The Maze: Germany’s digital-economy association, BVDW, wants retailers to judge affiliate partners beyond the final click before a purchase. These partners include websites and creators that earn commissions for referring sales. Its October 22 paper builds an operating framework around older German data: affiliates appeared in twice as many purchase journeys as final-click credit revealed.
The Details:
Two figures, two questions. Tracking providers attriXus, easy Marketing and Ingenious Technologies supplied data covering 185 advertisers. Affiliates received final-click credit in 12.41% of transactions. Count any recorded affiliate contact before purchase, and their participation reached 24.92%. A shopper might discover a product through a review, then return through a search for the retailer’s name. The earlier contact survives in the second measure. The original study dates data collection to early 2026; BVDW had already published the comparison in June. October’s development is the management proposal, rather than newly discovered sales.
The money is substantial, but attributed. The paper cites €932 million invested in German affiliate marketing in 2025 and €18.7 billion in associated revenue. These are separate market totals, not the value of the 185-advertiser sample. The June methodology says spending includes commissions, fixed payments and network service costs. It combines 13 direct network submissions with modeled missing data and an estimate for Amazon’s affiliate programme. Modeling accounts for less than 10% of reported volume. None of this revenue is a measured estimate of sales that would vanish without affiliates, or of merchants’ profit.
Paying and judging can use different rules. The paper proposes retaining final-click attribution for commissions while adding a second view that distributes credit across earlier contacts. It also recommends putting affiliate and other marketing data in one analytics environment, with a common purchase definition. A separate test must then ask whether the partner added sales: the appendix suggests control groups for cashback and voucher programmes, for example. That means comparing outcomes with a group that did not receive the intervention. The paper proposes tests; it does not report their results or quantify the sales uplift they would find.
MarketMaze Take: The useful change is a second lens for budget decisions. A commission rule settles who gets paid for a recorded sale. It cannot, by itself, tell a merchant which partner persuaded someone to buy. Counting every earlier contact improves visibility, but still leaves that harder question open.
Give discovery a hearing, not a blank cheque. Earlier reviews and creator recommendations may influence customers who later return directly. Cutting them solely because they lose final-click credit can therefore miss value. Equally, appearing in a journey does not prove persuasion. The paper’s authors work in the affiliate industry, and the publications do not identify the advertisers’ sectors, transaction count or the period over which contacts were counted. The larger percentage earns investigation, not an automatic doubling of budget.
Export the distinction, not the percentages. Merchants outside Germany can separate commission settlement, recorded participation and tested additional sales. They cannot assume these German shares describe their own customers. A broader report can reveal which partners deserve a test; a credible comparison can establish which deserve more money. Those are different decisions, even when the same purchase appears in both reports.
