Same word, different maths (Meta vs Google, decoded)
Chinmay Raibagkar · reconciles the two dashboards every Monday⚖️ with a Monday checklist, not a lecturePublished September 10, 2026 · Updated September 14, 2026
The 60-second version
- A conversion means different things per platform — different events, credit rules, and repeat-counting.
- Windows move numbers without moving sales. Shorten the lookback and reported conversions plunge 20–40%.
- The fix: never sum platforms; reconcile weekly against one warehouse number.
Last Tuesday a founder sent two screenshots taken ten minutes apart. Meta Ads reported a 4.2x ROAS for the week. Google Ads reported 3.1x for the exact same week, on the same store, selling the same products. He asked which platform was lying.
Neither platform was lying. They were simply answering different questions using identical metric names. A conversion in Meta's dashboard and a conversion in Google's dashboard are two entirely different mathematical calculations.
Why conversions mean different things on each platform
A conversion is not an objective fact that both platforms observe the same way. It is a set of counting rules applied to user events, and every rule differs across platforms:
- What counts as an event. In Google Ads, you define conversion actions manually (like a purchase or lead form), each with custom values and counting rules. In Meta, events stream in from the pixel or Conversions API, and reporting tables often combine standard events and custom conversions.
- What gets the credit. Google defaults to data-driven attribution across Google search, shopping, and display clicks, splitting fractional credit. Meta defaults to its own 7-day click or 1-day view rule, giving 100% of the credit to the last Meta touchpoint.
- How repeat purchases are counted. Google lets you choose between counting "every" purchase or "one" conversion per user. Meta defaults to total conversion events, meaning one active customer purchasing three times can produce three conversions on Meta and one on Google.
Here is a clear example: a customer clicks a Google ad on Monday, taps a Meta ad on Wednesday, and places a single order for ₹4,999 on Friday. Google claims one conversion worth ₹4,999. Meta also claims one conversion worth ₹4,999. Your bank account and warehouse show one order worth ₹4,999. If you sum both dashboards, they report ₹9,998 in credited revenue from a single ₹4,999 checkout.
The golden rule of ad reporting: never add attributed conversions or revenue across platforms. Each ad platform calculates its numbers as if it were the only channel in the world. Adding them together counts shared customer journeys twice.
Attribution windows change numbers without changing sales
An attribution window is the lookback period: how long after a click or ad impression a sale still gets credited to that campaign.
By default, Meta credits purchases within a 7-day click and 1-day view window, while Google Ads uses data-driven attribution with a click lookback of up to 30 days.
This difference creates large reporting swings. If an agency switches a Meta ad set from a 7-day click and 1-day view window down to a 1-day click window, reported conversions immediately plunge by 20% to 40%. Nothing broke in your ads, and your customers did not stop buying. The measuring tape simply got shorter.
Meta reported conversions dropped from 312 to 201 week-on-week, but your warehouse order count stayed completely flat at ₹39L. What is the most likely cause?
Always log every change to attribution settings with a specific calendar date. When a metric drops sharply from one week to the next, check the ad set settings before rewriting your ad creative.
Platform ROAS versus blended business return
Platform ROAS and blended ROAS share the same mathematical formula—revenue divided by spend—but their revenue numbers come from different sources:
Two Numbers Called ROAS
Data JourneyPlatform ROAS
The platform's own credited revenue divided by its own ad spend, inside its own window. Meta 4.2x and Google 3.1x comfortably coexist.
Blended ROAS (MER)
Total bank revenue divided by total ad spend across all channels. No attribution windows or double counting.
The relationship
The sum of platform claims will almost always exceed warehouse revenue. Blended ROAS sets budget; platform ROAS guides internal tweaks.
Consider a typical weekly report for an Indian ecommerce brand:
- Meta Ads: claims ₹14.2 lakh in revenue on ₹3.4 lakh spend (4.2x ROAS).
- Google Ads: claims ₹9.6 lakh in revenue on ₹3.1 lakh spend (3.1x ROAS).
- Sum of platforms: ₹23.8 lakh in claimed ad revenue.
- Warehouse truth: ₹39.1 lakh total net revenue on ₹7.8 lakh total marketing spend (5.0x blended ROAS).
Both ad platforms took credit for the same shoppers who clicked both ads. Run this warehouse query every Monday to track the gap between platform claims and actual cash received:
Use platform ROAS to decide which creative or ad set performs best inside that channel. Use blended ROAS to decide whether to raise or lower overall marketing budgets.
Reach and frequency cannot be added together
Reach and frequency look like simple audience measurements, but each platform calculates them inside its own walled garden:
- Reach counts distinct users within that platform. Meta counts logged-in Facebook and Instagram accounts. Google counts unique users across Search, YouTube, and the Display Network. Neither platform knows who saw an ad on the other network. Adding 500,000 Meta reach to 400,000 Google reach does not mean you reached 900,000 people.
- Frequency divides impressions by that network's reach. A frequency of 3.2 on Meta and 2.1 on YouTube does not mean your average customer saw 5.3 ads. Many saw ads only on Meta, some saw ads only on YouTube, and a core group saw both.
Treat reach and frequency as internal diagnostic metrics. Inside Meta, a frequency rising past 4.0 on a cold prospecting audience warns you of creative fatigue. But never combine frequency figures across different ad platforms in an executive report.
A practical Monday reconciliation checklist
You do not need to spend hours trying to make disparate platforms agree. Run this 5-step checklist every Monday in under an hour:
Monday reconciliation, under an hour never sum platforms
- Freeze windows on every ad set and conversion action to spot silent changes
- Split Meta click revenue from 1-day-view before comparing channel performance
- Compute blended ROAS first in the biggest font so leadership sees business truth
- Watch overclaim movement (±0.15 week-on-week) rather than its baseline level
- Write the one-paragraph bridge explaining the gap as overlap and organic demand
When writing your Monday summary, close with a single bridge paragraph:
"Warehouse revenue was ₹39.1L on ₹7.8L total marketing spend (5.0x blended ROAS). Meta claimed ₹14.2L and Google claimed ₹9.6L under their individual windows. The ₹15.3L difference reflects shared journey overlap, view-through impressions, and organic demand."
That short paragraph explains the discrepancy clearly and prevents pointless debates about which platform gets credit.
Frequently asked questions
Which platform's conversion number should I trust?
Neither platform should be used as the single source of truth for the whole business. Trust platform numbers for relative choices inside that platform—such as picking a winning ad variation. For company-wide performance and cash decisions, trust your warehouse and blended ROAS.
My agency changed Meta's window to 1-day click for accuracy. Is that a good idea?
It is more conservative, not necessarily more accurate. A 1-day click window ignores customer consideration journeys longer than 24 hours. For products with an order value above ₹2,000, shoppers routinely take several days to purchase. Keep the default window for campaign bidding and use blended metrics for executive reporting.
Why does Google show fractional conversions like 0.6?
Google Ads uses data-driven attribution by default. When a customer clicks three different Google ads before buying, Google splits credit among those clicks based on machine-learning weights. Meta gives 100% of credit to the final touchpoint. Do not round Google's fractional conversions.
How often should we run the reconciliation checklist?
Run the full checklist once a week on Monday morning. Review your attribution window settings quarterly or immediately after an agency handover, as that is when settings are most commonly altered without notice.
Takeaways for your next report
- Conversions and ROAS share identical names but use fundamentally different mathematical definitions across platforms.
- Changing an attribution window cuts or inflates reported numbers without changing real customer purchases.
- Summing platform revenues double-counts shared customer journeys; always use blended ROAS for executive reporting.
- Reach and frequency are internal platform diagnostics that cannot be added together across channels.
Related: how attribution windows shift revenue, how to calculate blended ROAS, and why blended CAC protects your cash flow.
MER Calculator
Total revenue divided by total marketing spend — the attribution-agnostic efficiency number, plus its contribution-margin-adjusted variant.
Chinmay Raibagkar
About author →Founder of DataLens AI. He helps non-technical teams read their ad and database numbers with confidence — which number to trust, what to do next, and what to ignore.