How last click attribution works
The rule is as simple as it sounds. Whichever click came immediately before the conversion receives the whole of the credit, and every click before it receives none. A customer who found you through a broad research search, came back through a video, and finally converted after searching your brand name has, under this rule, converted because of the brand search alone.
There are close relatives worth knowing. Last non-direct click ignores a final visit where the customer typed the address in, and gives credit to the last identifiable marketing source instead — a small adjustment that changes reporting a great deal for businesses with strong repeat traffic.
Why last click attribution matters
It is the model everyone understands, and that is not a trivial virtue. You can explain it to a client in a sentence, trace any single conversion to the click that earned it, and produce whole numbers that behave like a count. When the question is whether a specific ad produced a specific enquiry, this is the model that answers it without argument.
It is also the honest floor of your reporting. Because it makes no estimate about anything, it cannot flatter a channel or restate last month’s figures. For a small business making a first decision about where to spend, that plainness is often more useful than a sophisticated model nobody in the room can interrogate.
Where last click attribution goes wrong
It systematically overpays the bottom of the funnel and underpays the top. Brand campaigns, retargeting and comparison searches all sit late in the journey and collect credit for demand that something earlier created. Meanwhile the awareness work — the video, the informational article, the broad search — reports nothing at all and looks like an easy budget cut. Cut it, and the well-performing brand campaign quietly runs out of people to convert.
It also flatters direct traffic and misreads long journeys. A customer who researches on a phone over several evenings and buys on a laptop from a bookmark can appear as if no marketing touched them. And because every platform tends to claim the last click it saw, adding the last-click totals of two ad platforms together will count some sales twice.
What to do about it
Keep it, but do not let it decide budget on its own. Use it as the auditable baseline: the number you can defend line by line, and the one to reach for when a client wants to know exactly which ad produced an enquiry. Then read a broader view beside it to check what the baseline is hiding.
Before cutting a campaign that looks unprofitable under this rule, look at whether it appears earlier in journeys that converted — an assisted conversion view will show you. Where journeys routinely involve several touches, a modelled split such as data-driven attribution gives a fairer picture, provided you report its figures as estimates. And whichever you choose, use the same model for a full period rather than switching mid-way and comparing two different things.
