What Is Attribution? (Marketing Attribution)
The set of rules that decides which ad touchpoint gets credit for a conversion.
Book a free callAttribution is the process of determining which marketing touchpoint an install or purchase came from. Because users may see several ads before converting, an attribution model decides which touchpoint gets the credit. Since where your budget goes depends directly on this model, your attribution setup is the foundation of your marketing decisions.
How does attribution work?
The user journey rarely has a single touchpoint: someone watches a video on Instagram, sees another ad on TikTok two days later, then searches for the app on Google the next day and installs it. All three channels contributed, but most systems pick a single winner.
Two parameters drive that choice: the attribution model (who gets the credit) and the attribution window (the maximum time allowed between a touchpoint and the conversion).
Attribution models
| Model | Who gets the credit | When it fits |
|---|---|---|
| Last click | The last click before the conversion | The mobile default; simple and repeatable |
| First click | The first touchpoint | When evaluating discovery and brand awareness channels |
| Linear | Equal credit to every touchpoint | For long, multi-touch journeys |
| Time decay | More credit to touchpoints closer to the conversion | For short decision cycles |
| Data-driven (DDA) | An algorithm calculates each touchpoint’s contribution | When you have enough conversion volume |
The de facto standard in the mobile app ecosystem is last click. Its advantage is that it’s simple and neutral; however, it systematically undercounts the contribution of channels that feed the upper funnel. That’s why mature teams use last click for operational decisions while validating channel value with incrementality tests.
Attribution windows and view-through attribution
The attribution window is the maximum amount of time during which a touchpoint can receive credit. The common mobile standard is 7 days for clicks and 24 hours for impressions.
View-through attribution (VTA) gives credit when a user only sees an ad without clicking it. It captures the contribution of video and display campaigns, but it’s open to abuse: very long VTA windows assign installs that were actually organic to paid channels, making performance look better than it is.
Rule of thumb: use the same windows across all channels. Using a 7-day VTA window on one channel and a 1-day window on another makes comparison meaningless.
Attribution on iOS after the privacy changes
With Apple’s ATT framework and SKAdNetwork, user-level deterministic attribution on iOS has largely come to an end. Data now arrives aggregated, delayed and with limited granularity. In this environment, three approaches are used together:
- SKAN postbacks: Anonymous, campaign-level signals provided by Apple.
- Deterministic data: Classic attribution for users who grant ATT permission.
- Modeling and incrementality tests: Measuring a channel’s true contribution with geo-based on/off tests.
Media mix modeling (MMM) has regained importance now that granular attribution has weakened; it’s used to validate channel contribution at a high level, especially with large budgets.
Frequently asked questions
Why don’t ad dashboard numbers match MMP data?
Each platform uses its own attribution rules and windows, and tends to claim credit for itself. The same conversion can be counted in more than one dashboard. That’s why cross-channel comparisons should be done on independent MMP data, with dashboard data used only for in-campaign optimization.
Which attribution model should I choose?
For mobile app campaigns, last click is a practical starting point: simple, neutral and applicable across every channel. But don’t judge upper-funnel channels by this model alone; validate their true contribution periodically with incrementality tests.
How long should the attribution window be?
It depends on your product’s decision cycle. A short window is enough for a game people download on impulse; a wider window makes sense for a finance app with a long evaluation period. What’s critical is using the same window across all channels to keep them comparable.
Let’s clarify your attribution setup
We build a consistent measurement framework that’s comparable across channels and allocate your budget based on true contribution.
Book a free call