---
title: "4x Meta ROAS, a $2,499 Loss: The Revenue Reconciliation"
description: "An illustrative $10,000 Meta spend reports 4x ROAS but loses $2,499 after revenue reversals, margin, media and management. Here is the reconciliation."
url: "https://groas.com/post/4x-meta-roas-2-499-in-the-red-a-refund-adjusted-revenue"
image: "https://pub-87da24ecbbfc4c3bad6875f3aa013712.r2.dev/generated-images/4481786e-03d9-4f91-9644-e8940d37ca24.png"
published: "2026-11-06T06:30:46.778Z"
modified: "2026-11-10T21:19:00.826Z"
---

November 6, 2026 · 9 min read

# 4x Meta ROAS, a $2,499 Loss: The Revenue Reconciliation

[](https://groas.com/author/alexander-perelman)

[Alexander Perelman](https://groas.com/author/alexander-perelman)Personally Tested

Head Of Product @ groas

[LinkedIn](https://www.linkedin.com/in/alexander-433793253/)

Updated November 10, 2026

![A gold trophy engraved '4x' with no bottom: banknotes poured in fall straight through onto returned parcels, showing impressive ROAS hiding a real loss.](https://pub-87da24ecbbfc4c3bad6875f3aa013712.r2.dev/generated-images/4481786e-03d9-4f91-9644-e8940d37ca24.png)

In this article

1. [Start with the inputs, not the ROAS screenshot](#start-with-the-inputs-not-the-roas-screenshot)
2. [At 4x, the initial cushion is $501](#at-4x-the-initial-cushion-is-501)
3. [Reconcile reported purchases with retained sales](#reconcile-reported-purchases-with-retained-sales)
4. [Find the break-even point before you call 4x a win](#find-the-break-even-point-before-you-call-4x-a-win)
5. [Know what this table cannot prove](#know-what-this-table-cannot-prove)
6. [Make the manager show the reconciliation](#make-the-manager-show-the-reconciliation)
7. [Put the $1,499 fee through the same test](#put-the-1499-fee-through-the-same-test)

A Meta account can report 4x ROAS while the store loses $2,499 after ads and management. In this model, $10,000 in media produces $40,000 in reported purchase value. If 25% of that value never becomes retained sales, the screenshot still says 4x. The store’s contribution says something else.

**This is a model, not a client result.** I chose the spend, revenue, margin and reversal rates to show where the arithmetic turns. The management fee is the published input: **$1,499 a month for Meta Ads, with media spend separate**, listed on [groas’s Meta Ads pricing page](https://groas.com/pricing). With no revenue reversed, the example leaves $501 after media and management. I used to tell clients a 4x was comfortable. I was wrong when I said it without checking what settled.

## Start with the inputs, not the ROAS screenshot

Say you spend $10,000 a month, Meta reports $40,000 in purchase value, and the store keeps a 30% contribution margin on realized sales before advertising and management. Here, that margin means sales revenue after product, fulfillment, shipping, packing, and payment and platform costs. It **does not already include a refund reserve**; the revenue reversal is applied separately. Use one definition throughout or you will subtract the same loss twice.

| Input                                         |              Value in this model | Basis                                        |
| --------------------------------------------- | -------------------------------: | -------------------------------------------- |
| Media spend                                   |                          $10,000 | Illustrative store input                     |
| Meta-reported purchase value                  |                          $40,000 | Illustrative store input; 4.0x reported ROAS |
| Contribution margin before ads and management |            30% of realized sales | Illustrative assumption                      |
| Management fee                                | $1,499 per month, media separate | Published groas pricing                      |

These are not benchmarks for your store. Replace the first three inputs with your own records; keep the fee only if you are evaluating that quote. The 30% margin also assumes the same contribution rate on sales that remain after reversals. It does not separately model costs you cannot recover on a returned order, so an actual store calculation may be worse.

The margin matters more than the headline multiple. Before a management fee, break-even ROAS is 1 divided by contribution margin: at 25% margin, media alone needs 4.0x; at 40%, it needs 2.5x ([break-even ROAS calculation](https://eshoppick.com/guides/how-to-calculate-break-even-roas-for-ecommerce)). Once you add a fee, the required multiple rises. A platform ROAS figure cannot tell you which side of that line you are on.

**Free tool:** the [ROAS Calculator](https://groas.com/free-tools/roas-calculator) works out your return on ad spend, your break-even ROAS and the budget a revenue goal takes.

## At 4x, the initial cushion is $501

The first pass is deliberately plain. Apply the assumed 30% margin to the full $40,000 reported value, then subtract the $10,000 media bill and $1,499 management bill.

| Calculation                                 |             Illustrative amount |
| ------------------------------------------- | ------------------------------: |
| Reported purchase value × 30% margin        | $12,000 contribution before ads |
| Less media spend                            |                        −$10,000 |
| Less management fee                         |                         −$1,499 |
| **Contribution after media and management** |                       **+$501** |

That $501 is **not company net profit**. Rent, salaries, software, design, taxes and other overhead still come out of it. A separate worked store example reaches the same kind of uncomfortable result from different inputs: $15,000 in Meta spend reporting 4.2x ROAS, or $63,000, still lost about $230 after product, shipping, returns, fees and apps ([worked break-even example](https://merchcalc.com/blog/shopify-break-even-roas-cac-calculator)). The ratio looked healthy. The contribution did not.

What the deck calls _profitable ROAS_, I call contribution minus the two checks you wrote for media and management. Management does not appear inside platform ROAS, but it belongs in the decision about whether management pays for itself. And so far, this generous first pass has treated every reported dollar as a dollar the store kept.

## Reconcile reported purchases with retained sales

A purchase event can stay in a platform report even when the store later cancels or refunds the order, depending on how conversion values are sent and updated. Platform and store totals can also diverge for other reasons ([common reconciliation gaps](https://mcp.adray.ai/blog/why-marketing-numbers-never-match/)). One published reconciliation describes $48,000 claimed by Meta plus $31,000 claimed by Google against $61,000 in Shopify sales; its walk-down identifies $4,000 in refunds alongside overlap and modeled conversions, leaving $49,000 matched as ad-attributable ([reconciliation example](https://www.usedaymark.io/blog/meta-google-revenue-doesnt-match-shopify)). Those are that example’s figures, not a refund rate to apply to every account.

For this model, a _revenue reversal_ means some of the $40,000 reported purchase value is not retained as sales. It might be cancelled or refunded. I use 10% and 25% as sensitivity cases, **not claimed industry rates**. The calculation applies the same assumed margin to the revenue that remains, then subtracts the same media and fee.

| Share of reported value not retained | Realized revenue | Realized revenue ÷ media | Contribution after $10,000 media and $1,499 fee |
| ------------------------------------ | ---------------: | -----------------------: | ----------------------------------------------: |
| 0%                                   |          $40,000 |                     4.0x |                                       **+$501** |
| 10%                                  |          $36,000 |                     3.6x |                                       **−$699** |
| 25%                                  |          $30,000 |                     3.0x |                                     **−$2,499** |

At 10%, $36,000 × 30% leaves $10,800 before ads. Subtract $11,499 and the initial $501 cushion becomes a $699 loss. At 25%, $30,000 × 30% leaves $9,000; after media and management, that is a **$2,499 loss**. The ad account can still display its original 4x if its reported purchase value has not been adjusted.

Returns are not a hypothetical category of cost: one summary puts US returns in 2025 at about $849.9 billion, or 15.8% overall, with roughly 19.3% of online sales returned ([return figures](https://synctrack.io/blog/cost-of-returns/)). Those broad figures do not validate either sensitivity case for your store. Your own order records decide that.

There is one guardrail before you copy the table: check whether the values sent through your pixel or Conversions API already account for refunds. **If reported conversion value is already net of those reversals, do not subtract them again.** The useful comparison is reported value against retained value for the same orders and the same window, not an automatic haircut applied to every Meta report.

![Ledger showing reported purchase value reconciled to smaller retained sales](https://pub-87da24ecbbfc4c3bad6875f3aa013712.r2.dev/generated-images/b1e8df31-2eaa-4e8f-90cd-6c0276901620.png)

## Find the break-even point before you call 4x a win

The table gives you a sign: positive or negative under each assumption. You can also solve for the gross reported revenue this example would need to break even at a 25% reversal rate.

Media plus management totals $11,499. At a 30% margin, that requires $38,330 in realized sales. If only 75% of reported purchase value remains, $38,330 ÷ 0.75 is about $51,107 in gross reported value. Against $10,000 in media, that is **roughly 5.11x reported ROAS** just to cover the modeled media and fee. It is not a universal Meta target. Change the margin, fee or reversal rate, and the threshold changes with it.

This is why a higher reported ROAS is not, by itself, evidence that an agency improved the store’s economics. It may be an improvement in the platform’s ratio. The decision needs the revenue and costs underneath it.

## Know what this table cannot prove

Refund adjustment settles one question: under these assumptions, does retained revenue cover variable costs, media and management? It does **not** establish that Meta caused every sale it reports. Platform attribution can claim an order that another channel also claims, and modeled or view-through conversions add another reason to separate a platform claim from a store order. In the [reconciliation example](https://www.usedaymark.io/blog/meta-google-revenue-doesnt-match-shopify), Meta and Google together claimed $79,000 against $61,000 in Shopify sales. Adding their reports was not a way to count distinct sales.

Incrementality is a different test. A discussion of Meta measurement describes substantial gaps between reported and incremental results, with prospecting and retargeting behaving differently ([reported versus incremental ROAS](https://www.gofunnel.ai/blog/incrementality-meta-framework)). Those estimates are not adjustments I would paste into this model. A Conversion Lift test or a four-to-six-week geo holdout addresses the causal question; a refund-adjusted ledger addresses the revenue question. Neither substitutes for the other.

Repeat purchases and customer mix sit outside the table too. A $30 margin on a first order looks different if that buyer reorders twice without another ad cost. Pooled new and returning revenue can hide which part an agency actually found. I keep the margin discussion in [_A 3x Meta ROAS Can Lose Money: The Margin Table That Matters_](https://groas.com/post/ignore-roas-benchmarks-by-industry-your-meta-breakeven-roas) next to this calculation for that reason. **Use the table to stop grading gross ROAS; use an incrementality test to examine whether the next $10,000 of spend creates sales.**

## Make the manager show the reconciliation

I have seen the Ads Manager tab say 4x while the store showed something smaller. That gap needs a walk-down, not a better-looking slide. Before you renew an ecommerce Meta manager, ask them to put three things beside their reported ROAS: realized order revenue for the same window, the conversion-value rules behind the platform figure, and the treatment of cancellations and refunds. Buying advice for ecommerce Meta agencies similarly points readers toward contribution and sales, rather than clicks alone, and toward asking who works on the account and what they optimize ([agency evaluation criteria](https://velocityppc.com/blog/best-facebook-ads-agencies-ecommerce/)).

I would run the monthly review in this order:

1. **Match the window and the orders.** Pull net-of-refund store actuals for the purchases being compared with the platform report. Keep the comparison window fixed; platform click dates and store order dates can drift across month end.
2. **Get the conversion-value rule in writing.** Ask what purchase value is sent at event time, whether it changes after a refund, and which attribution window produced the report. This tells you whether a reversal belongs in the walk-down at all.
3. **Calculate contribution after media and the fee.** Apply your actual contribution margin to retained sales, then subtract the bills. Keep other overhead separate so nobody quietly calls this company net profit.
4. **Ask what management changed.** Compare the account work with realized contribution per dollar of media, not just a movement in reported ROAS.

If the provider cannot put those answers next to the ROAS claim, you cannot tell from that claim whether management earned its fee.

![Whiteboard showing refund-adjusted contribution math for a Meta ad account](https://pub-87da24ecbbfc4c3bad6875f3aa013712.r2.dev/generated-images/fc707eac-0bb7-4294-8eb1-d5668098e9d7.png)

## Put the $1,499 fee through the same test

I ran groas’s published $1,499 monthly Meta fee through this model on purpose. **The flat fee is an input to test, not proof of performance.** At $10,000 in media, a hypothetical percentage shop charging 15% would bill $1,500, almost the same amount. At $30,000 in media, that same rate becomes $4,500 while the quoted groas fee stays $1,499. An [ecommerce agency roundup](https://velocityppc.com/blog/best-facebook-ads-agencies-ecommerce/) lists monthly minimums from $1,500 to $5,000, as well as hourly prices of $100 to $199. That comparison puts groas’s fee near the low end before anyone makes a performance claim. It does not make a losing account profitable.

The companion calculation in [_The ROAS Lift a Meta Manager Needs to Earn Their Fee_](https://groas.com/post/the-roas-lift-your-meta-manager-has-to-deliver-just-to-pay) asks how much lift covers management at your spend, before revenue reversals enter the picture. Ask every finalist for that number, then ask them to redo it on retained sales.

I spent years doing the manual version of this work: mining negatives at 1am, nudging budgets, rewriting tired creative. groas pairs continuous execution on audiences, budgets and ads, including testing new ads and scaling winners, with a named strategist responsible for direction and the numbers. That is the operating argument for a flat fee instead of more hours or a rising share of spend. The reconciliation still decides whether the arrangement works for the store.

So run the same three-line test on all of us: take last month’s retained revenue for one fixed window, multiply by your contribution margin before ads, then subtract media and the quoted fee. Check the 10% and 25% reversal cases only as sensitivities if you do not yet know how much reported value remains. If the answer flips negative, a prettier platform ROAS does not repair the economics; margin, reversals or costs have to change.

**Decision: do not renew a Meta manager on reported ROAS alone.** Renew on realized contribution after media and fee, reconciled to net-of-refund orders in one fixed window. In this illustrative account, that rule turns a $501 cushion into a $699 loss at 10% reversals and a $2,499 loss at 25%. Run your last 30 days through the ledger before the next call. I would rather lose a client who wants a prettier 4x than keep one who pays me $1,499 to preside over negative contribution.

## Frequently Asked Questions

### Can a Meta account show 4x ROAS while the store actually loses money?

Yes. In the illustrative model, $10,000 of media produces $40,000 in reported purchase value, which is 4x ROAS, yet the store loses $2,499 once a 25% revenue reversal, 30% contribution margin, media spend and the $1,499 management fee are applied. The screenshot shows the platform ratio; the store contribution shows the real result.

### Does contribution margin include refunds, or are refunds subtracted separately?

In this model the 30% contribution margin before ads and management does not already include a refund reserve. The revenue reversal is applied separately, so subtracting refunds again inside the margin would count the same loss twice. The margin also does not separately model unrecoverable costs on returned orders, so a real store's result may be worse.

### How do refunds turn a positive ROAS result into a loss?

The model treats part of the $40,000 reported purchase value as cancelled or refunded rather than retained as sales. At a 10% reversal, $36,000 in realized sales at a 30% margin leaves $10,800, which becomes a $699 loss after $10,000 media and the $1,499 fee. At 25%, realized sales of $30,000 produce a $2,499 loss, while the platform can still display its original 4x.

### Should I subtract refunds from Meta's reported conversion value in my reconciliation?

Only if the reported conversion value is not already net of them. Check whether the values sent through your pixel or Conversions API already account for refunds; if they do, do not subtract reversals again. The useful comparison is reported value against retained value for the same orders in the same window, not an automatic haircut applied to every report.

### What ROAS do I actually need to break even once fees and refunds are counted?

Before a management fee, break-even ROAS is 1 divided by your contribution margin, so 4.0x at a 25% margin and 2.5x at 40%. Adding a fee raises the threshold. In the model with a 30% margin, $10,000 media and the $1,499 fee, a 25% reversal rate means reported purchase value must reach roughly $51,107, or about 5.11x, just to cover media and management. The threshold changes with your own margin, fee and reversal rate.

### Does a high reported ROAS prove Meta actually caused the sales it claims?

No. Reported ROAS does not establish that Meta caused every sale it reports, because platform attribution can claim orders other channels also claim, and modeled or view-through conversions add further overlap. Incrementality is a separate test addressed by a Conversion Lift test or a four-to-six-week geo holdout, while a refund-adjusted ledger addresses the revenue question. Neither substitutes for the other.

### What should I ask my Meta agency for before renewing, instead of the ROAS number?

Ask them to put realized order revenue for the same window, the conversion-value rules behind the platform figure, and the treatment of cancellations and refunds beside their reported ROAS. Renew on realized contribution after media and the fee, reconciled to net-of-refund orders in one fixed window, not on reported ROAS alone. If they cannot provide these, the ROAS claim alone cannot tell you whether management earned its fee.

### Does a flat management fee make a losing ad account profitable?

No. A flat fee is an input to test, not proof of performance. In the model, the $1,499 fee is an input alongside media spend and margin, and even with no revenue reversed the example leaves only $501 before overhead. The fee still has to be covered by realized contribution for the arrangement to work for the store.

## Run Your Marketing On groas

Plug ChatGPT, Claude or any AI agent into groas and let it do the work of a whole marketing team: free to ask, $49 a month flat with every feature included, and autopilot whenever you’re ready.

[Start Free](https://groas.com/pricing)

[Or have groas run it for you](https://groas.com/done-for-you)

## 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 the marketing infrastructure for your own AI: plug Claude, ChatGPT or any AI agent into groas and run all your marketing from it, for a tiny fraction of the usual cost, with the groas app there whenever you want to see more. It is the layer ChatGPT, Claude and any AI assistant run a business’s online marketing on. Through groas an assistant can do anything you could do natively in Google Ads, Bing Ads, Meta Ads and ChatGPT ads, on your website and in AI search, and, where the owner switches it on, can run connected campaigns by itself, so AI agents do the work a whole marketing team used to (free to read, $49 a month to make changes, delivered as an MCP server). Separately, its done-for-you service has senior people and proprietary quantitative systems run paid ads, SEO and AI search for businesses, and white label for agencies, from $5,000 a month.","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/4x-meta-roas-2-499-in-the-red-a-refund-adjusted-revenue#webpage","url":"https://groas.com/post/4x-meta-roas-2-499-in-the-red-a-refund-adjusted-revenue","name":"4x Meta ROAS, a $2,499 Loss: The Revenue Reconciliation","description":"An illustrative $10,000 Meta spend reports 4x ROAS but loses $2,499 after revenue reversals, margin, media and management. Here is the reconciliation.","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/4481786e-03d9-4f91-9644-e8940d37ca24.png"},"breadcrumb":{"@id":"https://groas.com/post/4x-meta-roas-2-499-in-the-red-a-refund-adjusted-revenue#breadcrumb"},"dateModified":"2026-11-10T21:19:00.826Z","mainEntity":{"@id":"https://groas.com/post/4x-meta-roas-2-499-in-the-red-a-refund-adjusted-revenue#article"}},{"@type":"BreadcrumbList","@id":"https://groas.com/post/4x-meta-roas-2-499-in-the-red-a-refund-adjusted-revenue#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":"4x Meta ROAS, a $2,499 Loss: The Revenue Reconciliation","item":"https://groas.com/post/4x-meta-roas-2-499-in-the-red-a-refund-adjusted-revenue"}]},{"@type":"BlogPosting","@id":"https://groas.com/post/4x-meta-roas-2-499-in-the-red-a-refund-adjusted-revenue#article","url":"https://groas.com/post/4x-meta-roas-2-499-in-the-red-a-refund-adjusted-revenue","mainEntityOfPage":{"@id":"https://groas.com/post/4x-meta-roas-2-499-in-the-red-a-refund-adjusted-revenue#webpage"},"isPartOf":{"@id":"https://groas.com/blog#blog"},"headline":"4x Meta ROAS, a $2,499 Loss: The Revenue Reconciliation","description":"An illustrative $10,000 Meta spend reports 4x ROAS but loses $2,499 after revenue reversals, margin, media and management. Here is the reconciliation.","datePublished":"2026-11-06T06:30:46.778Z","dateModified":"2026-11-10T21:19:00.826Z","image":{"@type":"ImageObject","url":"https://pub-87da24ecbbfc4c3bad6875f3aa013712.r2.dev/generated-images/4481786e-03d9-4f91-9644-e8940d37ca24.png"},"author":{"@id":"https://groas.com/author/alexander-perelman#person"},"publisher":{"@id":"https://graph.groas.com/entity/groas"},"keywords":"000, roas, meta, revenue, media, 499, management, margin","wordCount":1989,"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/4x-meta-roas-2-499-in-the-red-a-refund-adjusted-revenue#faq","mainEntity":[{"@type":"Question","name":"Can a Meta account show 4x ROAS while the store actually loses money?","acceptedAnswer":{"@type":"Answer","text":"Yes. In the illustrative model, $10,000 of media produces $40,000 in reported purchase value, which is 4x ROAS, yet the store loses $2,499 once a 25% revenue reversal, 30% contribution margin, media spend and the $1,499 management fee are applied. The screenshot shows the platform ratio; the store contribution shows the real result."}},{"@type":"Question","name":"Does contribution margin include refunds, or are refunds subtracted separately?","acceptedAnswer":{"@type":"Answer","text":"In this model the 30% contribution margin before ads and management does not already include a refund reserve. The revenue reversal is applied separately, so subtracting refunds again inside the margin would count the same loss twice. The margin also does not separately model unrecoverable costs on returned orders, so a real store's result may be worse."}},{"@type":"Question","name":"How do refunds turn a positive ROAS result into a loss?","acceptedAnswer":{"@type":"Answer","text":"The model treats part of the $40,000 reported purchase value as cancelled or refunded rather than retained as sales. At a 10% reversal, $36,000 in realized sales at a 30% margin leaves $10,800, which becomes a $699 loss after $10,000 media and the $1,499 fee. At 25%, realized sales of $30,000 produce a $2,499 loss, while the platform can still display its original 4x."}},{"@type":"Question","name":"Should I subtract refunds from Meta's reported conversion value in my reconciliation?","acceptedAnswer":{"@type":"Answer","text":"Only if the reported conversion value is not already net of them. Check whether the values sent through your pixel or Conversions API already account for refunds; if they do, do not subtract reversals again. The useful comparison is reported value against retained value for the same orders in the same window, not an automatic haircut applied to every report."}},{"@type":"Question","name":"What ROAS do I actually need to break even once fees and refunds are counted?","acceptedAnswer":{"@type":"Answer","text":"Before a management fee, break-even ROAS is 1 divided by your contribution margin, so 4.0x at a 25% margin and 2.5x at 40%. Adding a fee raises the threshold. In the model with a 30% margin, $10,000 media and the $1,499 fee, a 25% reversal rate means reported purchase value must reach roughly $51,107, or about 5.11x, just to cover media and management. The threshold changes with your own margin, fee and reversal rate."}},{"@type":"Question","name":"Does a high reported ROAS prove Meta actually caused the sales it claims?","acceptedAnswer":{"@type":"Answer","text":"No. Reported ROAS does not establish that Meta caused every sale it reports, because platform attribution can claim orders other channels also claim, and modeled or view-through conversions add further overlap. Incrementality is a separate test addressed by a Conversion Lift test or a four-to-six-week geo holdout, while a refund-adjusted ledger addresses the revenue question. Neither substitutes for the other."}},{"@type":"Question","name":"What should I ask my Meta agency for before renewing, instead of the ROAS number?","acceptedAnswer":{"@type":"Answer","text":"Ask them to put realized order revenue for the same window, the conversion-value rules behind the platform figure, and the treatment of cancellations and refunds beside their reported ROAS. Renew on realized contribution after media and the fee, reconciled to net-of-refund orders in one fixed window, not on reported ROAS alone. If they cannot provide these, the ROAS claim alone cannot tell you whether management earned its fee."}},{"@type":"Question","name":"Does a flat management fee make a losing ad account profitable?","acceptedAnswer":{"@type":"Answer","text":"No. A flat fee is an input to test, not proof of performance. In the model, the $1,499 fee is an input alongside media spend and margin, and even with no revenue reversed the example leaves only $501 before overhead. The fee still has to be covered by realized contribution for the arrangement to work for the store."}}]}]}
```
