Tips & Tricks

Facebook Ad Agency Reporting: What Insurance Agents Should Actually See

9 min read · October 13, 2026

Every agent who has hired an ad agency has gotten the report. It is a nicely designed PDF or a dashboard link. It has a chart going up and to the right. It says your ads reached tens of thousands of people, earned a healthy click-through rate, and generated a pile of “results.” And when you finish reading it, you still cannot answer the only question you care about: did the money I spent come back with profit?

That gap is not always the agency hiding something. A lot of the time it is just an agency reporting what it can see from inside Meta, which is ad activity, not insurance sales. But whether the gap is innocent or not, it is your money, and you are the one who has to close it. This post is about what a useful report looks like, which numbers are the scoreboard and which are just diagnostics, and the red flags that tell you a report is built to impress rather than inform.

The one question a report has to answer

You are not running ads to collect impressions. You are running them to write business. So a report is only useful if it walks the whole chain, from dollars in to premium out, for the same date range:

  1. How much did I spend?
  2. How many real leads reached me?
  3. How many of those did I talk to?
  4. How many did I sell, and for how much premium?
  5. What was left after the ad spend?

Anything that does not help answer one of those five questions is supporting detail. Some of it is useful supporting detail for whoever is building the ads. None of it is the scoreboard.

The scoreboard numbers

Here is the short list I would want on the first screen of any report, and who actually has the data for each one. That second column matters more than it looks.

NumberWhere it comes fromWhy it matters
Ad spendYour ad account and your card statementThe cost side of everything else. Should match to the dollar.
LeadsThe form, and what reached your phone or CRMReal submissions that got to you, not “results” in general
Cost per leadSpend divided by leadsUseful trend line, but only half the story
Contact rateYou, from your dispositionsSeparates bad leads from leads nobody reached in time
Sales and premiumYou, from your apps and carrier portalsThe point of the whole exercise
Cost per saleSpend divided by salesWhat a policy actually costs you to acquire
Profit after spendYour commission minus ad spendThe only number that tells you to keep going or stop

Notice the pattern. The top three rows the agency can pull straight out of the ad account. The bottom four depend on you. That is the honest reason most agency reports stop at cost per lead: the rest of the chain lives on your phone and in your carrier portals, and nobody told the agency.

Cost per lead on its own can mislead you in both directions. A campaign with a cheaper lead can lose money if those leads never pick up. A campaign with a pricier lead can be your best one if it closes. I went deeper on reading each metric in the guide to Facebook ad metrics for life insurance agents.

The diagnostic numbers, and why they are not the scoreboard

Ads Manager is full of metrics, and a good media buyer uses a lot of them. They are how you figure out why cost per lead moved. They are not how you decide whether the ads are working.

MetricWhat it tells the ad builderWhy it is not your scoreboard
Impressions and reachHow many times and to how many people the ads showedSpend more and these go up. That is not success.
CPMWhat the auction is charging to show adsExplains cost swings, does not measure results
Click-through rateWhether the creative stops the scrollClicks are not leads, and leads are not sales
FrequencyHow often the same person sees the adAn early warning for tired creative, nothing more
EngagementLikes, comments, sharesComments can be a moderation job, not a win

A report that leads with these is not automatically dishonest. It might just be a generic template the agency uses for every client, from restaurants to roofers. But if these are the only numbers you ever see, the report is measuring activity, and you are paying for outcomes.

Check the numbers against things you control

A report is a claim. You should be able to verify it without asking the agency for anything. Three checks cover most of it.

Spend against your card

If the ad account is yours and your card is on billing, Meta charges you directly and the receipts live in your billing settings. The spend in the report should match those charges for the same dates. If spend runs through the agency's card and gets rebilled to you, ask to see the account's billing summary. You are entitled to know what was spent on your behalf.

Leads against what reached you

Count the leads that actually landed on your phone or in your CRM for the month. Compare that to the lead number in the report. They will rarely match perfectly because of duplicates, tests, and junk submissions, but they should be close, and the agency should be able to explain the difference. A report that shows a lot more leads than you received is either counting something other than leads, or your lead routing is broken and nobody noticed. Both are worth finding out about this week.

The ad account itself

The simplest check is logging into Ads Manager and looking. That only works if you have access to an account in your own name, which is why I keep coming back to who should own the ad account. If you cannot open the account your money is being spent in, every report is something you take on faith.

Closing the sales gap is partly your job

Here is the uncomfortable part. Most agents who complain that their agency only reports cost per lead have never told the agency what sold. The agency cannot see your carrier portal. It does not know the lead from Tuesday turned into a policy on Thursday unless someone records it.

Recording sales does two jobs. The first is the report: once sales are logged against the leads that produced them, cost per sale and profit stop being guesses. The second is the ads themselves. When a sale is sent back to Meta as a conversion event, the system gets a signal about which kind of person actually buys, not just which kind fills out forms. The mechanics are in the pixel and Conversions API guide.

The other thing only you can supply is what happened on the phone. If you are calling leads hours later or giving up after one attempt, cost per sale will look bad no matter how good the ads are. A fair report separates “the leads were bad” from “the leads were never worked,” and it can only do that if your dispositions are honest.

How often to look, and how to read small numbers

Checking daily is fine for one thing: catching something broken. If spend is running and no leads are arriving, you want to know today, not at the end of the month. Beyond that, daily numbers are noise.

Judge results over a month or more. New and heavily edited ad sets go through Meta's learning phase, and an agent budget produces a modest number of leads and a smaller number of sales. Small numbers swing hard. To use made-up round numbers: if you spend the same amount two months in a row and sell three policies one month and four the next, your cost per sale drops by a quarter without anything about the ads changing. That is why one bad week is not a verdict, and one great week is not a trend.

The fix is to look at rolling totals. Running spend, running leads, running sales, running profit. Those lines settle down over time and tell you the truth. The weekly snapshot mostly tells you about luck.

Red flags in an agency report

  • Only platform metrics. Reach, impressions, CTR, and engagement, month after month, with no lead count you can check and no room for sales.
  • “Results” instead of leads. Meta's results column counts whatever the campaign is optimized for. Make sure the number is completed lead forms, not clicks or page views.
  • Lead counts that do not match your CRM. A gap with no explanation means something is miscounted or misrouted.
  • Spend you cannot reconcile. Especially when spend is billed through the agency and you have no view of the account.
  • Definitions that change. Cost per lead calculated one way in March and another way in April makes the trend line meaningless.
  • Reports that only show up on time when things are good. The bad months are the ones you most need to see clearly.
  • Blame without data. “Your follow-up is the problem” might be true, but a good agency shows you the contact rate that proves it.

If you are screening a new agency, ask to see a sample report before you sign. It tells you more than the sales call will. The rest of the screening list is in the questions to ask a Facebook ad agency.

A simple report is fine if…

Not every agent needs a dashboard. Be honest about where you are.

  • You already track everything yourself. If you keep a clean spreadsheet of leads, contacts, and sales and you can see the ad account, a plain spend-and-leads summary from the agency is enough. You are closing the gap on your end.
  • You are in the first few weeks. Early on there is not enough data for cost per sale to mean much. Watch spend, leads, and whether leads reach you fast. Judge the rest later.
  • The agency answers your questions straight. A modest report from someone who will pull up the account and walk you through it beats a glossy report from someone you cannot reach.

How we handle it

This is the reason the FEXmagnet portal exists. FEXads, our ad-management service, runs Facebook and Instagram campaigns for life insurance agents, and a real person hand-builds every one inside the agent's own ad account. The portal is metrics-only on purpose: spend, leads, cost per lead, sales, and profit, updated nightly. Logging a sale takes one click, and that same click feeds your pixel, so the report and the ads get better from the same action.

Because the account is yours, you never have to take the portal's word for it. Open Ads Manager and check. You keep the page, the pixel, the ad account, and every lead. That is the difference between renting a lead flow and owning one: when you own it, the numbers are yours too.

Honest verdict: A useful agency report walks the whole chain for one date range: spend, leads, contacts, sales, premium, cost per sale, and profit. Impressions, CTR, and CPM are tools for the person building the ads, not the scoreboard. Verify spend against your own billing and leads against what reached your phone, and keep the ad account in your name so you can check anything yourself. Then do your part: log every sale. Without that, no agency on earth can tell you whether the ads made money.

Frequently asked questions

What should a Facebook ad agency report to me?

Ad spend, leads, cost per lead, sales, premium written, cost per sale, and profit after spend, all for the same date range. Spend should match your own billing and leads should match what reached you. Platform metrics like reach and CTR are supporting detail.

Why does my agency not know how many policies I sold?

Sales happen on your phone and in carrier portals, not inside Meta. The agency only knows what sold if you record it somewhere it can see. Until then, cost per lead is the best it can report.

How often should I review my ad results?

Glance often enough to catch something broken, like spend with no leads. Judge results over a month or more, using running totals. Small numbers and the learning phase make weekly snapshots noisy.

What are the red flags in an agency report?

Only platform metrics, lead counts that do not match your CRM, spend you cannot reconcile, no access to the ad account, definitions that change month to month, and reports that run late when results are bad.

Want leads without buying leads?

Buying leads is renting. FEXads runs Facebook ad campaigns for life insurance agents inside your own ad account, so the page, the pixel, and every lead stay yours. A real person builds every campaign and you approve every ad. $200 to set up, then $500 a month flat starting at your first sale. No contract.

See how FEXads works