How to ask AI when your ROAS numbers don’t match
The same campaign can show different ROAS in Ads Manager, GA4, and your order system. Start by naming the revenue and spend sources behind each ratio.
On this page
When one campaign has different ROAS figures
Imagine preparing a weekly report. For the same campaign and dates, Meta Ads Manager shows ROAS of 4.2, GA4 shows 2.8, and your order-based calculation shows 3.5. These are illustrative figures. To choose a number for the report, first identify the revenue and spend behind each one.
Take a second illustrative example: Meta credits ₩4.2 million in revenue, Google ₩3.6 million, and Naver ₩1.8 million, while actual paid orders total ₩7 million. Multiple platforms may credit the same purchase, so attributed revenue can add up to more than sales. ROAS is a revenue-to-spend ratio; summing platform ratios does not produce overall ROAS.
Meta Ads Manager and GA4 need not assign the same revenue to Meta. Their windows, attribution methods, and observed or modeled coverage can differ. Check the account settings rather than assuming one must always be lower.
| Source | Question the ROAS answers | What it’s for |
|---|---|---|
| Meta Ads Manager | How much credit does Meta give its ads within its own attribution window? | Comparing campaigns and optimizing creative and targeting within Meta |
| Google Ads | What’s the cost vs. conversion value under the selected conversion action and attribution model? | Informing Google bidding strategy and campaign structure |
| GA4 | How much revenue does the selected reporting attribution model assign across the site and app journey? | Comparing channels on a single standard and reporting |
| Back office | What’s ad spend vs. revenue once you account for actual payments, cancellations, refunds, tax, and shipping? | Budget shifts and P&L decisions |
Align the definitions before comparing ROAS
Start with the attribution window and model. How long after a click or view can a purchase count, and how is revenue assigned to touchpoints? Record the actual settings used by each source.
The second is view-through credit and modeling. The numbers shift depending on whether you credit conversions where the user saw the ad but never clicked, and whether you use modeling to fill the gaps that privacy and browser restrictions leave behind. Google Ads says outright that it uses modeled conversions to cover attribution it can’t observe directly.
The third is how you define conversion value. Whether the value your pixel fires includes VAT or shipping, and whether it nets out cancellations and refunds, naturally drives a wedge between platform ROAS and back-office ROAS. Google Ads’ Target ROAS, after all, optimizes against the conversion value the advertiser reports to it.
Five prompts you can use right now
| Scenario | Bad question | Better question in Neuro |
|---|---|---|
| Weekly check-in | How’d the ads do this week? | Compare ROAS for [target account] from last Monday through Sunday with the preceding week. First confirm revenue and spend sources, currency, and time zone. If the team has no reporting definition, ask which one to use. Show the three comparable campaigns with the largest ROAS changes. |
| Diagnosing cross-channel gaps | Why don’t Meta and GA4 match? | For Campaign X over the same period, show Meta Ads Manager ROAS next to the ROAS GA4 allocates to the Meta channel. Then work through the gap in this order: attribution window, view-through credit, modeled conversions, deduplication, time zone, and whether refunds are included. |
| Budget decisions | Where should we put more budget? | First check whether orders can be reliably linked to campaigns. If so, compare the last 14 days of net-revenue ROAS against the target and identify candidates with verified budget constraints. If the linking evidence or target is missing, withhold a budget recommendation and state what data is needed. |
| Suspected tracking issues | Anything look off? | Using an illustrative threshold, find campaigns where GA4 ROAS fell at least 30% against the prior seven-day average while platform ROAS stayed stable. Flag small samples or incomplete reporting separately. Suggest events and time zones to check without concluding that tracking gaps caused the difference. |
| Client reporting | Build the report. | Build the client-report ROAS on our team’s GA4 reporting standard. Put platform ROAS in a separate reference table. If the platforms’ attributed revenue adds up to more than actual revenue, add a footnote flagging possible double counting. |
Mark which explanations you verified and which remain open. If definitions or data are missing, identify what to retrieve next instead of forcing the figures to agree.
Save the revenue definition for recurring reports
Save recurring revenue definitions in a Gem or team instructions. For example: use net revenue after refunds for budget review, and platform ROAS for comparisons within that platform. If there is no supported way to link net revenue to campaigns, explicitly prohibit allocating total sales by guesswork.
Questions that need more data
AI can’t settle every ROAS gap with certainty. There’s no way to audit, from the outside, how accurate each platform’s internally adjusted modeled conversions are. AI can say “this gap could be down to modeling,” but it shouldn’t claim that some exact percentage of revenue is the result of modeling.
Without order and refund data, you cannot calculate net-revenue ROAS. Compare what is available and list the missing inputs. Even a verified campaign-level figure does not, on its own, explain margins or repeat-purchase value.
A prompt for the first comparison
Sources
- Google Ads Help: About attribution models
Official help article explaining Google Ads attribution models and how to compare ROAS across models in the Model comparison report.
- Google Analytics Help: Select attribution settings
Official documentation for GA4 on configuring reporting attribution models, crediting channels, and setting lookback windows for key events.
- Google Ads Help: About modeled online conversions
Explains how Google Ads uses modeling to fill conversion links that can’t be directly observed due to privacy and technical limitations.
Related posts
When to start a new chat for ad work
Decide whether to continue or split an ad-analysis chat when accounts, goals, and reporting assumptions change. Includes prompts for carrying verified findings into a new task.
When AI groups campaigns incorrectly: check names and UTMs
Find campaigns whose names leave product groups or audiences unclear, then use settings, UTMs, and a mapping table to classify them.
How to use Gems in Neuro for ad operations
Create a Neuro Gem with reusable analysis and reporting instructions, then add the dates and account details for each task.
Five areas to check in an AI-assisted ad account review
Review goals, campaign structure, budget, conversion quality, and creative while separating verified findings from causes that need more evidence.
Before AI changes your ad budget
How to review AI proposals for budget and bid changes, including approval settings and scheduled execution in Neuro.