traps, because this is where weak-SERP hunting goes wrong.

Trap one: forums that are the intent. Sometimes people want opinions and experiences, not a guide. If most of the top ten are discussions, Google may be serving exactly what searchers want. Check the intent from the SERP first, the way we did in episode thirty. A forum in a SERP full of guides is a crack. A forum in a SERP full of forums is the answer.

Trap two: old but still the best. An evergreen page from years ago can still be the most complete answer. Age is a signal to look closer, not proof of weakness.

Trap three: weak because nobody searches. Very often a SERP is weak simply because the demand is too small for anyone to bother. A weakness index says nothing about demand. That's why every keyword on the hunt list still has to pass a demand check, which is what episode thirty-five is about.

And one more: a weak SERP can change fast. If others find the same crack, it closes. In the next episode we look at how to measure that kind of movement.

Wrap-up

So: four weak signals, weighted by position, checked against intent and demand. That's how you find results pages you can actually win. The weak-SERP checklist and scoring sheet are linked in the description. Next episode: how much results pages move over time, and how to tell when Google wants fresh content. See you there.