Baseline first, always
The most common mistake is switching the system on and then looking for a change. Without at least four weeks of pre-installation baseline on the metrics you care about, any movement afterwards is indistinguishable from seasonality.
Capture the baseline before the hardware arrives. It costs nothing and it is the difference between a result and an anecdote.
Pick metrics the business already tracks
Resist the urge to invent a scent-specific KPI. Use what finance already reports: spend per visit, conversion rate, average dwell time, length of stay, repeat visit rate, guest or patient satisfaction scores, staff-reported comfort.
Published research across retail and hospitality consistently associates ambient scenting with double-digit improvements in dwell time and spend per visit, and with materially higher brand recall. Your own numbers will differ — what matters is that they are your numbers, measured the same way before and after.
Run a control if you can
Multi-site operators have an advantage here: scent two comparable stores and leave two unscented for the measurement period. It removes seasonality, promotions and macro conditions from the comparison in a way that a before-and-after at a single site never can.
If a control is impossible, stagger the rollout instead and compare cohorts by activation date.
Do not forget the qualitative signal
Unprompted mentions of smell in reviews and feedback forms are a strong leading indicator, and they are free to collect. Before installation, mentions of odour are almost always negative. Afterwards, the useful signal is not praise for the fragrance — it is the disappearance of complaints about the air.