The Ad Spend Trap

There's a specific kind of business owner I see regularly. Running Google Ads for two years. Spending $1,500 to $3,000 a month. When I ask them what their cost per lead is, they don't know.

Their agency sends a report each month. Impressions, clicks, click-through rate. Professional and detailed. It just doesn't mention leads, or what each one cost.

So we do the math on the spot. Take last month's spend, divide it by the number of people who actually called or filled out a form. Sometimes the number is fine and they're relieved. More often it's a number they didn't expect, and the room gets quiet.

That gap — between what the report measures and what the business needs to know — is the trap. It isn't that ads don't work. It's that spending is easy to measure and results are easy to avoid measuring.

What ads actually do (and what they can't)

Paid ads buy attention. That's the whole product. You give Google or Meta money and they put you in front of someone who otherwise wouldn't have seen you.

What ads can't do is make someone want to hire you. That happens after the click, on your website, with your reviews, your pricing, your photos, and how fast someone answers the phone. Ads deliver traffic to whatever you already have.

This is why the businesses that get burned on ads are usually the ones who bought them to fix a problem ads don't fix. If your site converts one visitor in a hundred, running ads means you're now paying full price for ninety-nine people who leave. The ad platform will happily take that money for years.

There's a version of this that's easy to test. Look at the last 90 days: how many people came to your site from anywhere, and how many contacted you? If that ratio is bad organically, it will be bad from ads, except now every visitor has a price tag on it. Fix the conversion problem first and the same ad budget produces more. Every time.

Ads are an accelerant. They make whatever is already happening happen faster, in both directions.

How much are your ads actually returning?

The percentage-of-spend model and why it misaligns incentives

Most agencies managing paid ads charge a percentage of what you spend. Usually 10 to 20%. It's the standard, and on the surface it sounds fair: they make more when you invest more.

Sit with that for a second. Your agency's revenue goes up when your spend goes up. It does not go up when your cost per lead goes down.

I'm not saying agencies are deliberately wasting your money. Most aren't. But when the recommendation is always "we should increase budget," and never "we should pause these three campaigns and cut spend by 20% because they're not converting," the incentive is doing some of the talking.

You can see it in the reporting. Percentage-of-spend reports emphasize volume: impressions served, clicks delivered, reach expanded. Those numbers all go up when spend goes up. A report built around cost per lead and cost per customer would sometimes have to say the honest thing, which is that this month was worse than last month and here's what we're changing.

The fix isn't necessarily firing your agency. It's changing what you ask for. Tell them you want cost per lead and cost per acquired customer on the first page of every report, tracked month over month. A good partner will already have it. A bad one will explain why it's complicated.

It usually isn't complicated. It's spend divided by leads.

Quality Score — the hidden tax most businesses are paying

Google scores every keyword in your account from 1 to 10 based on three things: how relevant your ad is to the search, how relevant your landing page is to the ad, and how often people click your ad compared to expected.

That score directly changes what you pay. Two businesses bidding on the same keyword do not pay the same price. The one with a Quality Score of 8 pays meaningfully less per click than the one sitting at 4. Same keyword, same position, different bill.

This is the closest thing to free money in paid advertising, and most small businesses have never looked at it. It's not on the default dashboard. You have to add the column.

The usual cause of a bad score is structural. Someone built one ad group with forty keywords in it and wrote two generic ads to cover all of them. Nothing in that ad matches most of the searches, so the score sinks and the cost goes up on everything.

The fix is tedious, not difficult. Break keywords into tight groups by intent. Write ads that use the actual words people searched. Send the click to a page about that specific service instead of your homepage. Someone searching "emergency water heater repair" should land on a page about emergency water heater repair, not a general plumbing homepage where they have to find it themselves.

Businesses that do that cleanup usually see cost per click drop within a few weeks without touching their budget. Same money, more clicks, more leads.

Google Ads vs. Meta Ads — when to use which

These platforms do different jobs and get treated like they're interchangeable.

Google captures demand that already exists. Someone types "AC repair near me" at 8pm in July because their AC is broken. They are not browsing. They're buying, and they're going to buy from someone in the next hour. That intent is why Google clicks cost what they do.

Meta creates demand that wasn't there. Nobody opens Instagram looking for a med spa. But a good video ad in front of the right local audience can make someone think about something they'd been putting off. Cheaper clicks, colder audience, longer path to a sale.

The practical rule for most local service businesses: if you sell something people search for when they need it — plumbing, HVAC, legal, urgent repair, emergency anything — start with Google. If you sell something people choose rather than need — aesthetics, fitness, restaurants, home upgrades, anything discretionary — Meta usually wins.

If you're going to run both, don't split a small budget across them. A $1,000 budget divided between two platforms is two campaigns without enough data to optimize either. Pick the one that fits how people buy what you sell, get it working, then expand.

One more thing on Meta in 2026: video is not optional anymore. We're seeing static creative run 40 to 60% higher on cost per lead for local services, with everything else held constant. It doesn't have to be produced. Phone footage of actual work, actual staff, actual results outperforms polished stock most of the time.

The 30-minute check that keeps you out of the trap

You don't need to become an ads expert. You need to look at five things once a month.

What good actually looks like

A healthy paid account for a local business is unglamorous. A handful of tightly-themed ad groups. Ads that repeat the words people searched. Landing pages that match the ad and ask for one thing. Call tracking that works. A cost per lead you can recite from memory.

It's not a bigger budget. I've seen $1,200 a month outperform $4,000 a month in the same market, same industry, because one account was built with intent and the other was built to spend.

The businesses that get hurt by ads are almost never the ones spending too little. They're the ones spending steadily without ever asking the question underneath the spend.

Before you spend another dollar, make sure you can answer: what does a lead from this campaign cost me?

If you can't answer that today, that's the work. Not more budget. Not a new platform. Just the number.