The number everyone quotes is the casino one. Pipe a scent onto the gaming floor and slot-machine spend jumps 45 per cent. It gets repeated in almost every scent-marketing pitch, and it’s why so many buyers expect fragrance to behave like a switch: turn it on, watch the numbers rise.
The researchers who actually study this are far less certain, and that’s the more useful place to begin.
When Christoph Teller and Charles Dennis reviewed the field for the Journal of Marketing Management, they found something the sales decks tend to leave out. Across the published studies, the effect of ambient scent on what people actually did, spending, lingering, buying, was often small, indirect, or absent altogether. Scent reliably changed how people felt about a space and how they rated it. Turning that feeling into behaviour was much less dependable. The link most buyers assume is automatic is, in the evidence, conditional.
Conditional mostly on whether the scent belongs in the room.
Eric Spangenberg’s 1996 store experiment, one of the studies that put this subject on serious footing, shows the real shape of it. A pleasant ambient scent improved how people evaluated the store and its products, and it nudged their intentions. What it did not do was produce the mechanical spending surge the folklore promises. The gains were genuine and worth having. They were also modest, and they held only when the scent suited the setting rather than simply being present.
The systematic review by Davide Giacalone, Bartlomiej Pieranski and Barbara Borusiak makes the point harder to dodge. Read across the studies and the results are genuinely mixed. Where effects appear, the fine detail decides everything. One experiment they cite found a simple scent lifted actual spending while a more complex scent, in the same conditions, did nothing at all. Same intent, opposite outcome, settled by the composition of the fragrance itself.
This is where most scent programmes fail before they start. A space gets treated as a room that needs a nice smell. Someone chooses a fragrance they personally like, sets a diffuser to a default output, and waits for the graph to move. When it doesn’t, the verdict is that scent doesn’t work. The fairer verdict is that spraying was never the same as engineering.
Engineering means the scent is built for the actual conditions of the space and the specific job you want it to do. A high-ceilinged atrium disperses fragrance nothing like a fitting room does. A concourse moving three thousand people an hour thins scent at a rate a quiet boutique never will. Concentration, diffusion system, and calibration over the following months are the variables that decide whether any of the research applies to your building. Get them wrong and you have carefully recreated the conditions under which scent does nothing.
There’s a tell you can use before any of this. Ask a prospective supplier what they need to know about your building before recommending a scent. If the first questions are about cubic volume, ceiling height, airflow and footfall, you’re talking to someone who engineers. If the first questions are about which fragrance families you happen to like, you’re talking to someone who sprays.
I’d put it more strongly. The modesty of the academic findings is a reason to take the work seriously, not to dismiss it. A field promising guaranteed double-digit sales lifts would deserve suspicion. One that says, carefully, that a well-matched scent improves perception and can shift behaviour under the right conditions is describing something real, and something you can design against.
So when a supplier hands you the casino figure, ask the second question. What was the space, what was the scent, and would either survive being tested in yours? If the reply is a shrug, you’re being sold a smell, not a result. The businesses that get value from scent are the ones that stopped asking whether it works in general, and started asking whether it has been built to work here.
