The experiential stat everyone quotes has no study behind it. I went looking.
"Experiential campaigns deliver 3:1 to 5:1 ROI."
I've put that line in decks. You've probably seen it in one of mine. Last week I went looking for the study.
There isn't one.
The trail runs through four aggregator sites citing each other and then stops. One of them, to its credit, footnotes the figure as an industry estimate with no primary research behind it.
Same story for "78% of consumers can recall a brand after an experiential interaction" and "70% of attendees become repeat customers." Quoted constantly. Unsourceable.
Here's what does survive checking:
98% of attendees create digital or social content at branded events. That one's real — EventTrack, via Event Marketer. It's also from 2016, which means it predates TikTok existing as a consumer product. If anything it's now conservative.
Global experiential spend hit $138.9B in 2025 and PQ Media forecasts roughly 10% growth this year — call it $153B. The $120B figure still doing the rounds is years stale.
And the buying-intent number has moved in the wrong direction: 91% in EventTrack's 2021 read, 61% in the 2026 one.
The case FOR experiential:
It's one of the few channels where attention isn't rented, interrupted or skippable
Every attendee is a distribution node. The event is the shoot
Spend is growing double digits, which means your competitors are already there
The case AGAINST:
Half the business case is folklore. That should bother us more than it does
Cost per person reached is enormous next to paid social
500 people had an incredible night. Now what
Where I land: experiential probably does work. But we've been selling it with numbers we can't defend, and the first client who checks is going to be a bad meeting.
Build your own evidence. Pre and post brand lift. Code redemption. Repeat purchase on attendees versus a matched holdout. Unglamorous, defensible, yours.
Serious question for the people who run these: has anyone actually measured incremental revenue from an activation? Not attendance — revenue. I'd rather have one real number than ten borrowed ones.
Notes on the data
For the 3:1 ROI stat there is no link, which is the point. Happy to walk anyone through the citation chain — it's four sites deep and ends in a shrug.
If you have the primary study, I will genuinely and publicly correct this.