The report arrives on the third of the month. Leads are up, cost per lead is down, and there's a chart with a green arrow. Meanwhile your front desk says the phone was quiet, and the sales team can't name three new customers from ads.

Both things can be true at once. Ad platforms, analytics tools, and your CRM count different things in different ways. This guide shows you how to find out which numbers to trust, without accusing anyone of anything.

Why do agency reports and your CRM disagree?

Usually for one of four reasons:

  • Each ad platform counts its own credit. If a customer clicked a Google ad, then saw a Meta ad, both platforms may report that same customer as their conversion. Add the two reports together and you have counted one person twice.
  • "Lead" means different things. An agency may count every form submission, newsletter signup, or click to call. Your sales team counts people who were actually worth calling back.
  • Some conversions are modeled. Platforms estimate conversions they can't track directly. Those estimates are useful for steering campaigns, but they are not names in your CRM.
  • Tracking breaks quietly. A website update can double-fire a conversion tag or stop it firing at all. Nobody notices until the numbers look strange.

None of that requires bad intent. It does mean a report can be accurate by its own rules and still not tell you what you need to know.

What should you ask your agency?

Six questions, asked plainly, will tell you most of what you need:

  1. What exactly counts as a lead in this report? Get the definition in writing. Form fills only? Calls over a certain length? Booked appointments?
  2. Where does each number come from? The ad platform, GA4, or your CRM? Numbers from the ad platform describe the ad platform's view of the world.
  3. Are any conversions counted by more than one platform? Ask how they avoid double counting when reporting a combined total.
  4. How many of last month's leads became customers? If they can't answer, the reporting stops at the form, and that's the gap to close.
  5. When was conversion tracking last tested? A test is a real form submission or call, checked end to end, not a glance at a dashboard.
  6. What would you cut if the budget dropped by a third? A clear answer shows they know which channels pull their weight.

A good agency will welcome these questions. Many will already have answers.

How do you check the numbers yourself in an afternoon?

You don't need special software for a first pass:

  1. Export last month's leads from your CRM, booking system, or inbox, with the date and the source if you capture it.
  2. Put the agency's reported lead count for the same month next to it.
  3. Count how many of your real leads trace back to paid channels. A "How did you hear about us?" field helps, even if it's imperfect.
  4. Count how many of those became paying customers.
  5. Divide that month's ad spend by the number of paying customers. That is your actual cost per customer, the number that matters to your bank account.

If the agency's leads and your leads are within a reasonable distance, good. If the report shows three times what your CRM shows, the next conversation is about definitions and tracking.

What does good agency reporting look like?

  • Leads are defined in writing, and the definition matches what your sales team means.
  • The report shows platform numbers and CRM numbers side by side, and labels which is which.
  • Combined totals are deduplicated, or clearly marked as not deduplicated.
  • Cost per customer appears alongside cost per lead.
  • Tracking gets tested after every website change.

Where does this come from?

Raquel spent close to 20 years in analytics. At a marketing technology company, she was the analytics lead between enterprise clients and a data science team. Part of that work was attribution: connecting orphaned sales records back to the customers and campaigns that produced them. That let clients see which campaigns worked. She also led GA4 tracking governance through a migration in an in-house eCommerce role.

When is it worth a proper attribution review?

When a wrong answer on marketing spend gets expensive. When you pay for more than one channel. Or when you and your agency can't agree on the numbers.

Raquel's marketing attribution review costs $4,000 to $8,000 and takes 3 to 5 weeks. It matches spend to booked revenue across channels and leaves you with definitions you and your agency can both use. If tracking is the problem, the GA4 tracking audit is $2,500 flat.