What is Last-Click Attribution?
Last-click attribution is a single-touch attribution model that assigns 100 percent of the credit for a conversion to the final marketing interaction before the sale. It is the default model in Google Analytics 4, most ecommerce platforms, and the native reporting in most ad platforms.
If a customer first saw a Meta ad, clicked a paid search advert a week later, and finally converted after clicking a branded email, last-click credits the email and ignores the other two touchpoints.
Why Last-Click Became the Default
Last-click is dominant for historical and practical reasons, not methodological ones:
- It is easy to implement because the final touch is the one most directly observable at the moment of conversion.
- It is easy to explain to non-technical stakeholders.
- It produces unambiguous credit assignment with no need for statistical models.
- Most advertising attribution grew out of direct-response tracking, where the last click was all that mattered.
The Problem with Last-Click
Last-click systematically distorts marketing performance in ways that lead to poor budget decisions:
- It over-credits bottom-funnel and branded channels such as direct, email, and branded search. These channels often harvest demand that was created elsewhere.
- It under-credits top-of-funnel and discovery channels such as display, paid social awareness, and organic content. These channels rarely deliver the final click but are essential to creating the demand that is later harvested.
- It makes retargeting look stronger than it is, because retargeting tends to be the closing touchpoint even when it adds little incremental value.
- It incentivises marketers to invest in conversion-adjacent channels at the expense of awareness, which reduces the pipeline of new demand.
The Branded Search Trap
A classic example of last-click distortion is branded search. A customer sees a Meta ad, remembers the brand, searches Google for the brand name, clicks the paid brand advert, and converts. Last-click credits the paid brand search and gives Meta nothing.
The team, reading last-click data, concludes that paid brand search is highly profitable and Meta is weak. They increase brand search budget and cut Meta spend. A quarter later, branded search traffic starts declining because fewer people are being introduced to the brand.
The root cause was not that brand search was bad. It was that last-click hid the role Meta played in creating the brand awareness that drove the branded search in the first place.
When Last-Click Is Acceptable
Last-click is acceptable in limited situations:
- Very short sales cycles where only one or two touchpoints exist
- Single-channel campaigns where no other touchpoints are active
- As one lens among several in a multi-model attribution analysis
It is not acceptable as the sole attribution model for any business with a considered purchase, multiple channels, or long sales cycles.
Moving Beyond Last-Click
The alternatives to last-click fall into two broad categories:
Multi-touch attribution models (linear, time-decay, position-based, full-path) distribute credit across the journey in different ways, each reflecting a different assumption about how credit should be allocated.
Marketing Mix Modeling (MMM) uses aggregate spend and revenue data to infer channel contribution without relying on individual user journeys, making it resilient to privacy-driven tracking loss.
Most mature measurement stacks use both approaches in combination and treat last-click as one reference lens rather than the single source of truth.