Attribution Windows Are Skewing Your Decisions
Different attribution windows can make the same campaign look like a hero or a flop. To compare channels fairly, lock the window, measure incrementality, and allocate budget by margin.
Meta Ads shows a ROAS of 4 while GA4 puts the same campaign at 1.8, and you can't decide which one to trust. The problem isn't channel performance—it's that each platform uses a different attribution window and model. In this article you'll find a concrete way to compare channels on equal footing in 2026, isolate the window effect, and allocate budget by margin.
First, Lock the Attribution Window
The attribution window is the lookback period that determines which touchpoint gets credit for a conversion. Google Ads uses a 30-day click window by default; Meta Ads typically starts with 7-day click + 1-day view; in GA4 it varies by report, from 30 to 90 days. The same user journey gets credited to different channels under different windows. So first level the playing field: pick a single window across all channels and don't change it for at least 8 weeks.
Practical thresholds for choosing a window
- Short cycle (0-7 days): Low-ticket e-commerce, impulse purchases. A 7-day click window is enough.
- Medium cycle (8-30 days): Considered-purchase products, subscription starts. Use a 30-day click window.
- Long cycle (31-90 days): B2B, high-ticket services. 90-day click + 1-day view, but report view-through credit separately.
Once the window is locked, compare channels by reading ad management and GA4 data side by side. Don't make platform ROAS your sole decision metric.
Isolate the Window Effect with a Formula
To see the gap different windows create, run a simple breakdown. Pull the same campaign's 7-day and 30-day conversion counts, and mark the difference as the window contribution:
Window contribution (%) = (Conversions30d − Conversions7d) / Conversions7d × 100
Example: a brand gets 400 conversions in 7 days and 520 in 30 days. Window contribution = (520 − 400) / 400 × 100 = 30%. That means 30% of conversions are tied to touchpoints between days 8 and 30. If this gap is above 20%, don't cut budget based on the 7-day window—otherwise you'll shut down upper-funnel channels too early.
| Window | Conversions | CPA ($) | ROAS | Comment |
|---|---|---|---|---|
| 7-day click | 400 | 250 | 2.4 | Bottom-funnel weighted |
| 30-day click | 520 | 192 | 3.1 | Includes upper-funnel credit |
| 30-day click + 1-day view | 610 | 164 | 3.6 | View-through credit must be separated |
In this table spend is held constant at $100,000; the only thing changing is the window. Write down which row you base your decisions on. Define a single "decision window" for the team.
Don't Trust the Window Without Measuring Incrementality
The window distributes credit; incrementality measures the incremental conversions advertising actually creates. In 2026 there are two practical methods:
- Geographic holdout: Turn off the campaign in 10-15% of regions for 4 weeks and compare the conversion gap with the remaining regions. The difference gives an estimate close to the incremental effect.
- PSA (public service announcement) test: On Meta, allocate 5-10% of budget to a neutral message instead of your ad; the conversion rate gap between the two groups shows incrementality.
Repeat the incrementality test at least twice a year. Combine the results with channel-level margin within a 360° digital marketing framework, so you see attributed credit and real contribution in the same table.
Allocate Budget by Margin and Window
Once the attribution window is locked, tie budget decisions to margin. Calculate contribution margin, then derive target CPA from it:
Target CPA = Contribution Margin ($) × Acceptable Share (%)
Example: if a product's contribution margin is $60 and you allocate 30% to marketing, target CPA = $18. Even if platform CPA looks like $16.4, if CPA rises to $19.2 when the window extends to 30 days, you're above target. In that case, either improve creative/landing page or shift budget to a lower-CPA channel. Read funnel steps against shared data from SEO and paid channels; place organic search's upper-funnel contribution in the same window.
Weekly checklist
- Is a single window used across all platforms?
- If window contribution exceeds 20%, has the budget-cut decision been postponed?
- Is channel CPA below target CPA?
- Has an incrementality test been run in the last 6 months?
- Is view-through credit reported separately?
Once you establish this discipline, cross-channel comparison stops being guesswork; every decision is made on the same footing. If you need to connect your measurement infrastructure with your SaaS or app side, our SaaS product development team can help you build the event schema together.
Would you like to see how much the attribution window is skewing your budget decisions in your own account? In a free discovery call we can put your current platforms on the same footing and map out a 30-day measurement plan. Book a slot on the contact page and let's take a look together.