Before You Pay for Ads, Can You Prove They Work?

I talked to a business owner last month who wanted to run Facebook ads. He had a budget ready, some ideas for creative, and enthusiasm to "get in front of more people."

"Great," I said. "Walk me through what happens when someone clicks your ad."

He paused. "They go to my website."

"Then what?"

"They... fill out the form, I guess."

"And how do you know that form submission came from the Facebook ad versus someone who found you on Google?"

Longer pause.

The honest answer was: he didn't know. Couldn't know. Because nothing was set up to track it.

The Order of Operations Problem

Most small businesses approach paid advertising backwards.

They decide to run ads. They create the ads. They turn them on. Then—maybe—they start thinking about measurement.

This is exactly wrong. The tracking infrastructure needs to exist before the first dollar goes out. Not after. Not "when we have time." Before.

Here's why: once ads are running, you're spending money. If you can't prove which leads came from those ads, you have no way to know if the spend is working. You're flying blind—and paying for the privilege.

According to the Digital Marketing Institute, companies without proper attribution commonly misallocate up to 30% of their marketing budget. That's not a small leak. That's a third of your spend going to the wrong places because you can't see what's working.

What "Proving It Works" Actually Means

Proving an ad works isn't about clicks or impressions or reach. Those are vanity metrics. They feel good in a dashboard but don't pay your bills.

Proving an ad works means answering one question: How many customers did this create?

To answer that question, you need a chain of data:

1. Someone saw or clicked your ad. This is tracked by the ad platform.
2. They took an action. Visited your site, filled out a form, made a call.
3. That action is tagged with the source. You know it came from the Facebook ad, not organic search.
4. The lead entered your CRM. Now it's a real prospect, tracked in your system.
5. The lead converted (or didn't). You know if they became a customer and how much revenue they generated.

If any link in that chain is broken, you can't prove the ad worked. You might believe it worked. You might hope it worked. But you can't prove it.

The Minimum Setup Before Spending

Before you turn on any paid advertising, you need four things in place:

Google Analytics or equivalent. Something that tracks what happens on your website. GA4 is free and works for most small businesses. It needs to be installed correctly, with conversion events set up for the actions that matter (form submissions, phone clicks, booking requests).

UTM parameters on all ad links. UTM tags are those extra bits at the end of a URL that tell analytics where traffic came from. Every ad should have them. Every single one. "utm_source=facebook&utm_medium=paid&utm_campaign=spring-ac-promo" tells you exactly which campaign drove that click.

Call tracking (if you get phone leads). A lot of service businesses generate leads through phone calls. If someone clicks your ad and then calls the number on your website, how do you know that call came from the ad? You don't—unless you're using a call tracking service that assigns different numbers to different sources.

CRM integration. Your website forms should feed directly into your CRM with source information attached. When someone fills out a lead form, the CRM should record not just their name and contact info, but where they came from. That way, when they become a customer, you can trace the revenue back to the original campaign.

This isn't fancy marketing technology. This is the baseline. If you're spending money on ads without this infrastructure, you're guessing.

The "We'll Figure It Out Later" Trap

I know what some of you are thinking: "That sounds like a lot of setup. Can't I just start running ads and add tracking later?"

You can. People do it all the time. Here's what happens:

You run ads for three months. Your business gets busier—or maybe it doesn't, you're not sure. You finally set up tracking. Now you can see results going forward, but you have no data on those first three months.

Was the increased business from ads? From seasonality? From word of mouth? From that one Google review that went viral? You have no idea. The money you spent during that period is unaccounted for forever.

This is how businesses convince themselves that "ads don't work" or "Facebook is a waste." They ran campaigns they couldn't measure, saw ambiguous results, and concluded the channel was the problem. The real problem was the setup.

Start Small and Prove the Model

Here's what I recommend for businesses that haven't done paid advertising before:

Start with a small budget—$500 to $1,000 per month. Make sure all tracking is in place before you spend a dime. Run the campaign for 60-90 days.

At the end of that period, you should be able to say: "We spent $X. It generated Y leads. Z of those became customers. They generated $W in revenue."

If the math works—if the revenue exceeds the spend by enough to make it worthwhile—you scale. If it doesn't, you either optimize or cut.

That's the whole game. Test small, prove the model, then invest. Not the other way around.

The Math Has to Be Visible

I push hard on this with clients because I've seen too many small businesses burn through marketing budgets with nothing to show for it.

The agency says "awareness is up." Great—awareness doesn't show up in your bank account.

The platform says "your cost per click is below industry average." Great—clicks don't pay your mortgage.

The only thing that matters is: Did this spending create profitable customers?

If you can't see that math clearly—if you can't trace a customer back to the specific ad that generated them—then you don't have a marketing system. You have a marketing expense. And expenses without returns are just losses dressed up in nicer language.

Set up the tracking first. Prove the model works. Then spend.

In that order. Always in that order.