Common models: first touch, last touch, linear, time decay, position based, data driven
Attribution is a modeling choice, not a measurement. Every model distributes a fixed amount of credit across touchpoints according to a rule someone selected, and different rules produce different winners from identical data. Signal loss from privacy changes, cookie deprecation and cross-device behavior has made deterministic attribution steadily less reliable, which is why serious operators pair it with holdout testing.
One conversion involves an organic blog visit, a paid social click, a branded search click and an email click. First touch gives all the credit to organic. Last touch gives it all to email. Linear splits it four ways. The customer journey never changed. Only the reporting rule did, and each model would justify moving budget somewhere different.
Boards should treat attribution reports as directional. The right framing is that attribution allocates credit while a controlled holdout measures causation, and only the second one can justify a large budget shift. A marketing leader who explains that distinction unprompted is usually the one whose numbers end up being trusted.
If Attribution is the metric under pressure in your next board meeting, the work usually starts with analytics and reporting.
Work With Me