What Is a Look-Back Window?
A look-back window (also called an attribution window) is the defined time period during which an ad platform can credit a conversion to a prior ad interaction. If a user clicks or views an ad, then converts within the window, the platform assigns credit to that ad. Conversions that occur outside the window receive no attribution to that touchpoint.
Look-back windows are a core setting in any attribution model. They determine which ads appear to drive results, which affects budget allocation, optimization signals, and reported ROAS.
Click-Through vs. View-Through Windows
Most platforms offer two separate look-back windows that operate in parallel.
Click-Through Window
The click-through window starts when a user clicks an ad. A 7-day click-through window means the platform credits that ad for any conversion within 7 days of the click. Standard industry defaults range from 7 to 30 days.
View-Through Window
The view-through window starts when a user sees an ad but does not click it. If they convert within the window (commonly 1 day), the impression receives credit. View-through attribution is more contested because the causal link between an impression and a conversion is harder to establish.
| Platform | Default Click Window | Default View Window |
|---|---|---|
| Meta Ads | 7 days | 1 day |
| Google Ads | 30 days | 1 day |
| TikTok Ads | 7 days | 1 day |
| Pinterest Ads | 30 days | 1 day |
| LinkedIn Ads | 30 days | 7 days |
How Look-Back Windows Affect Reported Performance
The window length directly inflates or deflates conversion counts. A longer window captures more conversions per ad, making campaigns appear more efficient. A shorter window reduces attributed conversions, raising apparent CPA.
Example Calculation
Consider a direct-to-consumer brand running Meta Ads. Under a 7-day click / 1-day view window, the campaign reports 400 conversions at a $25 CPA. Switching to a 1-day click / 0-day view window produces a different picture. That tighter setting, closer to what privacy-restricted measurement actually returns, may show only 180 conversions and a $55 CPA from the same spend.
The formula for comparing windows:
Attribution Inflation Rate = (Conversions at Longer Window / Conversions at Shorter Window) – 1
Using the example above: (400 / 180) – 1 = 1.22, or 122% inflation. Neither number is “wrong,” but they measure different things.
Why Window Length Is a Strategic Choice
The appropriate window depends on the purchase cycle of the product. A $12 impulse purchase and a $1,200 software subscription have fundamentally different decision timelines.
- Short windows (1-3 days): Suited for low-consideration purchases, flash sales, or app installs. They reduce noise from organic intent and produce conservative conversion counts.
- Medium windows (7 days): A common default for e-commerce. Covers weekend research-to-purchase behavior without over-crediting ads from two weeks prior.
- Long windows (14-30 days): Better for considered purchases like furniture, travel, or B2B software. A user researching an enterprise tool may take three weeks from first ad exposure to demo request.
The Cross-Platform Double-Counting Problem
When running campaigns across Meta, Google, and TikTok simultaneously, each platform claims credit for conversions using its own window. A single customer may click a Google Shopping ad on Monday, see a Meta retargeting ad on Wednesday, and purchase on Thursday. Google credits the conversion under its 30-day window. Meta credits it under the 7-day window. Both platforms report it. The brand’s actual conversion count is 1.
This overlap is why platform-reported conversions routinely exceed conversions recorded in a brand’s own analytics or multi-touch attribution system. Industry estimates suggest platform-reported conversions can run 20% to 60% higher than independently verified conversions, depending on channel mix and window settings.
Marketers at companies like Glossier and Warby Parker have publicly discussed aligning all platform windows to match their customer data platform as a way to create comparable cross-channel metrics.
Look-Back Windows After Privacy Changes
Apple’s App Tracking Transparency (ATT) framework, introduced in iOS 14.5 in April 2021, sharply limited look-back windows for mobile. The SKAdNetwork framework Apple enforces limits conversion reporting to a postback window of roughly 24 to 72 hours with heavy aggregation and delays.
Meta responded by shifting its recommended comparison benchmark from 7-day click to 7-day click with modeled conversions. Statistical modeling estimates conversions that cannot be directly observed under the new privacy constraints. This creates a gap between raw attributed conversions and modeled totals that can confuse reporting if not clearly labeled in dashboards.
For web campaigns without mobile identity restrictions, look-back windows remain configurable, but third-party cookie deprecation has reduced the reliability of long windows. A 30-day cookie-based window depends on the cookie surviving long enough to match the conversion event. In many browsers, cookies are cleared or blocked within 7 days, making the longer window unreliable in practice.
Setting Windows Consistently Across Tools
The most common error is misaligned windows between the ad platform and the analytics layer. If Google Analytics uses a 30-day attribution window while the Google Ads account reports on a 7-day window, the two systems will produce different conversion counts for the same campaign. That mismatch makes performance reviews harder than they need to be.
Best practice is to:
- Define a standard window based on actual purchase cycle data from CRM records.
- Set identical windows in all active ad platforms.
- Mirror that window in the analytics or BI layer used for reporting.
- Document the window in all shared dashboards so stakeholders know what the numbers reflect.
Look-Back Window vs. Attribution Model
These two concepts are often confused. The look-back window defines when a touchpoint is eligible for credit. The attribution model defines how much credit is distributed among eligible touchpoints. A last-click model and a linear model can operate on the same 7-day window but produce different credit distributions across channels.
Adjusting the window changes the eligible touchpoint pool. Adjusting the model changes how credit is split within that pool. Both decisions affect reported channel performance independently.
Frequently Asked Questions
What is a look-back window in digital advertising?
A look-back window is the defined period during which an ad platform can credit a conversion to a prior ad interaction. It sets the time boundary for attribution: conversions inside the window are credited to that ad, conversions outside it are not.
What is the default look-back window for Meta Ads?
Meta Ads defaults to a 7-day click and 1-day view look-back window. That means conversions within 7 days of a click, or within 1 day of an impression, are credited to the ad.
Why do different platforms report different conversion totals for the same campaign?
Each platform tracks conversions using its own window and credits them independently. When a customer sees ads on multiple platforms before buying, each platform claims the same conversion. A brand’s actual conversion count can be a fraction of what platforms collectively report. A unified measurement layer, such as a multi-touch attribution system or customer data platform, is needed to reconcile the totals.
How did Apple’s privacy changes affect look-back windows?
Apple’s App Tracking Transparency (ATT) framework, introduced in iOS 14.5 in April 2021, limits mobile conversion reporting to roughly 24 to 72 hours through the SKAdNetwork framework. This compressed effective mobile attribution windows and pushed platforms toward modeled conversion data rather than directly observed data.
What is the difference between a look-back window and an attribution model?
A look-back window defines when a touchpoint is eligible for credit. An attribution model defines how much credit is distributed among those eligible touchpoints. Both settings affect reported channel performance, but they operate independently: the same window can run under last-click, linear, or data-driven attribution and produce different results.
Key Takeaways
- A look-back window sets the time boundary for crediting an ad interaction to a conversion.
- Longer windows inflate conversion counts; shorter windows deflate them relative to each other.
- Cross-platform overlap causes double-counting that only a unified measurement layer can resolve.
- Privacy changes have compressed effective mobile windows to near real-time, increasing reliance on modeled data.
- Consistent window settings across platforms and analytics tools are necessary for reliable performance comparison.