Shopify, Meta, Google Ads and GA4 Show You Four Different Revenue Numbers. Which One Can You Trust?

Why your dashboards show different revenue, which number should drive which decision and when your tracking is actually broken.

Jason Spizert

Jason Spizert

Updated on September 13, 2026

Published 4 min read

The short answer

Every platform credits purchases to itself by its own rules: if someone clicks a Meta ad first and later buys via Google, both can claim the same purchase. Shopify remains the reference for what was actually ordered. For decisions within a platform, use that platform's number; for the big picture, use total revenue relative to total ad spend.

Imagine you want to decide tomorrow whether Meta gets more budget. Shopify shows €31,840 in revenue, Meta €24,600, Google Ads €13,900 and GA4 €27,200.

The first reflex: one of them must be wrong. Not necessarily. Meta doesn't tell you how much revenue your business made, but how many purchases it credits to itself by its rules. Every system counts differently:

The same purchases, four systems, four ways of counting
Purchases in your storeone month, the same store
ShopifyReferenceCounts what was actually ordered.€31,840
GA4MeasurementMeasures what visitors do on the way to purchase.€27,200
Meta AdsAttributionCredits itself with purchases by Meta's rules.€24,600
Google AdsAttributionCredits itself with purchases by Google's rules.€13,900
Example month from this article. None of the four systems is broken; each answers a different question. Vendor logos in original colors.

Different numbers aren't the problem. The problem is making the wrong decision with the wrong number.

Which number to use for what

For each decision, take the number from the system that is meant to answer exactly that question. That's all you need to start.

How much was actually ordered?ShopifyYour orders and your clearly defined revenue figure.
Which Meta campaign performs better?Meta AdsCompare Meta with Meta, not Meta with Shopify.
Which Google campaign performs better?Google AdsCompare Google with Google.
How do visitors move through the store?GA4For behavior, paths and trends, not as your ledger.
How is paid doing overall?Total revenue ÷ ad spendThe simple big picture, before you get lost in channels.
AI image: a green package on a light background, lit by several light sources, casting shadows in different directions and lengths
One purchase, several perspectives: each system casts its own shadow on the same order. (Illustration, AI-generated)

Why Meta + Google can report more than your store sold

A customer clicks your Meta ad on Tuesday. Two days later, she googles your brand name, clicks your Google ad and buys for €120. Meta counts this purchase for itself, and so does Google. Shopify has one order.

One journey, three bookings
TuesdayClick on the Meta ad
ThursdayClick on the Google brand ad
ThursdayPurchase · €120
Meta counts €120, the click is within the 7-day window
Google Ads counts €120, last click
Order list: 1 purchase, €120
One €120 purchase, recalculated
The platforms report
Meta Adsclick on Tuesday€120
Google Adsclick on Thursday€120
Reported in total€240
Shopify counts
1 order€120
€240 reported, €120 ordered. Never add up platform revenue.
Both platforms count according to their rules, each for itself. That's why Meta, Google and Shopify don't have to arrive at the same number.

If Meta and Google together report more than your store sold, you haven't discovered a new business model. For steering, that means: compare Meta with Meta, Google with Google. That's also how our teams for Meta Ads and Google Ads work.

When you really should take a closer look

Different numbers alone are normal. It gets interesting when something suddenly becomes implausible.

Discrepancy or defect? One question decides
Do you see one of these four warning signs?01Shopify has orders, but an ad account suddenly reports no purchases at all.02A purchase is visibly counted twice.03The gap jumps massively after a technical change.04Numbers are mathematically impossible or change for no apparent reason.
NoNormal discrepancyEvery system counts by its own rules. Use the right number for each decision; there's nothing to fix in your tracking.
YesTechnical checkDoes the purchase event fire correctly?Is it reported twice?Is data missing?Are sources lost along the way?
Different numbers alone are no proof of a defect. Only when something becomes implausible is it worth looking at the setup.

These are real tracking questions. Left unchecked, they lead you astray: if your reporting doesn't count orders, you go hunting for a checkout problem that doesn't exist, and every A/B test measures on shaky ground.

Do you need Triple Whale, Hyros or WeTracked for this?

Not automatically. If you mainly run Meta and Shopify works cleanly, built-in tools take you a long way. Triple Whale, Hyros or WeTracked help when several ad channels grow in parallel. They don't decide for you which number counts for what.

Where customers heard about you is often revealed by a simple post-purchase survey: no cookies, no guesswork.

Order: problem →︎ built-in tools →︎ decision →︎ tool
Built-in tools first, free
Events ManagerChecks whether purchases and other events arrive cleanly at Meta.
GA4 DebugViewShows you live which events are arriving in GA4 right now.
GTM / Tag AssistantHelps you check which tracking tags fire on your site.
MER sheetCompares your total revenue with your total ad spend.
Double counting, purchase event, attribution window, timeline: you can check all of it here before you buy any tool.
The customer's side of the storyThis isn't the perfect truth either, since people don't always remember correctly. But it's a helpful counterpoint to your tracking data.
The links to WeTracked, Hyros and Zigpoll are affiliate links (advertising). The assessment above applies regardless.

What I would do on Monday morning

Monday morning: five steps, one number per question
01 · BaselineShopify: what was actually ordered?
02 · MetaMeta numbers only for Meta decisions.
03 · GoogleGoogle Ads numbers only for Google decisions.
04 · Big pictureTotal revenue ÷ total ad spend. That's your blended ROAS.
05 · TrackingOnly dig into the tech when something is genuinely off.
The goal isn't to force four dashboards onto the same number, but to know which one you listen to for each decision.

Why the sum of platform values regularly exceeds your actual revenue is calculated step by step in ROAS: formula, calculation and why the number misleads. Clean numbers are also the foundation of any conversion optimization.

If you don't want to run the checks yourself: exactly these steps, from matching attribution windows to checking for double counting, are part of our store audit, including a documented finding for each step.An audit only pays off once you steer meaningful ad spend with numbers you don't trust. Before that, this article is enough.
  1. 01Matching attribution windows, date ranges and timelines
  2. 02Checking the purchase event and double counting (pixel/CAPI)
  3. 03Documented finding per check: normal, broken or open
Request a store auditNumbers are right, but no sales? The diagnostic order in “Shopify traffic but no sales”

Frequently asked questions

Because Meta credits purchases to itself by its own rules. A customer can click a Meta ad first and buy later via Google. Then both platforms can credit themselves with the same purchase, even though Shopify of course only received one order.

For decisions within Meta, use Meta; within Google Ads, use Google Ads. For the question of how your marketing is doing overall, look at your total revenue relative to your total ad spend. Shopify remains the reference for what was actually ordered.

Not necessarily. Tools like these become interesting when you run several channels in parallel and need a shared view. They help you put things in context, but they are not a magic referee that suddenly knows the one perfect truth.

When purchases are missing, counted twice, numbers suddenly jump sharply or become mathematically impossible. Then you should have the setup checked technically. Different numbers alone are not proof of a defect.

See all insights

  • Reply within 24 hours
  • No obligation, no sales pitch

Request an intro call

How can we help?*
What budget do you have in mind?*

By submitting, you agree to the processing of your data as described in our privacy policy.

Rather pick a time directly?

Book a call directly
Patricia from the NextPercent team working on a laptop in an armchair

Jennifer Rosbach

Beauty brand (nagelati.de)

“As a young brand, it mattered to us to work with a partner who doesn't just execute, but really understands how ecommerce brands grow. That's exactly what we found. Clear communication, honest assessments and a good feel for conversion and brand presence.”