The pitch you have already heard about AR location campaigns, verified real-world reach, digital assets as media, the retail moment reimagined, is not this piece. This piece is for the person who has bought the pitch and now wants to know what to write on the brief. It is a working guide rather than a manifesto, and it is written from the position of a platform that has run a few hundred of these and knows where they succeed and where they die.
Why brands run them in 2026. The short answer is verified visit versus impression. A billboard billed in CPM is charging for the possibility that somebody with the right demographic drove past. An AR location campaign is charging for the fact that somebody physically walked to your door, opened your brand's app or a partner app, and claimed a thing you placed there. The unit economics are directly comparable and often win on cost-per-verified-arrival even before you count the wallet you now have a connection to.
The three shapes a campaign takes. Single-location drop: something worth having placed at one address, live for a short window. Multi-city relay: the same drop at every branch of a chain, sometimes with a race element. Event overlay: drops seeded across a defined venue for the duration of a festival, sports event or conference. Every campaign we have run is a variant of one of these three, and knowing which shape you are in changes almost everything else in the brief.
Shape one, in practice. A flagship store places a limited-edition asset in the shop window, visible only through the AR app, claimable only inside the store. The typical version is a hundred units, a two-hour window, and a Saturday afternoon. The reward can be a coupon, an NFT collectible, a token, or access to something else. What matters is that the window is narrow enough that the crowd forms, the drop that empties in twenty minutes with fifty people outside is the drop that gets the press.
Shape two, in practice. A coffee chain seeds a two-euro credit at every one of its forty-eight branches during a launch week. Different rewards per branch keep the completionists engaged; a leaderboard tracks visit count. This is a category of campaign where the interesting number is not the total pickups but the users who visited more than three locations, those are the ones the brand can now email, or send to a next campaign, or convert into loyalty members. The brief writes itself around that number.
Shape three, in practice. A music festival places branded collectibles at each stage and at the two food areas. Attendees discover them by wandering. The point is not the collectible per se, most attendees will not visit all of them, it is the fact that the festival's brand is in the app and in the wallet of every attendee who claimed even one, and stays there after the festival is over. Event overlays are the campaign shape with the longest post-event tail.
What a working brief actually contains. Six things and nothing else. The coordinate or coordinates. The claim radius (five metres for a shop doorway, five hundred for a festival ground). The reward, described in one sentence. The duration, in hours or days rather than weeks. The cap, how many copies exist, and what happens when they run out. And the trigger, which is usually a specific action the user takes in-app to receive the reward once inside the radius. Anything past those six is optional. Under those six, no campaign works.
The reward matrix. Rewards fall into four categories and the choice of which is the most consequential decision in the brief. Utility tokens (small amounts of a specific coin) are easy to explain and easy to over-supply, the brand that airdrops the same token every week trains its users that the token is worthless. Collectible NFTs work best when the art is genuinely nice, ideally by an artist the audience knows; a bad collectible is worse than no collectible. Coupons work well and are legally the most complex; check with counsel before placing anything redeemable for money. Experience access, a wristband to skip a queue, a table at the restaurant, a seat at the front, is the category with the best conversion and the hardest logistics.
Budget math. The way to think about the cost is per-verified-arrival, not per-drop. A campaign that places a hundred rewards worth five euros each is not a five-hundred-euro budget; it is five hundred plus platform fees plus the cost of the design work plus your own team's time. Divide the total by the number of verified arrivals at the end and you have your CPVA. In the campaigns we have run, CPVA lands between three and eleven euros depending on the reward attractiveness and the location. Compared to good OOH billboard CPMs at similar total budget, the arrival number is usually one to two orders of magnitude higher.
A case with numbers. A coffee chain we worked with last quarter placed a five-euro credit at 34 branches over a Wednesday-to-Sunday launch week. Total budget including our platform fee was around eighteen thousand euros. Verified pickups: 1,712. Coupon redemption rate at the branch counter: 71%. Follow-on visits within the next fortnight from the pickup wallets: about 2.2 per user, most of them paid rather than discounted. CPVA for the campaign as measured came in at ten and a half euros. The client's internal number was that in-store CPMs from any other channel would have landed in the same range but without the wallet connection at the end.
A case without numbers. A streetwear label placed a hundred branded AR wearables at their flagship for a Saturday afternoon. The reward emptied in twenty-two minutes; there were three hundred and change people outside the store when it did. Local press covered it; three national outlets picked it up the following week; the brand's mailing list from that single event grew by four thousand names. This is the kind of campaign that does not appear in a CPVA calculation because the value is elsewhere, and it is also the campaign shape most likely to justify itself before any other outcome.
What campaigns fail at, and why. Indoor drops without a beacon fail because GPS is bad indoors and users cannot find the coordinate. Weather-sensitive drops fail; nobody walks three streets to a virtual coffee credit in a storm. Drops with no follow-up fail, the brand gets the pickup, does nothing with the wallet, and the user forgets they ever engaged. And drops that misjudge cap-versus-window fail, either the reward emptied before anyone could get there, or so much of it remained at the end that scarcity did not motivate anybody.
Measurement, from your side. Instrument the walk-in on your end regardless of what the platform reports. Point-of-sale flag for coupon redemption. Optional email capture at the counter if the campaign is coupon-based. Simple event log for the AR-triggered in-app action. If the platform's numbers and yours disagree, both are right about different things and the gap is diagnostic, it usually tells you where the drop-off is in your funnel between claim and conversion.
Post-campaign. The most-neglected step. Every wallet that claimed your drop is now a wallet with a wallet-app installed on a phone, and can be reached by future campaigns without buying the connection again. Segment those wallets by which drop they claimed, how many, in what order; that segment is the audience for the next drop. Brands that treat AR campaigns as one-shots leave the compounding on the table. Brands that treat them as a channel see the second campaign convert at two to three times the rate of the first at half the CPVA.
When to run one, when not to. Categories with a natural physical destination, retail, food and drink, live events, tourism, hospitality, benefit most and have the shortest brief. Categories where the customer never has to visit anywhere physical benefit least; those brands are usually better off with digital-only crypto distribution and should not be pretending otherwise. The novelty premium, where the campaign gets press because it is one of the first of its kind in a market, is real, and it decays quickly. If you are running one in 2026 in the Netherlands you can still get a local newspaper to write about it. That will not be true in 2028.
How to start. Pick one location. Pick one reward. Pick one four-hour window on a Saturday. Cap the reward at whatever feels sensible and halve it. Publish it. Watch what happens. The learning curve on the first campaign is what makes the second one much better; the second-campaign brief is where most of the compounding starts. If the publisher side of Seekprotocol is where you want to try one, the tooling for the first three shapes above is what the panel is built around, and we help walk clients through the first brief until it feels self-serve.
The short version. AR location campaigns work when the brief is boring and the discipline is high, one location, six variables, one clear reward, a narrow window. Everything else about them is discretionary, and most first-timer failures come from moving faster than the discipline can keep up. Start small, instrument well, treat the first campaign as a data-gathering exercise, and the second one is where the medium starts paying back.
