---
title: "I Tore Apart a Typical Meta Agency Report. Its Biggest Numbers Miss the Point"
description: "A page-by-page teardown of a typical Meta agency report: what its traffic and ROAS figures show, what they leave out, and the two revenue lines to demand."
url: "https://groas.com/post/i-tore-apart-a-typical-meta-agency-monthly-report-14-numbers"
image: "https://pub-87da24ecbbfc4c3bad6875f3aa013712.r2.dev/generated-images/a0265484-9c33-4f79-908f-6092d2b19614.png"
published: "2026-11-03T06:26:05.525Z"
modified: "2026-11-03T06:26:05.593Z"
---

November 3, 2026 · 11 min read

# I Tore Apart a Typical Meta Agency Report. Its Biggest Numbers Miss the Point

[Alexander PerelmanHead Of Product @ groas](https://groas.com/author/alexander-perelman)[LinkedIn](https://www.linkedin.com/in/alexander-433793253/)

![Cartoon of a shop owner holding her ad agency's report up to a lamp like a suspect banknote, while the beaming agency rep waits for praise.](https://pub-87da24ecbbfc4c3bad6875f3aa013712.r2.dev/generated-images/a0265484-9c33-4f79-908f-6092d2b19614.png)

In this article

1. [Page 1: big numbers, small answers](#page-1-big-numbers-small-answers)
2. [Page 2: Meta grades its own homework](#page-2-meta-grades-its-own-homework)
3. [Page 3: the paragraph that says nothing](#page-3-the-paragraph-that-says-nothing)
4. [The two lines the report leaves out](#the-two-lines-the-report-leaves-out)
5. [The budget decision, without the green arrows](#the-budget-decision-without-the-green-arrows)
6. [The one-page report I would demand instead](#the-one-page-report-i-would-demand-instead)

I tore apart a composite built from a dozen real Meta agency monthly reports. I chose it because, for many owners paying someone to run Facebook and Instagram ads, **the PDF is the product they actually see**. It opens with [reach, spend and Website Purchase ROAS, then splits performance by platform and creative](https://www.adbeacon.com/attribution-tool-vs-ads-manager-black-friday/). The targeting and creative may be doing good work. This report is supposed to show it.

My claim is flat: **the standard report is built to look busy.** It gives you a stack of numbers and a paragraph of insights, yet often leaves you unable to answer whether the spend produced revenue your business can account for. Two lines matter more than the rest: reconciled revenue and a return measured against your actual costs. Here is the report in the order an owner reads it, with a verdict on each piece before I turn the page.

## Page 1: big numbers, small answers

**Lines 1 and 2: Reach 182,400. Impressions 311,000.** The report puts both in large type because they are large numbers, and large numbers calm clients. I used to put them first too. I was wrong. Reach counts people who could have scrolled past. Impressions count times an ad appeared on a screen. Neither requires anyone to stop, read, click or buy.

**Verdict:** Proof of delivery, not proof of value. Keep both in an appendix if you like history. Do not let them open the budget decision.

**Line 3: CPM $11.42, down 18% month over month.** There is usually a green down arrow beside it. The implication is clear: cheaper eyeballs, better buying. But Meta can find cheaper impressions by drifting toward broader audiences or lower-intent inventory. A falling CPM can mean [cheaper, lower-quality inventory rather than better performance](https://contentstudio.io/blog/meta-ads-benchmarks). For context, [2026 all-industry medians sit around $14 CPM, up about 20% year over year](https://contentstudio.io/blog/meta-ads-benchmarks). So $11.42 looks cheap against that benchmark. Cheap only helps if the people who saw the ad did something.

**Verdict:** CPM tells you what exposure cost. Until you connect it to sales, the green arrow is decoration.

**Lines 4 to 6: Link CTR 1.84%, up from 1.61%. Link clicks 5,722. CPC $0.62.** This is where the report starts to sound like performance. It is still traffic accounting. A [link click is a tap](https://meetadmiral.com/blog/meta-link-clicks-vs-landing-page-views). A landing page view requires the destination to load and the pixel to fire. The gap can include accidental taps and pages that never finish loading. You paid for the click either way.

I check that gap before I admire the CTR. Say you spent $3,550 for 5,722 link clicks but recorded only 3,910 landing page views. Roughly 32% of those clicks did not register as arrivals. Your cost per click is $0.62; your cost per recorded landing page view is about $0.91. [Optimising for link clicks trains delivery toward cheap taps](https://meetadmiral.com/blog/meta-link-clicks-vs-landing-page-views), which makes that distinction worth watching. Against [2026 medians of 1.5% to 2.2% CTR and $0.78 traffic CPC](https://contentstudio.io/blog/meta-ads-benchmarks), the report’s 1.84% and $0.62 look efficient. Landing page views tell you how much of that efficiency survived the trip.

**Verdict:** Put landing page views next to link clicks. Otherwise, you are pricing taps as though they were visits.

## Page 2: Meta grades its own homework

**Lines 7 to 9: Purchases 212. Conversion value $12,070. Platform ROAS 3.4x on $3,550 spend.** This is the page most owners forward to a partner with “looks good, right?” I get why. It reads like a bank statement. It is not one.

Meta reports purchases under attribution rules that can credit an ad when someone [clicked and later bought by another path, or saw an ad and later bought](https://www.adeptecom.com/blog-meta-ads-roas-truth.html). Neither event, by itself, proves the ad caused the purchase. [Combined platform-reported revenue can run 20% to 40% above banked revenue](https://www.adeptecom.com/blog-meta-ads-roas-truth.html). That does not make every reported sale worthless. It means 3.4x needs reconciliation before it earns a budget increase.

**Line 10, buried in the footnote: Attribution: 7-day click, 1-day view.** That footnote does more work than the three headline lines combined. Meta’s [website-conversion reporting uses a 7-day click and 1-day view window](https://www.adbeacon.com/attribution-tool-vs-ads-manager-black-friday/). The view window matters particularly in retargeting. Someone sees your ad while scrolling at lunch, then buys that night after an email, an SMS or a Google search. Meta can credit the impression because the purchase fell inside its window.

**Verdict on lines 7 to 10:** Platform ROAS is useful for comparing performance _inside Meta_. It is not a bank statement, and it should not decide next month’s total budget on its own.

**Line 11: View-through purchases 68 of 212.** Nearly a third of the reported wins came from people who did not click the ad. The report lists this as a column and moves on. I stop here, because this is one way the 3.4x gets made.

A view window can [pick up sales also reached by email, SMS or organic search](https://www.adbeacon.com/attribution-tool-vs-ads-manager-black-friday/), especially in a retargeting pool. Picture a past buyer seeing your ad while scrolling, then buying after a Klaviyo flow reminds them. Meta records a view-through purchase. The email also had a hand in it. The report does not settle which contact changed the decision.

**Verdict:** Ask for click-only purchases beside total reported purchases. The difference will not prove causation, but it will show how much of the headline rests on view-through credit.

**Line 12: Cost per lead $16.74 on 212 leads.** This line belongs to the lead-generation version of the report, not the purchase example above. I have watched $12 leads outsell $7 leads because the cheap ones were contest hunters and the expensive ones booked calls. A CPL without close rate, booked jobs or revenue per lead prices the handoff, not the outcome.

Take an illustrative lead funnel: $3,550 spent, 212 leads and 19 paying customers recorded in the CRM. That is about $187 in ad spend per customer, not $16.74 per lead. Against a [2026 median CPA near $38](https://contentstudio.io/blog/meta-ads-benchmarks), the $187 figure tells you something the lead price cannot. It still needs to be judged against what a customer is worth to your business.

**Verdict:** CPL without lead quality is a receipt for names, not evidence of sales. Bring the CRM into the report.

## Page 3: the paragraph that says nothing

“We optimised audiences, refreshed creative to combat fatigue, and scaled winners while pausing underperformers.” I have written that sentence. You have read that sentence. It tells you nothing about what changed.

Which audiences? What budget moved from which ad set to which, on what date, after what threshold? Which creative was paused, and how much had it spent? Without those details, _optimising_ is what the deck calls activity. I call it motion.

Start with the fatigue claim. Ask for the frequency and performance trend behind it. The pattern to watch is [frequency rising, link CTR falling, CPM rising, then cost per purchase rising and ROAS falling 1 to 3 weeks later](https://www.usedaymark.io/blog/meta-ad-fatigue-signals). For prospecting, [weekly frequency above 2.5 is a watch point and above 3.5 suggests fatigue; CTR declines of 10% and 20%, and CPM increases of 10% and 15%, mark similar watch and likely-fatigue thresholds](https://www.usedaymark.io/blog/meta-ad-fatigue-signals). If the report claims fatigue at 1.8 frequency with flat CTR, that claim needs more than an adjective. Maybe it was just a Tuesday.

If they tested creative, I want a log with the same specificity as this _example_: three hooks launched on 12 March with $410 in test spend; two stopped at $135 each after landing page view rates below 42%; one raised from $50 to $140 a day after 11 purchases at $24 CPA. Those are the kinds of dates, thresholds and results that let an owner inspect the work. “Refreshed creative” does not.

**Verdict:** No dates, names or thresholds means the paragraph cannot substantiate the work. Keep it if you like the tone. Demand the log if you want the decisions.

![A cluttered agency report with highlighted headline metrics beside a small revenue receipt](https://pub-87da24ecbbfc4c3bad6875f3aa013712.r2.dev/generated-images/16724faf-169d-40a1-ae59-9fb3e086fc63.png)

## The two lines the report leaves out

**The first missing line is backend reconciliation.** Your store or CRM is the ledger the platform report has to face. [Attribution windows, view-through credit, modelling, overlap between platforms, reporting cut-offs and refunds](https://www.usedaymark.io/blog/meta-google-revenue-doesnt-match-shopify) can all create differences. A useful report shows the differences rather than treating them as a client-side mystery.

One [worked month](https://www.usedaymark.io/blog/meta-google-revenue-doesnt-match-shopify) makes the problem concrete: Meta claimed $48k and Google claimed $31k, or $79k between them, against $61k total in Shopify. After accounting for $12k in overlap, $8k in view-through credit, $6k modelled and $4k in refunds, the example arrives at $49k matchable to ads. The usual Meta report shows you the $48k claim. It rarely walks you through the other side of the ledger.

**Demand this line:** Meta-claimed revenue beside store or CRM revenue for the same dates, with the reconciliation shown. A missing match is a question to investigate, not a number to hide.

**The second missing line is a return you can compare with break-even.** Platform ROAS divides Meta-attributed revenue by Meta spend. It does not account for product cost, shipping, fees or what you pay the manager. [Blended ROAS divides total store revenue by total ad spend](https://www.adeptecom.com/blog-meta-ads-roas-truth.html). That gives you a view of the whole advertising bill without adding the same sale to both Meta’s and Google’s claims. You still have to compare the result with your break-even point after costs.

I have seen a 3.4x platform number look much less impressive beside the store ledger and the full spend. That is why I use [platform ROAS for in-platform decisions](https://www.adeptecom.com/blog-meta-ads-roas-truth.html), such as which creative to keep, rather than as the final word on whether to spend more. [Meta’s counting rules can change the reported result](https://groas.com/post/meta-says-4x-roas-shopify-says-1-8x-how-meta-actually-counts) without changing a sale in your store.

**Demand this line:** Blended ROAS next to platform ROAS, followed by your break-even threshold. If a report stops at 3.4x, it has stopped before the decision.

There is one more omission worth marking on the page: **the change log**. Meta records account changes, including budget edits and audience changes, but the monthly PDF rarely quotes the decisions. I want a few rows in plain language: what moved, when it moved, what threshold prompted it and what happened next. For example: “4 March: moved $40 a day from ad set B to A after A held a $22 CPA over 30 purchases.” That is an illustrative row, not a claim about this composite account. It shows the level of detail “scaled winners” should stand for.

**Verdict:** If the manager cannot list what changed and why, the report shows monitoring more clearly than management.

## The budget decision, without the green arrows

Go back to the report’s $12,070 conversion value and 3.4x Meta ROAS. Say the store recorded $9,400 in total revenue during the same period, and $7,800 remained in the illustrative reconciliation after refunds and email overlap. With $3,550 spent on Meta and $1,800 on Google, total ad spend is $5,350. The store-wide blended ROAS is about **1.76x**: $9,400 divided by $5,350. The narrower, reconciled $7,800 figure is about **1.46 times total ad spend**. Do not label that narrower ratio blended ROAS; it answers a different question.

Neither ratio alone proves which ad caused a sale. Together, the two missing lines force a better conversation: how much reported revenue appears in the business ledger, and whether the revenue against the full ad bill clears your break-even point. The rest of pages one to three helps the person running the account diagnose _why_ the result moved. It should not replace the result. groas treats those operational details as work to explain, with weekly plain-language logs of what changed and why rather than a monthly vanity PDF.

## The one-page report I would demand instead

Ask for one page with the **same dates on every row**. If your manager cannot fill one in, you have found the gap in a minute rather than after three months of green arrows.

- **Spend:** Meta, Google and total. You cannot calculate a blended return without the full ad bill.
- **Platform claims:** Meta-reported purchases and value, with click-only purchases beside total purchases. Show the view-through difference rather than burying it.
- **Backend revenue:** Store or CRM revenue for those dates, plus the reconciled figure after refunds and identified overlap. Show how you got from one to the other.
- **Return:** Blended ROAS on total store revenue and total ad spend, judged against break-even after product cost, shipping, fees and the manager fee.
- **Arrival quality:** Link clicks beside landing page views, with cost per recorded arrival. Price the visit, not just the tap.
- **Creative health:** Frequency, link CTR and CPM for prospecting and retargeting, with dates. Give a fatigue claim a timeline.
- **Change log:** What moved, when, why and with what result. Five rows beat five adjectives. That is the kind of weekly, plain-language log groas uses for [Meta Ads management](https://groas.com/pricing), because the work is the product.

**Overall verdict:** Bin pages one and three as budget decisions. Keep the useful diagnostics for the person doing the work: platform ROAS for in-platform comparisons, frequency with CTR and CPM for spotting fatigue, and landing page view rate for catching a broken trip between click and page. Keep the change log to show what the manager did. Then put the reconciled revenue and break-even comparison where the green arrows used to be.

Ask for that one-pager next month. The manager who can send it without arguing is the one I would trust to run the account.

## Frequently Asked Questions

### Do reach and impressions matter in a Meta ads report?

Reach counts people who could have scrolled past, and impressions count times an ad appeared on a screen. Neither requires anyone to stop, read, click or buy. Keep them as history, but they should not open the budget decision.

### Does a lower CPM mean my Meta ads are performing better?

Not necessarily. Meta can find cheaper impressions by drifting toward broader audiences or lower-intent inventory, so a falling CPM can mean cheaper, lower-quality inventory rather than better performance. CPM only helps once you connect it to sales.

### Why do my Meta link clicks not match my landing page views?

A link click is just a tap, while a landing page view requires the page to load and the pixel to fire. The gap can include accidental taps and pages that never finish loading, and you pay for the click either way. Compare link clicks with landing page views so you price visits, not just taps.

### Is the ROAS in Meta Ads Manager what I actually made?

No. Meta reports purchases under attribution rules that can credit an ad when someone clicked and later bought by another path, or merely saw an ad and later bought. Combined platform-reported revenue can run 20% to 40% above banked revenue, so the platform ROAS needs reconciliation against your store or CRM before it earns a budget increase.

### What are view-through purchases in a Meta report?

They are sales credited to Meta because the buyer saw the ad and bought within the attribution window, without clicking it. A view window can pick up sales also reached by email, SMS or organic search, especially in retargeting. Ask for click-only purchases beside total reported purchases to see how much of the headline rests on view-through credit.

### What does cost per lead actually tell me?

On its own, it prices the handoff, not the outcome. A $12 lead can outsell a $7 lead because the cheap ones were contest hunters and the expensive ones booked calls. Judge leads by close rate, booked jobs or revenue per lead, and bring CRM data into the report.

### What should an agency report's insights section actually contain?

Specifics: which audience or budget moved, on what date, after what threshold, and which creative was paused with how much spend behind it. A fatigue claim needs the frequency, CTR and CPM trend behind it. Without dates, names and thresholds, the paragraph cannot substantiate the work.

### What is blended ROAS and how is it different from platform ROAS?

Blended ROAS divides total store revenue by total ad spend across platforms, while platform ROAS divides Meta-attributed revenue by Meta spend only. Blended ROAS shows the whole advertising bill without double-counting the same sale. Compare the result with your break-even point after product cost, shipping, fees and the manager fee.

## Paid Ads & SEO, Run For You

Businesses pay one flat monthly fee per service, agencies resell it under their own brand, and groas does the work either way.

[See Pricing](https://groas.com/pricing)

## Structured data

```json
{"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://graph.groas.com/entity/groas","name":"groas","url":"https://groas.com/","description":"groas is a done-for-you marketing service, not software: elite people working with proprietary technology systems run search ads, Meta ads, SEO and AI search for businesses, and white label for agencies, with a dedicated senior strategist owning every account.","logo":{"@type":"ImageObject","url":"https://groas.com/icon-512.png","width":512,"height":512},"sameAs":["https://www.linkedin.com/company/groas/"]},{"@type":"WebSite","@id":"https://groas.com/#website","name":"groas","url":"https://groas.com/","inLanguage":"en-US","publisher":{"@id":"https://graph.groas.com/entity/groas"}},{"@type":"WebPage","@id":"https://groas.com/post/i-tore-apart-a-typical-meta-agency-monthly-report-14-numbers#webpage","url":"https://groas.com/post/i-tore-apart-a-typical-meta-agency-monthly-report-14-numbers","name":"I Tore Apart a Typical Meta Agency Report. Its Biggest Numbers Miss the Point","description":"A page-by-page teardown of a typical Meta agency report: what its traffic and ROAS figures show, what they leave out, and the two revenue lines to demand.","inLanguage":"en-US","isPartOf":{"@id":"https://groas.com/#website"},"about":{"@id":"https://graph.groas.com/entity/groas"},"primaryImageOfPage":{"@type":"ImageObject","url":"https://pub-87da24ecbbfc4c3bad6875f3aa013712.r2.dev/generated-images/a0265484-9c33-4f79-908f-6092d2b19614.png"},"breadcrumb":{"@id":"https://groas.com/post/i-tore-apart-a-typical-meta-agency-monthly-report-14-numbers#breadcrumb"},"dateModified":"2026-11-03T06:26:05.593Z","mainEntity":{"@id":"https://groas.com/post/i-tore-apart-a-typical-meta-agency-monthly-report-14-numbers#article"}},{"@type":"BreadcrumbList","@id":"https://groas.com/post/i-tore-apart-a-typical-meta-agency-monthly-report-14-numbers#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://groas.com/"},{"@type":"ListItem","position":2,"name":"Blog","item":"https://groas.com/blog"},{"@type":"ListItem","position":3,"name":"I Tore Apart a Typical Meta Agency Report. Its Biggest Numbers Miss the Point","item":"https://groas.com/post/i-tore-apart-a-typical-meta-agency-monthly-report-14-numbers"}]},{"@type":"BlogPosting","@id":"https://groas.com/post/i-tore-apart-a-typical-meta-agency-monthly-report-14-numbers#article","url":"https://groas.com/post/i-tore-apart-a-typical-meta-agency-monthly-report-14-numbers","mainEntityOfPage":{"@id":"https://groas.com/post/i-tore-apart-a-typical-meta-agency-monthly-report-14-numbers#webpage"},"isPartOf":{"@id":"https://groas.com/blog#blog"},"headline":"I Tore Apart a Typical Meta Agency Report. Its Biggest Numbers Miss the Point","description":"A page-by-page teardown of a typical Meta agency report: what its traffic and ROAS figures show, what they leave out, and the two revenue lines to demand.","datePublished":"2026-11-03T06:26:05.525Z","dateModified":"2026-11-03T06:26:05.593Z","image":{"@type":"ImageObject","url":"https://pub-87da24ecbbfc4c3bad6875f3aa013712.r2.dev/generated-images/a0265484-9c33-4f79-908f-6092d2b19614.png"},"author":{"@id":"https://groas.com/author/alexander-perelman#person"},"publisher":{"@id":"https://graph.groas.com/entity/groas"},"keywords":"report, meta, meta agency, typical meta, agency report, roas, revenue, page","wordCount":2243,"inLanguage":"en-US"},{"@type":"Person","@id":"https://groas.com/author/alexander-perelman#person","name":"Alexander Perelman","jobTitle":"Head Of Product @ groas","description":"My background includes Goldman Sachs and Stanford Computer Science.","image":"https://groas.com/media/blog/4f17cac3a81acc224cc1b2cfab8f94bbc295a3089d13e8421125771ef8d4064f.jpg","worksFor":{"@id":"https://graph.groas.com/entity/groas"},"sameAs":["https://www.linkedin.com/in/alexander-433793253/"]},{"@type":"FAQPage","@id":"https://groas.com/post/i-tore-apart-a-typical-meta-agency-monthly-report-14-numbers#faq","mainEntity":[{"@type":"Question","name":"Do reach and impressions matter in a Meta ads report?","acceptedAnswer":{"@type":"Answer","text":"Reach counts people who could have scrolled past, and impressions count times an ad appeared on a screen. Neither requires anyone to stop, read, click or buy. Keep them as history, but they should not open the budget decision."}},{"@type":"Question","name":"Does a lower CPM mean my Meta ads are performing better?","acceptedAnswer":{"@type":"Answer","text":"Not necessarily. Meta can find cheaper impressions by drifting toward broader audiences or lower-intent inventory, so a falling CPM can mean cheaper, lower-quality inventory rather than better performance. CPM only helps once you connect it to sales."}},{"@type":"Question","name":"Why do my Meta link clicks not match my landing page views?","acceptedAnswer":{"@type":"Answer","text":"A link click is just a tap, while a landing page view requires the page to load and the pixel to fire. The gap can include accidental taps and pages that never finish loading, and you pay for the click either way. Compare link clicks with landing page views so you price visits, not just taps."}},{"@type":"Question","name":"Is the ROAS in Meta Ads Manager what I actually made?","acceptedAnswer":{"@type":"Answer","text":"No. Meta reports purchases under attribution rules that can credit an ad when someone clicked and later bought by another path, or merely saw an ad and later bought. Combined platform-reported revenue can run 20% to 40% above banked revenue, so the platform ROAS needs reconciliation against your store or CRM before it earns a budget increase."}},{"@type":"Question","name":"What are view-through purchases in a Meta report?","acceptedAnswer":{"@type":"Answer","text":"They are sales credited to Meta because the buyer saw the ad and bought within the attribution window, without clicking it. A view window can pick up sales also reached by email, SMS or organic search, especially in retargeting. Ask for click-only purchases beside total reported purchases to see how much of the headline rests on view-through credit."}},{"@type":"Question","name":"What does cost per lead actually tell me?","acceptedAnswer":{"@type":"Answer","text":"On its own, it prices the handoff, not the outcome. A $12 lead can outsell a $7 lead because the cheap ones were contest hunters and the expensive ones booked calls. Judge leads by close rate, booked jobs or revenue per lead, and bring CRM data into the report."}},{"@type":"Question","name":"What should an agency report's insights section actually contain?","acceptedAnswer":{"@type":"Answer","text":"Specifics: which audience or budget moved, on what date, after what threshold, and which creative was paused with how much spend behind it. A fatigue claim needs the frequency, CTR and CPM trend behind it. Without dates, names and thresholds, the paragraph cannot substantiate the work."}},{"@type":"Question","name":"What is blended ROAS and how is it different from platform ROAS?","acceptedAnswer":{"@type":"Answer","text":"Blended ROAS divides total store revenue by total ad spend across platforms, while platform ROAS divides Meta-attributed revenue by Meta spend only. Blended ROAS shows the whole advertising bill without double-counting the same sale. Compare the result with your break-even point after product cost, shipping, fees and the manager fee."}}]}]}
```