Last-touch attribution model
Assigns 100% of revenue credit to the final touchpoint before a conversion. Simple to report, but systematically blinds you to the channels that built the relationship.
What is the last-touch attribution model?
The last-touch attribution model credits 100% of the conversion value to the final recorded marketing interaction before the conversion event. If a prospect clicked a branded paid search ad immediately before submitting a demo request, that ad receives all the credit, whether the prospect had been reading your content for six months or not.
Last-touch is the default setting in most advertising platforms, including Google Ads and Meta. It is also the implicit model used when sales teams record a "lead source" at the moment of conversion without tracking prior interactions. Because it is everywhere, it is also one of the most commonly misapplied attribution approaches in B2B marketing.
The model's strength is clarity. It answers a narrow but useful question: what was the proximate cause of this conversion? Its weakness is that it treats the final touch as the complete cause, ignoring every earlier interaction that moved the account through the funnel.
How credit is distributed
Last-touch assigns a weight of 1.0 (100%) to the final touchpoint and 0 to all preceding interactions. Journey length is irrelevant. A deal that touched 20 channels over eight months gets exactly the same report as one that touched two channels in two days: the last channel wins.
| Touchpoint | Position | Credit weight |
|---|---|---|
| Organic blog visit | 1st (first touch) | 0% |
| LinkedIn sponsored post | 2nd | 0% |
| Nurture email click | 3rd | 0% |
| Branded paid search click | 4th (last touch) | 100% |
Compare with first-touch (100% to the first interaction) and time-decay (exponentially higher weights as you approach the conversion date), which credits later touchpoints heavily but still distributes some credit earlier in the journey.
When to use last-touch attribution
Last-touch is most defensible in three specific situations:
- Very short sales cycles (under a week) where the buyer journey is genuinely one or two steps
- Direct-response campaigns where the ad is the offer: the click and the conversion are the same event
- Sales ops reporting where you need one "source of record" per deal for CRM hygiene, not budget allocation
For most B2B SaaS companies with multi-touch journeys, last-touch produces systematic budget distortions. It pushes spend toward bottom-of-funnel retargeting while starving the content, SEO, and events that build the pipeline in the first place. Pairing it with W-shaped attribution gives you both the closing-channel signal and the full pipeline picture simultaneously.
Pros and cons
Pros
- Directly links the conversion event to a specific channel, easy to report
- Useful for measuring which channels are most effective at closing deals
- Straightforward to implement: requires only the final touchpoint before conversion
- Widely understood by sales teams accustomed to "source of record" thinking
Cons
- ✕Assigns zero credit to every channel that built awareness and interest earlier
- ✕Frequently overcredits demos, sales calls, or branded search at the bottom of funnel
- ✕Distorts top-of-funnel investment decisions by making awareness channels look ineffective
- ✕Breaks down in long B2B sales cycles where the last touch may be a low-effort click
How AttriByte handles last-touch attribution
AttriByte calculates last-touch in real time alongside all five other models. Because all models share the same cookieless identity graph and warehouse event stream, the last-touch number on any report is directly comparable to the linear or W-shaped number without any data reconciliation work. You can see instantly how much revenue last-touch is overcrediting to closing channels relative to other models.
The cookieless identity layer ensures the "last touch" is genuinely the last recorded marketing interaction, not simply the last click that happened to be captured before a cookie reset. Identity stitching connects sessions across devices and over time, so the model operates on a complete journey rather than a fragmented one.
Model outputs write back to your own warehouse. You can query last-touch revenue by channel, campaign, or account segment directly in SQL without exporting from a vendor dashboard. See the full platform overview on the product page.
Explore other models
See last-touch next to five other models
AttriByte calculates all six attribution models simultaneously so you never have to choose just one.