A budget discussion can become a competition between dashboards. Google Ads reports one return, paid social another, and GA4 a third. Choosing the highest number makes the decision look easy. Choosing the lowest can look cautious. Neither approach explains what the next unit of spend is likely to deliver.
My recommendation is to start with a Finance-agreed view of net delivered revenue and contribution, then use attribution reports and experiments to decide where to invest. The business measure sets the commercial boundary. It cannot, by itself, tell us which channel caused the sale.
Why I would revisit the definitions now
Google's 13 October 2026 Demand Gen update announced a Data Strength Uplift metric launching in Google Ads, Data Manager expanding into GA and DV360, and enhanced conversions coming to those products. The announcement described rollout stages, not universal availability. Google's September explanation describes the uplift metric as additional conversions recovered through the first-party data setup. That makes it useful evidence about measurement coverage; it does not, alone, establish additional delivered sales. Google's October update, September measurement announcement.
Meta has also changed how it describes credit. Its 3 March 2026 announcement said click-through attribution for website and in-store conversions would move to link clicks, with other social interactions moving into engage-through attribution. It described a phased rollout beginning later that month. I would check the account's actual implementation date before comparing reporting periods. This is an earlier announcement rechecked this week, not a new October launch. Meta's official announcement.
My interpretation is straightforward: a reported improvement may contain better observation, a changed credit rule and a real performance change. The job is to separate them.
Give each number a clear job
I would organize a budget review around three measurement lanes.
Campaign optimization. Keep Google Ads, Meta, TikTok and Snapchat results visible within their own reporting rules. They help teams assess bidding, creative and delivery. Label the conversion action, attribution window, click or view basis and source of conversion value. Adding platform-attributed revenue across channels is not a dependable business revenue total.
Journey and attribution diagnosis. Keep GA4 and Adjust/MMP in separate panels. GA4's attribution model affects event-scoped reporting differently from session-scoped reporting. Adjust's install attribution windows concern the interval between an engagement and an install. Matching the label on two columns does not make their definitions equivalent. Document event, scope and window before interpreting the gap. GA4 attribution settings, Adjust attribution windows.
Commercial accountability. Agree the net delivered revenue definition with Finance, including cancellations, returns, discounts and VAT treatment. Review it alongside contribution and operating constraints. A business revenue total divided by all media spend can describe overall efficiency, but it should not be presented as channel-attributed ROAS or proof of advertising causality.
Give each number a clear job.
Choose the decision. Keep each source's reporting rules visible.
Start with the commercial outcome. Agree delivered revenue, contribution and operational limits with Finance. Attribution and experiments help decide where to invest.
Guide the campaign
Read each platform separately.
Use for
Bidding, creative and delivery decisions.
Keep visible
- Conversion action and value source
- Click, view or engagement basis
- Attribution window and reporting date
Limit
Attributed return does not establish profit or causality.
Explain the difference
Event, session and channel reporting.
App acquisition and attribution rules.
Use for
Investigating channel, funnel and tracking differences.
Limit
Keep GA4 and MMP results in separate panels. Do not average them.
Set affordable scale
Net delivered revenue and contribution.
Use for
Commercial limits, realized value and operating capacity.
Agree the definition
- Cancellations and returns
- Discounts and VAT treatment
- Revenue date and adjustment lag
Limit
Business totals do not assign causal credit to individual channels.
Specify the outcome and evaluation period. Check holdout or geo-test feasibility. Use the result with commercial guardrails and measurement diagnostics.
Text takeaway: platform reports guide optimization; GA4 and Adjust/MMP diagnose different parts of the journey; Finance and order records define business value. Experiments help evaluate additional impact.
| Measurement lane | Source to keep visible | Decision it supports | Limit |
|---|---|---|---|
| Campaign optimization | Each ad platform separately | Bid, creative and delivery changes | Attributed return is not proof of profit or causality |
| Journey diagnosis | GA4 | Channel and funnel investigation under stated reporting settings | Scope and attribution settings affect interpretation |
| Journey diagnosis | Adjust/MMP, separately from GA4 | App acquisition and attribution investigation | Install and reattribution rules need their own definitions |
| Commercial accountability | Finance and order records | Affordable scale, delivered value and contribution | Does not assign causal credit to individual channels |
Visual and table: my decision framework, informed by the linked documentation. They contain no company results, benchmark values or invented conversion totals.
When reported ROAS moves, investigate before reallocating
Imagine a tracking or attribution change is followed by a rise in platform-reported return while delivered revenue remains broadly unchanged. This is a hypothetical diagnostic situation, not a Nahdi result.
I would first inspect the change history, conversion actions, reporting dates and conversion lag. I would also check whether the report now includes a different class of conversion.
Google's current Demand Gen documentation provides a concrete reason to do that. It describes view-through conversion optimization as an open beta, applicable to video assets across YouTube, Display and Discover Feed. When enabled, eligible view-through conversions enter the main conversion reporting used for bidding. That can affect comparisons with third-party reporting. Before treating a jump as growth, I would segment by ad event type and compare equivalent definitions. Google Ads documentation.
An explained gap can be acceptable. An unexplained gap should remain visible, with an owner and a next check.
The next budget decision needs a testable claim
For a substantial reallocation, I would write down the proposition: the proposed spend change should create additional delivered value at an acceptable contribution level.
Then I would choose a suitable holdout or geo experiment, agree the outcome and evaluation period in advance, and account for conversion delay. Google documents Conversion Lift as an incrementality tool, but access is not available to every account. Account eligibility and study feasibility need checking before promising a test. Google Conversion Lift guidance.
For the next review, my practical action is a one-page measurement agreement: the business KPI, the conversion value used for bidding, each source's reporting rules, and the evidence required to increase spend. Assign the marketing, analytics and Finance owners together.
That agreement gives the team a shared basis for deciding what to change, what to investigate and what still needs proof.
Source review: 27 October 2026. This is a public-source perspective and a proposed working method. It contains no private employer figures or claims about a specific account's results.
If your team is working through conflicting performance reports, contact ahmed@galal.app or explore analytics and measurement support.