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.
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.
- Players = channels; value = average marginal contribution
- Order-independent and 'fair' under game-theory axioms
- Computationally heavy; usually approximated in practice
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
- Assuming Shapley credit is exact when the channel set is incomplete.
- Ignoring that undertracked channels get understated values.
- Treating a fair split as proof of causal lift without experiments.
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
- Data-driven attribution: promise and caveats
Data-driven attribution (DDA) assigns credit using a model trained on a site's own conversion paths rather than a fixed rule like last-click. Done well it credits assist touches more fairly. Its caveats are real: it needs enough conversion volume, it is a model not a measurement, and it cannot see touches that were never tracked.
- Markov chain attribution
Markov chain attribution models customer journeys as a probabilistic graph of transitions between channel states, ending in conversion or null. Each channel's credit is derived from its 'removal effect' — how much the overall conversion probability falls if that channel (and its transitions) are removed from the graph. It is a leading algorithmic alternative to Shapley-based attribution.
- Algorithmic vs rules-based attribution
Attribution models split into two families. Rules-based models apply fixed, human-chosen weights to touchpoints by position — last-click, first-click, linear, time-decay, U/W-shaped. Algorithmic (data-driven) models learn credit from observed conversion paths using methods like Shapley values or Markov chains. This page contrasts the two and explains when each is appropriate.
- Attribution analytics
First-party channel-presence data to reason about contribution.
Sources and verification notes
- Google Analytics Help — Data-driven attributionDescribes data-driven attribution using an algorithmic approach to distribute credit; Shapley value is the game-theory basis of this family.
- Shapley value — concept (cooperative game theory)Definition of the marginal-contribution averaging method.
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.