Multi-touch attribution

Attribution that splits credit for a conversion across several tracked touchpoints, using fixed rules or a statistical model.

By Azande PorterUpdated 3 min read

What is multi-touch attribution?

Multi-touch attribution (MTA) splits the credit for a conversion across several tracked touchpoints on the path to it, instead of giving everything to the first or the last click. It exists to fix the obvious flaw in last-click attribution: content that starts a journey gets no credit there.

For example, a buyer's tracked path might be: a LinkedIn post click, a newsletter signup page visit, an edition click, then a purchase. A multi-touch model decides how much of that purchase each of the four steps receives.

What are the common multi-touch attribution models?

Model How credit is split Four-touch example
Linear Equally across every touch 25% each
Time decay More to touches closer to the conversion Last touch gets the most
Position-based (U-shaped) 40% to the first touch, 40% to the last, 20% shared across the middle 40%, 10%, 10%, 40%
Data-driven Estimated from many paths, based on which touches change the chance of converting Varies by account

Google's legacy Analytics documentation describes the position-based split as 40% to each of the first and last interactions, with the rest divided evenly across the middle. Those percentages are a convention someone chose, not a measurement of your readers.

Google has since narrowed the menu. Its current Analytics help says the first click, linear, time decay, and position-based models are no longer available as of November 2023. What remains is data-driven attribution and two last-click variants, and Google Analytics uses data-driven attribution by default.

Does multi-touch attribution work for organic content?

Only partly, and the gaps are worst for creators and small teams.

  • It only splits credit among touches it can see. Those are clicks with readable sources, usually from links carrying UTM parameters. Someone who read twenty of your posts and clicked one has a one-touch path in the data.
  • Private sharing is missing. Visits from DMs, group chats, and forwarded emails often arrive with no source, so dark social never enters the path.
  • People switch devices. Without a login that ties visits together, a phone session and a laptop purchase look like two strangers.
  • Data-driven models need volume. They learn from many conversion paths. A business with fifteen sales a month gives the model very little to learn from, and the output can look precise while resting on a handful of journeys.

The result is a precise-looking split of an incomplete path. Credits to two decimal places do not make the underlying record more complete.

What should a small business use instead?

Combine cheaper methods and keep them labeled:

  1. First-party facts: what you published, when, and with which tagged link.
  2. Provider-reported facts: signups and their stored sources from your newsletter platform, orders from your checkout provider.
  3. What people tell you: a signup or checkout question gives you self-reported attribution, which reaches the parts of the path no tracking sees.
  4. Period comparisons: compare quarters, not individual posts, and treat a pattern as an observation to test again.

The guide to tracking newsletter subscribers who purchase shows how to hold those evidence types side by side without turning a sequence into a cause. That is the honest goal of any attribution work.

Platform details checked

How Distinctful helps

Distinctful does not split credit across touchpoints or assign revenue to posts. It keeps your publishing history beside subscribers and editions reported by beehiiv, Kit, or Ghost, so you can read patterns across them as observations to test rather than as credit to distribute. See newsletter analytics.

Related terms

Go deeper

Read patterns without splitting credit

Compare publishing history with newsletter growth and test what a pattern suggests.