Last-Touch Attribution

Last-touch attribution is a single-touch crediting rule that awards a conversion entirely to the final interaction recorded before it. If a shopper saw a connected-TV ad, later read a review, then bought after clicking a branded search ad, last-touch credits the search ad and nothing else. It is the default in most analytics platforms — partly because it is simple to compute and partly because the closing click is the easiest thing to observe — which is exactly why it has shaped so many budgets. The model rests on an appealing but narrow assumption: that the last thing a customer touched is what convinced them. Sometimes that is roughly true. Often it is not, because the final touch is frequently a low-effort confirmation of a decision the customer had already made for reasons last-touch never sees.

The danger of last-touch is that it systematically rewards channels that stand near the finish line, whether or not they did the persuading. Branded search, retargeting, and email recapture sit right before the purchase and scoop up the credit, while the awareness work that created the demand — broad social, content, video, sponsorships — looks worthless. Acting on that picture, teams cut upper-funnel spend, watch the pipeline thin, and only later connect the two. Last-touch is not wrong so much as incomplete: it answers "what was the last step" precisely and "what caused the sale" badly. Treat it as a closing-channel report and a starting point for diagnosis, never as a faithful account of which efforts actually moved the customer.

Last-touch and first-touch attribution are mirror images. Last-touch gives the whole conversion to the closing interaction and ignores how the customer arrived; first-touch gives it all to the opening interaction and ignores how the deal was sealed. Both discard everything in between, so both describe a real journey with a single point. The practical consequence is that they flatter opposite channels: last-touch makes finishers like branded search and retargeting look indispensable and discovery look like waste, while first-touch does the reverse. Because the default tool in most stacks is last-touch, the more common real-world distortion is over-crediting the bottom of the funnel — which is why so many brands quietly underinvest in the awareness that fills it.

Multi-touch attribution is the response to both extremes. Rather than handing one hundred percent to a single step, it distributes credit across the touches in the path — linearly, with more weight to the first and last, decayed by how recent each was, or fitted statistically. Last-touch is simply the most aggressive end-loaded case of this family, a model that zeroes every weight except the final one. Naming it that way exposes its blind spot. When you genuinely want to know which channels close, last-touch is a clean read. When you want to know what contributed overall, use a multi-touch model and, where the decision matters, validate the answer with an incrementality test, because the last click is often a passenger taking credit for a journey other channels drove.

Use last-touch attribution for what it actually measures — closing performance — and label it as such so no one mistakes it for the whole truth. It is a legitimate way to compare bottom-of-funnel channels against each other and to spot which finishers convert efficiently once a customer is ready. The mistake is letting the platform default quietly become your strategy, then defunding everything that does not appear in the final click. Counterbalance it with a first-touch or multi-touch view to see the demand-creation side, and where budgets are large, run an incrementality test, because branded search and retargeting frequently harvest conversions that would have happened anyway. The right posture is to use last-touch as one report among several, not the scoreboard the rest of the plan bends toward.

The failures cluster around mistaking convenience for accuracy. Teams accept last-touch because it is the default, credit the closers, slash upper-funnel spend, and then puzzle over a shrinking pipeline. Others let it overstate retargeting and branded search, which sit closest to the purchase and absorb credit for demand other channels created. And many forget that a low-effort final click — typing a brand name into search — is often a symptom of persuasion that happened earlier and elsewhere. The discipline is to state last-touch's closing-only bias plainly, read it only as a finishing-channel measure, pair it with discovery and multi-touch views, and confirm the genuinely incremental contribution with controlled testing rather than the convenient last click.

Last-touch attribution — a single-touch rule crediting the final interaction with the entire conversion — is the analytics default, measuring which channels close while ignoring the demand creation that fed them.

Etymology: source.

Search interest for this term over the last five years:

View interest-over-time on Google Trends →

Areas of marketing where last-touch attribution is a core concern: