WebmasterID logoWebmasterID
Attribution models

Shapley value attribution

Shapley value attribution applies a concept from cooperative game theory: it treats channels as players in a coalition and assigns each one credit equal to its average marginal contribution across all possible orderings of channels. The result is a principled, order-independent way to split conversion credit. It underpins data-driven attribution in several analytics products.

Verified against primary sources

What this means

The Shapley value comes from cooperative game theory (Lloyd Shapley, 1953). In attribution, the 'game' is producing a conversion and the 'players' are the marketing channels. A channel's Shapley value is its average marginal contribution: for every possible ordering in which channels could be added to the coalition, you measure how much the conversion probability rises when that channel joins, then average those marginal gains.

This makes the credit order-independent and provably 'fair' under axioms like efficiency (credits sum to the whole) and symmetry (channels that contribute identically get equal credit).

Why teams use it and its costs

Shapley-based attribution avoids the arbitrariness of positional rules: it derives credit from observed combinations of channels rather than from where a touch sits. Google's data-driven attribution is described as using an approach in this family.

The cost is complexity. The number of channel orderings grows factorially, so practical implementations approximate or restrict the channel set. Results are also only as trustworthy as the data: if certain channels are undertracked (offline, walled-garden, or consent-blocked touches), their Shapley values are biased downward.

How it appears in analytics and logs

Credit that reflects how much each channel adds when present versus absent — averaged over orderings — indicates a Shapley-based model; a channel's value reflects its marginal lift across coalitions, not its position.

Diagnostic use case

Use Shapley value attribution when you want a theoretically grounded split of credit that accounts for how channels combine, rather than a fixed positional rule.

What WebmasterID can help detect

WebmasterID gives you first-party channel-presence data per path, the raw input a Shapley-style computation needs, so you can reason about marginal contribution without third-party tracking.

Common mistakes

Privacy and accuracy notes

Shapley attribution operates on aggregated path data — which channels appeared and whether a conversion followed — not on individual identity. It is a statistical computation over coalitions.

Related pages

Sources and verification notes

Last reviewed 2026-06-24. Facts are checked against primary/official sources where available; uncertain specifics are marked “Data not yet verified” rather than guessed.