SSSampling is one of the most powerful acquisition tools in pet nutrition. A pet either eats the food or it does not, and that first bowl decides more about future purchase than any advertisement ever will. But most sampling programmes are run as distribution: push as many units out as possible, hope some convert, never really know what happened.
Working on Mars Petcare brands, we ran sampling as a funnel with explicit stages — reach, request, delivery, trial, first-party conversion — and measured each stage the way a paid channel is measured.
Where most sampling budgets leak
- Audience waste: samples reaching homes with no pet, or homes loyal to a competitor that will never switch.
- No capture: the trial happens anonymously, so the brand never owns the relationship it paid to create.
- No follow-up: the highest-intent moment — the first bowl — passes with no CRM or commerce journey attached.
Treating the sample as an acquired user
The shift is simple to state and hard to execute: a sample is not a cost line, it is the cost of acquiring a qualified first-party prospect. Once that framing lands, every decision changes — targeting, creative, partnerships, and above all measurement.
We paired integrated media (reach and request generation) with commerce partnerships and sampling mechanics that required a simple sign-up, so every sample shipped with a relationship attached. Lifecycle CRM then followed up at the moment of trial with content relevant to breed, life stage and pack size, plus a clear first-purchase path.
A sample without a follow-up journey is a donation. A sample with one is acquisition.
What actually moved the cost curve
- Tighter audience definitions cut waste before a single unit shipped.
- Request-based mechanics — the pet parent asks for the sample — improved qualification and delivery success rates.
- Media planned and bought against reach benchmarks delivered +22% overdelivery on the same budget.
- Cohort-level trial-to-purchase tracking showed which audiences were worth sampling again, and which were not.
The result: cost per sample down 40% while distribution scaled — because every stage of the funnel was owned, measured and optimised like a performance channel.
Figures in this article are drawn from the author's prior professional experience at Mars Petcare and are presented as learnings, not as client results.