How do you prove an activation worked?
The event went well. Everyone says so. There are good photographs. Now finance wants to know what it returned, and nobody wrote down the one number that would have answered them.
6 min read
The report everybody writes
Footfall: a large number. Samples distributed: another large number. Photographs: many, all of them showing people smiling.
None of it answers the only question being asked, which is whether the money did anything.
This is not because activations don't work. It is because the measurement was designed after the activation instead of before it, and by then the only things left to count were the things that count themselves.
Why footfall is not a result
Footfall past your site is a number the mall or the market gives you. It would have been roughly the same number if you had not been there.
Even footfall *at* your site is weak on its own. A person who paused for four seconds because something was moving and then kept walking is in that number. So is a child who wanted the balloon.
Counting people who came near is measuring the venue. Measuring the activation means counting what only your activation could have caused.
Decide the number before you build anything
This is the whole discipline, and it costs nothing except the discomfort of committing early.
Before the first sketch, someone has to answer: what will be different afterwards, and how will we see it?
The answer differs by objective:
- Trial. Not samples handed out — samples *used*. A coupon code, a QR, a redemption at a nearby store. The number that matters is the one that required a second action.
- Enquiries. Contact details given willingly, with permission, and — critically — routed to whoever will call them. See what happens when nobody owns the lead.
- Reappraisal. Hardest to measure and often the real objective. A short, consistent question asked to a sample of people before and after, in the same location. Crude, but it is the difference between an opinion and a measurement.
- Sell-in and sell-out. For retail-adjacent activations, the stockist four hundred metres away has a number, and it is usually the most honest one available.
The comparison that makes any of it meaningful
A number on its own is not evidence. Twelve hundred redemptions is good or bad depending entirely on what you compare it against.
Three comparisons are usually available, and at least one almost always is:
Before and after, same place. The store's own sales in the two weeks before and the two weeks after. Imperfect — other things change — but immediate and cheap.
A location you did not activate. The closest thing to a control group most brands can get. If similar markets moved and your activated one moved more, that difference is the closest thing to proof this discipline offers.
The same activation, previous edition. Once you have run something twice with the same measurement, every subsequent one gets easier to judge.
What honest reporting looks like
Say what the activation caused, say what it plausibly influenced, and say what you cannot separate from everything else happening that month.
That last category is not a failure. Every activation runs alongside other marketing, and pretending otherwise produces reports nobody senior believes. A report that admits its own limits is more persuasive than one that claims a clean attribution the method cannot support. It is also the only kind of report worth putting a name to on a brand activation.
Questions people ask
- What is the single most useful number to capture at an activation?
- Something that required a deliberate second action from the participant — a code redeemed, a form filled, a store visited. Anything that happens passively (people walking past, samples pushed into hands) measures your location, not your idea.
- How do we measure brand impact if there are no direct sales?
- A short intercept survey — the same two or three questions asked to a sample of people in the same location, before the activation and again after. It is not laboratory-grade, and it does not need to be. It is the difference between having a measurement and having an impression.
- Is footfall completely useless?
- Not useless — it tells you whether the site and timing were right, and it is genuinely useful for choosing venues next time. It is just not evidence that the activation worked. Treat it as an input you paid for, not a result you produced.
- Who should own measurement, the agency or the brand?
- The agency should build measurement into the plan and be honest about its limits. But numbers that only the brand can see — sell-out, enquiries that converted, actual sales — have to come from the brand. An activation measured only by the people who ran it is missing exactly the half that decides whether it was worth doing.
Read next
The first fifteen seconds of a mall activation
Nobody comes to a mall to meet your brand. They came for a film, a burger, or the air conditioning. You get about fifteen seconds to become part of somebody else's afternoon.
Performance marketing cannot sell what nobody wants
A performance campaign is a very good net. It is not a very good reason. If nobody wants the thing, a better net just proves it faster and more expensively.
One brand, two markets: the Vectus activation
The same brand had to work at a village haat and outside a city retailer. Those are two different audiences, two different attention spans, and almost nothing in common except the logo.