Last-click attribution gives 100% of the credit for a conversion to the last marketing touchpoint before it happened, and 0% to everything earlier. If someone sees a social ad, opens an email, then clicks a paid search ad and buys, paid search gets the whole sale.
It is simple, easy to audit, and the default in many ad platforms. It misleads when channels that start or nurture demand are never the final click, so they look worthless on paper.
A Worked Example
One illustrative $100 order. The customer's path, in order: a social ad, an email, an organic search visit, then a click on a branded paid search ad right before buying. Here is how the same order is credited under four common rules.
| Touchpoint | Last click | First click | Linear | Position-based (40/10/10/40) |
|---|---|---|---|---|
| 1. Social ad | $0 | $100 | $25 | $40 |
| 2. Email | $0 | $0 | $25 | $10 |
| 3. Organic search | $0 | $0 | $25 | $10 |
| 4. Branded paid search (last) | $100 | $0 | $25 | $40 |
Same order, same customer, and the ranking of channels flips depending on the rule. That is the central point: an attribution model is an assumption about credit, not a measurement of cause.
Where Last-Click Works Well
- Short purchase cycles with one or two touches, where the final click really is most of the story.
- Direct-response campaigns judged on a quick, same-session conversion.
- Reporting that has to be easy to reconcile with platform and billing data.
Where It Misleads
- It over-credits bottom-funnel channels. Branded search and retargeting capture people who were already coming, so they collect credit for demand something else created.
- It under-credits awareness and nurture channels. Social, video, content and email rarely get the last click, so budget gets cut from the channels that fed the funnel.
- It cannot see cross-device or offline steps. A visit on a phone followed by a purchase on a laptop can show up as direct.
- It rewards the channel that is easiest to measure, not the one that changed behavior.
How to Check Whether It Is Misleading You
- Compare models on your own data. Run last-click, first-click and a multi-touch rule over the same orders. Channels whose credit swings the most are the ones the model choice is deciding for you. See attribution models compared.
- Look at assisted conversions. A channel that appears in many paths but is rarely last is a likely under-credited one.
- Run a holdout. Switch a channel off (or hold back a random audience) and measure what happens to total conversions. Attribution reports can only say who got credit; only an experiment says what the channel caused. See incrementality measurement.
- Make small moves first. If a different model suggests shifting spend, shift a small share, watch total results, and continue only if they hold.
For a deeper comparison of how last-click and multi-touch models change budget decisions, see last-click vs multi-touch.
Frequently Asked Questions
What is last-click attribution?
Last-click attribution assigns all of the credit for a conversion to the final marketing touchpoint before it. Every earlier touchpoint in the customer's path gets nothing.
What is the difference between last-click and first-click attribution?
Last-click credits the final touchpoint before the conversion; first-click credits the first one. In a path of social ad, email, organic search and paid search, last-click gives paid search everything and first-click gives the social ad everything.
Why is last-click attribution misleading?
It over-credits channels that capture existing demand, such as branded search and retargeting, and under-credits channels that create demand earlier in the path, such as social, video, content and email.
Is last-click attribution still worth using?
Yes, for short purchase cycles and direct-response campaigns, and as a simple baseline that is easy to reconcile. It is best read next to a multi-touch view and checked with a holdout test before large budget moves.
What should I use instead of last-click attribution?
Compare it with linear, time-decay or position-based multi-touch models on the same orders, and confirm the biggest differences with an incrementality test, where you hold back a channel or audience and measure the change in total conversions.