The First Thing I Do When Someone Asks Me to Fix Their Ads

A friend reached out last week about his heating installation business.

"Are we doing the right thing on Google Ads? How do we generate more leads?"

Solid question. And my first instinct should be to log into his ad account and start poking around.

But I don't.

I haven't opened his account yet. And I won't—not until I've done something else first.

Here's what most people do when they want to "fix" their ads:

They open the account. Stare at the metrics. Try to figure out what's wrong based on... what, exactly?

Their gut? Some best practices they read about? A course they took three years ago?

They're auditing in a vacuum.

They have no idea what "good" actually looks like in this specific market, for this specific service, at this specific price point.

So they make changes based on theory. Or they compare against benchmarks that have nothing to do with heating installation in their friend's city.

This is backwards.

Here's what I've learned after years of running paid ads for service businesses:

Someone in your market is already winning.

They've already tested the keywords. They've already figured out the landing page. They've already dialed in the offer.

And they're not hiding it. They can't hide it. It's all right there if you know where to look.

My job isn't to be a genius. My job is to find what's already working and reverse-engineer it.

Before I touch his account, I pull up SimilarWeb, SEMRush, or SpyFu.

I'm looking for heating companies pulling significant monthly traffic with heavy paid search spend. Not a one-month spike—consistent spend, month after month.

Why does consistency matter?

Because nobody burns cash on ads that don't convert. If a company is spending $20K/month on Google Ads for six months straight, they're making money.

The ads are working. Full stop.

These are the companies I want to study.

Now I build a swipe file. I'm looking at everything:

Keywords: What exactly are they bidding on? Branded terms? Service-specific long-tail? Emergency keywords? What are they not bidding on that you'd expect?

Landing pages: What does the page look like? What's above the fold? How many trust signals? Reviews, certifications, guarantees—what's present on every top competitor's page?

The offer: Are they leading with price? Financing? Speed of service? Free estimates? What's the actual hook?

Facebook Ad Library: What creative are they running? What angles are they testing? How long have specific ads been live? (Longer = working.)

I'm not guessing anymore. I'm building a map of what the market has already validated.

Only after I have this data do I log in.

And suddenly the audit becomes obvious.

Maybe he's bidding on keywords the winners aren't touching—probably for a reason.

Maybe his landing page is missing trust signals that every top competitor has.

Maybe there's a channel the winners are dominating that he's ignoring completely.

I'm not auditing against theory. I'm auditing against reality.

Here's what most people miss:

Best practices are averages. They're what works across all industries, all markets, all price points.

But you're not competing against "all industries." You're competing against the three or four companies actually winning in your specific space.

Their practices are the ones that matter. And they might look nothing like the generic advice.

The heating company dominating your friend's market might be doing something that would be "wrong" according to every Google Ads course out there. But it's working—because it's working for that market.

I use this approach for every channel now.

Meta ads? I check Facebook Ad Library before I write a single piece of creative.

Landing pages? I screenshot the top five competitors before I design anything.

Email sequences? I sign up for every competitor's list before I write copy.

The market has already done millions of dollars worth of testing. The least I can do is look at the answers before I start guessing.

I won't pretend this is fast. Building a proper competitive swipe file takes a few hours.

But those hours save weeks of "testing" that's really just expensive guessing.

And honestly? It makes the work more interesting. You start seeing patterns. You notice what every winner is doing that the losers aren't. You develop real intuition—not theoretical knowledge, but market-specific understanding.

That's the edge.

The market already figured out what works.

Your job is to find it.