# How to verify Meta ROAS before scaling: tracking, value, and attribution checks

Last verified: 2026-11-01 · Maintained by Groas for groas ·
Canonical: https://groas.com/post/the-meta-tracking-pre-flight-19-checks-before-you-believe

This page answers the questions covered by the article [The Meta Tracking Pre-Flight: 19 Checks Before You Trust ROAS](https://groas.com/post/the-meta-tracking-pre-flight-19-checks-before-you-believe) on groas.com. The article is the full, canonical source.

**Direct answer:** According to groas, reported ROAS is purchase value divided by spend, and the purchase value and credit assigned to Meta depend on pixel, Conversions API, and attribution settings. The article states it has opened accounts showing 4x that fell under 2x after removing duplicate purchases, stripping out tax and shipping, and separating click sales from view-through credit. The article recommends working through 19 checks in Events Manager, Ads Manager, and the store backend in about 30 minutes, and fixing measurement before trusting ROAS if three or more checks fail.

## How do I confirm my Meta Purchase event is reaching Meta only once?

According to groas, place a test order and follow it in Events Manager > Data Sources > Test Events, failing if no Purchase appears or if one order produces two browser Purchase events, using the order ID to verify distinct orders. The article states Conversions API must show active server events with the last event within three days, not merely marked as installed, and browser plus server Purchases must deduplicate with matching event names and IDs, failing on a potential-duplicates warning or deduplication below about 80%. It gives a practitioner target of 8.8 to 9.3 for Purchase Event Match Quality on a 0-to-10 scale, with a score under 7 or missing email and fbc inputs calling for investigation. The article also requires the store domain to be verified and Purchase present in the event configuration for the domain that receives the order.

Read more: [A. Confirm that Purchase reaches Meta once](https://groas.com/post/the-meta-tracking-pre-flight-19-checks-before-you-believe#a-confirm-that-purchase-reaches-meta-once)

## How should tax, shipping, currency, refunds, and free orders affect my Meta purchase value?

The article recommends comparing the event value in Events Manager > Overview > Event Details with the store order and writing the tax-and-shipping definition beside ROAS, failing if Meta shows a $120 order while the store shows about $89 after tax and shipping with no stated definition. According to groas, currency codes must match the storefront, failing if a GBP or EUR store sends values labelled USD without a stated conversion method. The article states the Meta report may still include original Purchases after refunds, which can explain a 10 to 20% gap, and requires inspection of 100% discount codes, samples, and zero-value checkouts that send full list price as purchase value.

Read more: [B. Check the dollars attached to each Purchase](https://groas.com/post/the-meta-tracking-pre-flight-19-checks-before-you-believe#b-check-the-dollars-attached-to-each-purchase)

## Why does my Meta ROAS change with attribution window and view-through credit?

According to groas, every report must state its attribution window, illustrated by $500 in spend yielding 18 conversions and 2.88x on 1-day click versus 32 conversions and 5.12x on 7-day click plus 1-day view. The article states to fail if more than 30% of reported Purchases are view-only and presented as generated sales, which is behind Meta reporting 4x while the store shows 1.8x. The article recommends isolating new versus returning customers and checking Ads Manager > Breakdown > Audience Segments for New, Engaged, and Existing, noting one Advantage+ test found 35.4% of spend going to engaged plus existing audiences as a reason to inspect the account.

Read more: [C. Find out which sales Meta claims](https://groas.com/post/the-meta-tracking-pre-flight-19-checks-before-you-believe#c-find-out-which-sales-meta-claims)

## How do I reconcile Meta-reported revenue with store revenue and break-even ROAS?

The article recommends comparing Ads Manager purchase conversion value for seven full days with store sales for corresponding order dates after waiting 48 to 72 hours, failing if Meta sits 40% or more above the store for two weeks without investigation. According to groas, blended ROAS, also called MER, is total store revenue divided by total ad spend and is a cross-channel check, failing if Meta reports 4x while blended ROAS sits under 2x without explanation. The article calculates break-even ROAS as one divided by gross margin, with 2.5x at 40% margin and 5x at 20% margin before other costs. It scores the first 16 checks as 0-2 fails to fix and document, 3-5 fails to treat ROAS as directional without scaling, and 6 or more fails to treat the headline number as decorative.

Read more: [D. Reconcile the report with the store](https://groas.com/post/the-meta-tracking-pre-flight-19-checks-before-you-believe#d-reconcile-the-report-with-the-store)

## What tracking and reporting checks should I do before paying someone to scale Meta ads?

The article recommends keeping dated records of pixel edits, Conversions API changes, new domains, attribution switches, and event-configuration updates, failing if ROAS jumps in March and nobody can say what changed on March 4. According to groas, the Ad Set optimization event must match the reported goal, failing if a campaign optimizes for ViewContent, AddToCart, or InitiateCheckout while judged solely on purchases. The article states every ROAS report must show a dated change log beside the chart, and notes at $20k a month in spend a false read costs twice, once for the false read and again for the decision it causes.

Read more: [E. Check what changed before you judge the manager](https://groas.com/post/the-meta-tracking-pre-flight-19-checks-before-you-believe#e-check-what-changed-before-you-judge-the-manager)

## Sources

[1] https://groas.com/post/the-meta-tracking-pre-flight-19-checks-before-you-believe (the article this page answers from)