The 8,000-follower doctor beat the 30-million celebrity by 52 points.
- The five real creator archetypes under audit
- The two product lines and their claims
- The buyer segments in the category
- The influencer budget question the brand faced
- Each creator's modelled purchase-authority score (0–100)
- The credibility audit each buyer ran on each creator
- The gap between reach and authority
- Which archetype actually moved purchase
The budget was allocated on the assumption that follower count buys authority.
A functional-food brand with two product lines wanted to audit its influencer spend — was it generating purchase authority, or just impressions? Five creator archetypes, from an unpaid practitioner to a 30-million-follower celebrity, were tested across 40 synthetic buyers in three decision rounds. The going-in belief was the industry default: bigger audience, bigger effect.
Credibility beat reach, and it wasn't close.
Every buyer silently asked: why is this person saying this?
Buyers in this category run a fast credibility audit on any endorsement — why is this person recommending this? An unpaid clinician with 8,000 followers has an obvious, disarming answer: a doctor with that small an audience isn't optimising for brand deals, and "I've been recommending this to my patients" reads as practice, not sponsorship. A 30-million-follower celebrity fails the same audit instantly — the incentive to be paid is visible, so the endorsement is discounted. Reach delivered impressions; it did not deliver the authority that converts.
Reallocate budget from reach to credibility. Weight spend toward practitioner and expert micro-voices whose incentive to endorse is not visibly commercial, and treat large-follower celebrity placements as awareness, not as purchase drivers — the audit buyers run will discount them either way.
These are modelled outcomes, not measured ones. Every score and probability here is a synthetic-population prediction, produced before any real-market test. Forty synthetic buyers across five archetypes — built to rank the mechanism, not to price a specific creator deal. Treat the numbers as directional — the direction and size of the effect is the finding, not a forecast. A design partner validates the top interventions against real behaviour before a decision commits to them.
“A doctor with 8,000 followers isn't optimising for brand deals. That is exactly why the endorsement lands.”SIMULATION SYNTHESIS · CREDIBILITY-AUDIT CLUSTER
Run your version of this decision.
This is the real engine. Bring the thing you're weighing and see how the population responds — before you commit to it.