Glossary · MeasurementTOFU
Attribution
The short answer
Attribution is the method of assigning credit for a conversion to the marketing touchpoints that came before it. A customer might see an Instagram ad, search your brand on Google, read a review and click an email before buying. Attribution decides how much of that sale each touchpoint "earned".
Every model is a set of assumptions, and every ad platform uses a model that favours itself.
Common models
- Last click — 100% to the final click. Simple; over-credits retargeting and brand search.
- First click — 100% to the first touch. Over-credits discovery channels.
- Linear — equal credit to every touch.
- Time decay — more credit to touches closer to conversion.
- Data-driven — statistical credit based on observed paths; GA4's default.
- Platform attribution — each ad platform's own view, which can't see other platforms.
The fundamental limit
Attribution describes who was present, not who caused the sale. To learn what would have happened without a channel, you need incrementality testing. Use attribution to steer within channels and MER plus experiments to judge between them.
Attribution in practice
A customer sees a YouTube ad, later clicks a Meta ad, then searches the brand on Google and buys. Last-click attribution credits Google; data-driven attribution spreads credit; each platform may claim the sale. Blended metrics and incrementality tests help resolve the differences.
Common attribution models
- Last click
- First click
- Linear
- Time decay
- Data-driven
Attribution in India
Indian customer journeys often cross Instagram, YouTube, Google, WhatsApp and marketplaces — and many conversations happen on WhatsApp or phone, invisible to standard tracking. Offline conversion uploads and blended metrics fill the gaps.
Common attribution mistakes
- Adding up platform-reported conversions
- Ignoring WhatsApp and phone conversions
- Judging upper-funnel channels on last click
Frequently asked questions
Which attribution model is best?
None is "true". Data-driven models are usually better than last click for allocating within a platform, but no model replaces controlled experiments.
Why do Meta and Google both claim the same sale?
Each sees only its own touchpoints and applies its own window. Neither is lying; they're just counting from different vantage points.
What is an attribution window?
The time after a click or view during which a conversion is credited to the ad — for example, 7-day click or 1-day view.
Can attribution be perfectly accurate?
No. Some influence is never tracked. Use attribution for direction and tests for proof.
Why do platforms report more conversions than our store?
Each platform counts conversions it touched, so the same sale can be claimed several times.
What's the simplest reliable measure?
Blended MER or total CAC from your own data.
Can WhatsApp sales be attributed?
Yes, with click-to-WhatsApp tracking, CRM tagging and offline uploads.
Related terms
Full glossaryTalk to a strategist
Thirty minutes. Your numbers. A straight answer.
Book a strategy call with the people who'd actually run your account. We'll tell you what we'd do — or that you don't need us yet.