Linear Attribution Model

The linear attribution model splits credit for a sale or conversion equally across every touchpoint a customer interacted with. If a journey contained four interactions, each receives 25% of the credit, regardless of when it happened or how influential it was.

How it works

Suppose a customer clicks a paid social ad, later opens an email, returns via organic search, and finally converts through a branded paid search click. Under linear attribution each of those four channels is credited with a quarter of the order value. With a €200 order, every channel is assigned €50.

The calculation is simply the conversion value divided by the number of recorded touchpoints, which makes the model easy to explain to stakeholders and easy to audit.

When it is a reasonable choice

• Long consideration journeys — for considered purchases with many interactions, equal weighting is less distorting than crediting one moment.
• Full-funnel visibility — it surfaces upper-funnel channels that a last-click view would show as contributing nothing.
• A neutral starting point — when you have no evidence about which position in the journey matters most, equal weighting encodes fewer assumptions than a decay curve.

Where it misleads

The model's central assumption is usually false: touchpoints are not equally influential. A fleeting display impression receives the same credit as the product-page visit where the decision was actually made. This has two practical consequences.

First, channels that appear frequently but persuade little — retargeting and broad display in particular — accumulate credit in proportion to their impression volume rather than their effect. Second, because credit scales with the number of recorded touchpoints, journeys with more tracking coverage look different from journeys with less, even when the underlying behaviour is identical.

Linear attribution also inherits the limitation shared by every attribution model: it can only divide credit among interactions that were recorded, and it cannot distinguish sales that marketing caused from sales that would have happened regardless.

What to use alongside it

Treat linear attribution as a description of journey composition rather than a measure of causal impact. To establish what spend actually caused, use incrementality testing; to measure the contribution of the whole mix without relying on user-level tracking, use Marketing Mix Modeling.