Ask three New Zealand brands what their last sampling activation cost and you will get three numbers that are not comparable. One will quote the agency fee. One will quote the agency fee plus the product. One will have quietly written off the pallet that came back from the depot and stopped talking about it.
That is not sloppiness. Sampling costs are genuinely hard to see, because most of them sit in other people's budgets.
The line items everyone remembers
Three costs make it into every sampling budget, because someone invoices for them.
- The product. Units given away, at cost rather than at retail, plus whatever is lost, damaged or never returned.
- The agency. Staffing, briefing, uniforms, a supervisor, and a report at the end.
- The site. Whatever the retailer or the venue charges for the space and the time.
Those three are the number that gets approved. They do not capture every cost of the activation.
The line items that hide
The rest of the cost is real, but it arrives as time rather than as an invoice.
- Freight and handling. Product has to get to a depot and then to a store, in cartons nobody planned for, on a timeline that suits the retailer.
- Retail coordination. Someone in your team agrees dates, planograms and access with the retailer.
- Internal time. Briefing, approvals, a site visit, and the Saturday somebody spends checking that the table went up.
- Opportunity cost. One catchment, one store, one day. Everyone who shops on a Tuesday is not in the sample.
None of that is unreasonable. It is simply the true price of putting a folding table in a supermarket, and it belongs in the comparison whenever someone says an alternative sounds expensive.
What you have at the end
This is the part that matters. At the end of a sampling day you have a count of shoppers approached, a stack of leftover product, and a report that says the activation was well received.
You do not have the identity of a single person who tried it. You cannot re-contact them, you cannot see whether they bought it again, and you have nothing to run as media on Monday. The spend has been converted entirely into a moment that has now passed.
That is the real cost, and it does not appear on any line.
A different way to count it
The alternative is to spend the same money on trial that leaves something behind. In a purchase-first sampling campaign, the customer buys the product themselves through normal retail, uses it at home, and posts what happened. The brand refunds the purchase and pays for the content.
The cost structure changes shape. There is no freight, because nothing is shipped. There is no retail coordination, because nothing is placed in a store. There is no demonstrator, because the person doing the demonstrating is the customer.
What replaces those costs is the participant payment and the moderation work behind it, which is the unglamorous part: matching people to the brief, checking proof of purchase against the content that comes back, and rejecting what does not meet the standard.
What it leaves behind
Three things, all of which outlive the campaign.
A verified trial count. Not shoppers approached. Individual purchases, matched to a person and to the content they posted. It is a smaller number than a demo-day tally and it is a number you can put in front of a category manager.
A content library you own. Dozens to hundreds of pieces, licensed for paid, organic, retail and the trade presentation. That is the deliverable a demo day cannot produce at any price.
Paid media that works harder. The content that performs organically can then be run as advertising. In a New Zealand campaign for Kiwicare Weed Weapon in December 2025, customer content ran at 2.78% click-through against 1.07% on the brand's own paid ads in the same campaign, at NZ$0.20 a click.
Source: Kiwicare Campaign Report, January 2026.
The comparison worth making
Not cost per sample. Cost per sample is the metric that makes a demo day look efficient, because it counts a spoonful in a paper cup as a trial.
The useful comparison is cost per verified trial, plus what you hold at the end. Once the content library and the paid performance sit on the same side of the ledger, the arithmetic usually stops being close.
If you are working through that comparison for a specific product, the retail sales lift page sets out how the result gets measured, and food and beverage covers the category rules that shape the brief.