Keyword research

Keyword research isn't just for ads

← Back to all posts
TLDRKeyword research tells you what people type into Google, how often, and what a click costs. I was only using it for ads, and even there it wasn't checking itself against the ad account. Now one pull feeds ads, the website and social, and the research and the account check each other.

I run keyword research for a practice I work with every time we touch a procedure. It's basically a list of what people type into Google about that procedure, how many people type it each month, what an ad click costs, and who already shows up on page one.

For months I treated it as an ads input. Pull the list, pick the searches worth paying for, hand it to the ads side, done.

Then I asked a question I should have asked earlier: is the research actually giving the ads engine what it uses to make decisions? I had Claude read how the ads engine decides things (it judges every ad against what a lead is worth to the practice, and it has its own block lists for searches we never want to pay for). The answer was partly.

What was off

Three things.

The research was quoting its own "what a lead is worth" number. The ads engine had updated its number the night before. So the research said a procedure could afford a lead at one price and the ads engine was using a much lower one. Two documents, two different answers to the same question.

The research would pick searches to buy without checking whether the account already blocked them. When we finally checked, an old block list had one line on it, "[procedure] cost", that quietly blocked every search with those words in it, in every campaign. Including the exact people you want, the ones asking what it costs. Someone added it months ago for a reason nobody remembers. It had been flagged once and then forgotten because that campaign was paused at the time.

And the ads template we start every campaign from had ad groups for "before and after" and "recovery" searches. Those people are looking, not booking. In one campaign that kind of search ate a real chunk of budget and didn't produce a single form fill.

The fix: the research and the account check each other

The rule now is the ads engine owns the "what a lead is worth" number. The research doesn't guess at it. What the research gives it instead is click prices from two places, side by side: Google's estimated range, and what we actually paid for that same search over the last 90 days. If we're paying way above Google's range, something's off. If we're under it (we mostly were, on the searches I checked), good to know before anyone raises a bid.

Same idea for blocked searches. The research builds its own list of searches we shouldn't pay for. Then it pulls the live block lists from the account (read only, it can't change anything) and marks every search one of three ways:

That conflict column is the one that found the "[procedure] cost" block. Nobody would have caught it by reading. It took two independent sources disagreeing.

And "before and after" and "recovery" searches are out of paid completely. They didn't stop mattering though, they just moved.

One pull, three uses

This was the bigger shift for me. The expensive part of keyword research is pulling the data. Once it's pulled you can group it three different ways.

Ads groups it by ad group. Only searches from people ready to book, with click prices and blocks checked.

The website groups it by page. Here the questions and "before and after" searches are gold. And the research shows something ads can't: where you rank and whether anyone clicks. For one search the practice sat near the top of page one and got almost no clicks over three months. Something above them (an AI answer, the map pack) was taking all of them. So "we already rank for free" turned out to be a bad reason to skip ads on that search. For another search the practice was buried on page four, and Google was showing a photo gallery page instead of the actual procedure page. That's a website fix, not an ads fix.

Social groups it by post or video. Every "what does recovery look like" or "is it worth it" search is a video topic in the patient's own words. And Google now shows TikTok, Reddit and Instagram right in the results for a lot of these searches, so the research can flag which searches are winnable with a short video instead of a blog post. (One limit: this only sees Google. It doesn't see what people search inside TikTok or Instagram themselves. That's a different data source.)

What I'd tell another practice

The research costs a couple of dollars per run. The block it found had been sitting there for months.

Most marketers stop at the click. I've worked on the inside.

If you're ready for someone who knows where leads actually turn into patients, let's talk.

I've spent more than a decade in marketing and operations, and the last 7 years in aesthetic marketing for surgeons and med spas. Today I work with eight-figure, multi-surgeon practices. I get into your CRM, your scheduling system and how your coordinators actually work, and I find where leads and numbers slip before I build anything. Every post here is a real example from that work.

Request a session

Tell me where leads are slipping. I'll tell you honestly if I'm the right fit.