Attribution

The practice of assigning credit for a conversion or revenue event to the marketing touchpoints that influenced it. Attribution models determine which channels, campaigns, or interactions get credit for driving outcomes.

Attribution answers a question every marketing leader asks: which of our efforts are driving results? When a customer converts, attribution models trace backward through the touchpoints they encountered and assign credit to each one.

The models range from simple to complex. Single-touch models (first-touch, last-touch) give all credit to one interaction. Multi-touch models distribute credit across multiple touchpoints using rules (linear, time-decay, position-based) or algorithms (data-driven attribution). Each model makes different assumptions about how influence works, and each produces different answers from the same data.

Without a model, allocation is guesswork

Without attribution, budget allocation is guesswork. You know how much you spent on paid search, email, and display. You know how many conversions happened. But you cannot connect the two without a model that maps touchpoints to outcomes. Attribution provides that map, however imperfect.

Attribution also creates accountability. When channels can be measured against outcomes, teams can optimize toward what works and reduce investment in what does not. The alternative is allocating budget based on precedent (“we have always spent this much on display”) rather than evidence.

Models as tools, not truth

The first mistake is treating any attribution model as truth. Every model is a simplification of a complex, multi-channel, multi-device buying journey. Last-touch over-credits the final interaction. First-touch over-credits the entry point. Even data-driven models carry assumptions about what counts as a touchpoint and how influence decays over time. The goal is not perfect accuracy. It is directionally useful decision-making.

The second mistake is relying on attribution alone for measurement. Attribution tracks individual-level paths. Media mix modeling measures aggregate channel effects. Incrementality testing measures causal impact. The strongest measurement strategies triangulate across all 3 methods rather than treating any single approach as definitive.

Frequently Asked Questions

What is the difference between first-touch and last-touch attribution?

First-touch gives all credit to the first interaction that brought a customer into your funnel. Last-touch gives all credit to the final interaction before conversion. Both are single-touch models that ignore everything in between. Neither reflects how buying decisions work.

Is multi-touch attribution dead?

Multi-touch attribution (MTA) relies on user-level tracking across channels, which has become increasingly difficult as cookies deprecate and privacy regulations tighten. MTA is not dead, but it is less reliable as a standalone method. Most measurement strategies now combine MTA with media mix modeling and incrementality testing.

Related Definitions

  • Media Mix Modeling (MMM) — A statistical analysis technique that measures the impact of marketing spend across channels on business outcomes like revenue, conversions, or market share. MMM uses aggregate historical data rather than individual-level tracking.
  • Incrementality Testing — An experimental method that measures whether a marketing activity caused an outcome that would not have happened otherwise. Incrementality testing isolates causal impact by comparing a group exposed to the marketing with a holdout group that was not.
  • Marketing Operations (MOps) — The function responsible for managing marketing technology, data, processes, and performance measurement. MOps is the operational backbone that enables marketing strategy to execute at scale.