Search insights should shape more than ads

A few months ago I was reviewing search data for a fintech startup. Their product helped small businesses manage cash flow.

In the search term report, I kept seeing variations of searches about whether business owners would be able to pay their staff that month. (I'm paraphrasing to protect the client, but the emotional intensity was unmistakable.)

That's not the kind of phrase you'd find in a keyword tool. It's too raw, too specific, too human. But real people were typing things like that into Google at probably 2am, stressed out, looking for answers.

The company's homepage talked about "cash flow optimisation" and "financial visibility." Those words are fine. But they don't speak to someone who's panicking about payroll.

That gap between how customers describe their problem and how businesses describe their solution — it shows up everywhere. And search data is one of the few places you can actually see it.

Where search insights usually stop

In most companies, search data lives with whoever manages the ads. They use it to decide what keywords to target, how much to bid, what copy to test.

That's useful. But it's also a waste of everything else the data could tell you.

Every search query is someone describing a problem in their own words, at the exact moment they're thinking about it. That's customer research that updates in real time. Except most companies treat it like a targeting spreadsheet.

Google even hints at this broader value — the Search Terms report is described as a way to discover what customers actually searched for and to use that insight to improve messaging and landing pages, not just targeting (Google).

What happens when you use it elsewhere

Rewriting a homepage. I worked with a company that sold project management software. Their homepage led with features: "Task assignment, timeline views, collaboration tools."

Their search data showed people were actually looking for things like: "how to stop projects going off track," "why do we keep missing deadlines," "team can't agree on priorities."

We rewrote the homepage to speak to those problems first. Features came later, after we'd shown we understood what was frustrating them. (I've seen this pattern work consistently across B2B software companies — leading with the problem, not the product.)

Fixing an email sequence. A SaaS company had a welcome email that opened with something generic like: "Welcome to [Product]! Here's how to get started with your dashboard."

Their search data showed that most people found them by searching for very specific problems — things like getting fewer status update requests from clients, or having better project visibility without micromanaging.

The new email opened by naming those problems directly. Open rates improved noticeably, and — more surprisingly — people actually replied to a previously one-way email.

Training a sales team. A B2B services company noticed their sales calls were getting stuck. Prospects would ask questions the team didn't have good answers for.

When we looked at search data, we could see the types of questions people were asking before they booked a call: whether it was worth it for small teams, what happened if it didn't work, how long until they'd see results.

The sales team started addressing those concerns proactively in the first five minutes. Close rates improved. (I can't share specific numbers, but the change was significant enough that the team kept the approach.)

The thing nobody does

Search data is available to almost every business. Most of them never share it beyond the ads team.

Content marketers write blog posts based on what they think people want to know. Product marketers write messaging based on internal positioning documents. Sales teams wing it based on what they've heard in calls.

Meanwhile, there's a live stream of exactly how customers describe their problems, and almost nobody looks at it. The best-performing content uses language directly sourced from customers — and search queries are one of the richest sources available (Content Marketing Institute, 2024).

How to actually start

Don't build a process. Just start sharing.

Pull up your search term report. Find the ten most surprising or specific queries — the ones that reveal something about how customers think. Send them to your content team. Share them in a marketing meeting. Show them to your sales team.

You don't need a dashboard or a system. You just need to get the insights out of the silo and into conversations where they can be useful.

Search data is customer research. Treat it that way.