Guerrilla marketing uses an unexpected experience or placement to attract attention with relatively limited media buying. The usual spelling is guerrilla, although people often search for “gorilla marketing”. A clever stunt is only useful if the intended audience understands the brand and has a reason to take the next step.
The effect you can estimate depends on the intervention, comparison and outcome you define. The incrementality and experiments learning path connects this idea to the assumptions, practical checks and next steps.
To work through the connected decisions, read Geo experiments: audit spillover before interpreting lift and Incrementality tests: choose a commercially useful MDE.
Start with the audience’s situation
Write a one-sentence brief: “We want this audience, in this place, to remember this benefit and take this action.” A commuter installation, a sampling event and a participatory street game create different experiences. Choose the format because it fits the buying situation, not because it is likely to look impressive in a case study.
Plan the practical details with the venue and relevant local authorities: permission to use the space, accessibility, staffing, cleanup and how to avoid obstructing people. An activation that annoys passers-by can generate attention while weakening the brand. The campaign budget should include these operational costs and staff time.
Give the experience a measurable next step
For a fictional coffee subscription, a tasting outside a partner venue might offer a location-specific landing page with a sample-box offer. A QR code helps connect an encounter to a visit, but a scan is not a new customer. Record visits, completed orders, cancellations and repeat purchases separately.
Suppose the event costs £1,200 and produces 60 first orders with £15 contribution per order before the event cost. Those orders contribute £900, leaving a £300 shortfall on that initial accounting. The calculation is still optimistic if some customers would have purchased anyway. Future repeat contribution belongs in a separate, explicitly stated assumption.
Design a comparison before launch
- Select comparable locations and review their pre-campaign sales patterns.
- Run the activation in a randomly selected subset if practical. Avoid simultaneous local discounts that would obscure the effect.
- Measure outcomes over a pre-agreed window that reflects the purchase cycle.
- Report exposure proxies separately from incremental sales and contribution.
The incrementality guide explains the counterfactual. For campaigns assigned by location, the geo-testing guide explains why geography, spillover and pre-period fit matter.
Decide what deserves a second run
Keep evidence of execution quality: photos with appropriate permissions, stock availability, staffing issues and whether people could actually understand the offer. A failed rollout and a failed idea are different diagnoses. If the event produces strong qualitative engagement but little measurable demand, revise the offer or audience before repeating it at greater scale.