Before you hire an agency to run your ads, ask to see the numbers from its own ads. An agency that buys ads for a living should be able to show you what its own money bought.

That is the fastest way to vet a marketing agency, and few owners ask. So we will go first. Below are the numbers from our own Facebook ads, line by line. You will see what we spent, what it brought in, the best ad, the ads we cut and the mistake that cost us.

We sell to portrait studio owners, so these ads were aimed at them. Your business is different. The point is not our cost per lead. The point is that you can see all of it.

What most owners ask for instead

Most owners vet an agency by its client wins. They see a case study, a screenshot of a great month or a story about a business like theirs. Those are worth seeing. They are also picked by the agency.

You see the account that went well. You do not see the ones that did not. You see the cost per lead, the price paid for each form fill. You rarely see what those leads turned into. And you almost never see the dates. A great week looks the same as a great quarter in a screenshot.

Why that costs you

A client win tells you the agency did well at least once. From the outside, you cannot tell if it was skill or luck. So you hire on the best story. Then you find out the real average on your own money.

There is a second cost. An agency that only shows wins is telling you how it reports. If the pitch skips the misses, the weekly report will too. You will hear about lead volume and booked calls. You will hear less about which ads failed and what they cost. As we covered in why your Meta cost per lead can mislead you, a cheap lead number can hide an expensive result.

This one is personal for us. Before Dynamite Growth, Gian Gomez, our founder, ran a solar company doing over $20 million a year. He went from zero leads on Facebook to over $500,000 a month from one advertising channel. Along the way, he paid for marketing that handed over activity instead of customers. That is why we are accountable to paying customers, not lead counts.

What we do instead: show our own

Our team has audited and worked with more than 40 portrait studios and their Meta accounts. Those numbers belong to those studios. Our own ads are the numbers we can show in full, misses included.

So here are our ads to portrait studio owners. These are our numbers, on our money, selling our service.

Run 1 went from 8/3 to 9/1. We spent $3,666. It brought in 30 Meta leads at $122 per lead, and 5 studios signed.

Run 2 went from 10/6 to 11/3. We spent $2,690. It brought in 27 leads at $100 per lead, and 6 studios signed.

Together, that is $6,356 spent, 57 leads, $112 per lead and 11 studios signed. It works out to about $578 of ad spend per signed studio.

That last number leaves things out, and you should know what. It does not count software or contractors. It is ad spend only. A full cost per signed studio would be higher.

A snapshot from inside the second run

Totals at the end of a run hide what happened in the middle. So here is a snapshot we took in mid-October. We had relaunched on 10/6 at $100 a day. By that read we had spent $1,077.54 for 16 leads, or $67.35 per lead. The last 7 days of that window ran $82.88 per lead.

Further down the pipeline, 11 of 13 done-for-you opportunities had reached a booked stage. An opportunity here is a studio owner who asked about having our team run it for them. And 6 of 8 first calls showed up. That is a small sample, so read it as a snapshot, not a rate.

By the end of run 2, the full-run number was $100 per lead. The early read was cheaper than the finish. That is one more reason to ask for full date ranges, not the best week.

The best ad, the cuts and the mistake

Across both runs, our best ad ran at $39.80 per lead and produced 2 closes. Six other ads ran at 2x or more of the best ad's cost per lead. They were cut.

That is the 2x rule we run on every account. Every live ad goes on one table next to the best ad. An ad that has spent twice the best ad's cost per lead, and keeps paying 2x or worse, comes off.

We did not always follow it. Over three days, from 10/9 to 10/12, our best ad ran about $26 per lead. Another ran $199 per lead, which is 7.6x. We kept the expensive one on twice because it had won a deal. In those 3 days it burned $253 and brought in no lead. That is why a closed deal never saves an ad on our accounts now.

What we would change

We would change three things.

First, we would hold the line with no exceptions. The money was small. But it was lost to a story instead of the table, and stories are how small losses repeat.

Second, we would track the full cost from day one. Ad spend per signed studio is useful. Ad spend plus software plus contractors is the number a business actually lives on.

Third, we would wait for bigger samples before reading anything past the lead. Six shows out of eight first calls is good to see. It does not prove anything yet.

Where this stance has limits

Asking for an agency's own numbers is a good test. It is not a perfect one.

Our offer is not yours. We sell a service to studio owners. You might sell a roof, a cleaning or a family session. Our cost per lead says nothing about what yours will be. What it shows is how we read numbers and whether we share the bad ones.

The samples are small. Two runs, 57 leads and 11 signed studios is real money and real results. It is not a big data set, and we would rather say so than let it pass as one.

Some good agencies do not advertise for themselves. They grow on referrals. That alone is not a red flag. In that case, ask for the full numbers from one client account instead: spend, leads, bookings and sales, with dates, including the ads that were cut.

Own numbers can be cherry-picked too. Ask for the full window, from the first dollar to the last.

What to ask any agency this week

Before your next agency call, write these down and ask them in order.

  1. What did you spend on your own ads, and over what dates?
  2. How many leads did that bring in, and how many became paying customers?
  3. What was your best ad, and what did it cost per lead?
  4. Which ads did you cut, and what rule did you use?
  5. What costs are left out of your numbers?
  6. What mistake did you make on your own account, and what did it cost?
  7. What do you report to clients each week?

Good answers have dates, dollar amounts and at least one miss. Vague answers are an answer too.

On the last question, here is ours. The wedge is closed revenue, not booked calls. We report closed revenue every week, and booked calls and lead volume are context.

FAQ

How do you vet a marketing agency before you hire it?

Ask for numbers from the agency's own ads, with dates, spend, leads and paying customers. Then ask which ads it cut, what rule it used and what costs it left out. An agency that answers with real figures and at least one mistake is showing you how it will report to you.

What if an agency does not run ads for itself?

That alone is not a red flag. Ask for the full numbers from one client account instead. You want spend, leads, bookings and sales, with dates, including the ads that were cut.

What should a marketing agency report every week?

We report closed revenue every week, with booked calls and lead volume as context. We never report "leads this week vs last week" on its own. A lead count says nothing about cost or bookings.

Why does Dynamite Growth publish its own ad numbers?

Because it is the test we would want an agency to pass. Client numbers belong to clients. Our own ads are ours to show in full, including the ad we kept on too long.

If you want to put these questions to our team directly, start with a discovery call.