Project SIMS

Method note · Solutions

Asking a real audience, before the budget commits

What a study answers, how one runs, what comes back, and where this sits against the alternatives you already use.

In short

A study puts one message in front of cohorts compiled from accounts that already posted, already followed and already reacted, and returns a distribution with confidence intervals — never a single confident answer. It is a triage instrument: it tells you which direction to take and which audience disagrees, in the time it takes to brief an agency. It is not a replacement for a panel when a number has to survive procurement, and it does not forecast absolute outcomes.

Cohorts
485
each ≥100 accounts
Accounts
265K
resolved, not sampled
Reactions
311K
observed, not inferred
Voice samples
9.9K
indexed for tone

01Where this sits against what you already use

The honest comparison first, because it decides whether the rest of the page is relevant to you. Nothing in the fast half of this chart replaces a panel.

slower ← turnaround → fasterindicative ← evidential weight → defensibleSurvey paneldays–weeksFocus groupweeksDesk researchdaysSynthetic personasminutesProject SIMSminutes
Figure 1 — Where each method sits. Nothing in the fast half replaces a panel for a claim that has to survive scrutiny. The gap this fills is the weeks of guessing that happen before anyone can justify commissioning one.
MethodStrongest atTurnaroundReach for it when
Survey panelReal humans; defensible for regulated claimsDays–weeksThe number must survive legal or procurement
Focus groupDepth, follow-ups, genuine surpriseWeeksYou do not yet know what to ask
Synthetic personasInstant and cheapMinutesNothing is riding on the answer
Project SIMSDirection and divergence, grounded in recorded behaviourMinutesYou need to choose before you can justify fielding

02What a study answers

Four question shapes cover most of what people bring. Each returns the same structure: a distribution, the split by cohort, and the quotes behind it.

UseThe question, as people ask itWhat comes back
Creative“Will this ad stop the scroll?”Stop rate · split by cohort · representative comments
Positioning“Which of these three lines lands, and with whom?”Per-variant distribution · pairwise gaps · ties called honestly
Risk“Who will read this the wrong way?”Divergence by cohort · the dissenting quotes
Audience“Who is actually engaging with this space?”Cohort profiles · observed topics · reach percentiles

03How a study runs

The step that separates this from prompting a model with a demographic description is the fourth. Reactions move along ties that were recorded, so a message entering a dense community behaves differently from one entering the periphery — because it does.

rounds 2…n — opinions carry forwardBrief
message + audience, in plain language
Match
cohorts drawn from the catalogue
Sample
independent draws per cohort
Propagate
opinion moves along recorded ties
Report
split, interval, divergence, quotes
Enforced at every stepNo cohort below 100 real accounts · no individual ever addressable · every share carries its interval
Figure 2 — One study, end to end. Steps three and four repeat: an agent answering in a later round has already seen how its neighbours reacted, which is how a message actually travels.

04What comes back

Would stop scrolling41%Would keep scrolling34%Would comment16%95% interval, not a pointshareBy cohortwhere the headline hides a disagreementUS · beauty commentary58%US · fitness & routine37%Global · food creators22%divergence0.31
Figure 3 — Anatomy of a result. The headline is an average of audiences that may disagree, so the per-cohort split and a divergence score sit beside it. Where two options’ intervals overlap, the report says they are tied rather than ranking noise.
ElementWhat it is for
Distribution, not a verdictEvery option with its share and a 95% interval. Overlapping intervals are reported as tied.
The split by cohortThe headline is usually an average of audiences that disagree. Each cohort is shown separately, with its own sample size.
Divergence scoreOne number for how far apart the cohorts are. High means the average is hiding the finding.
Representative quotesDrawn from the cohort's indexed voice, so tone is retrieved from what members actually wrote rather than invented.
Stated limitsWhich cohorts were thin, what the evidence does not cover, and the run's own parameters.

05What a study will not tell you

Stated plainly, because a buyer who is surprised in month three is a worse outcome than one who declines today.

Absolute forecasts

Direction and relative gaps hold up far better than a predicted percentage. Treat the numbers as a ranking, not a projection.

Audiences outside the catalogue

If a brief matches thin cohorts, the platform says so rather than substituting neighbours.

Anything about an individual

Every cohort is an aggregate of at least 100 accounts, by construction. There is no individual to address.

Reactions to breaking events

Evidence carries a capture latency. A cohort does not know what happened yesterday.

Bring the question you were going to guess at.