Simulation research

Simulate the world before you shape it.

Simulant simulates human behaviour so you can make more informed policy and business decisions.

Humanity at scale

Simulate the world before you shape it. The moves that matter most are the ones you can't take back — put into the world once, judged only when it's too late to change them. Simulant is a simulation of your world: a representative model of the people you're trying to reach, grounded in real data, so you can test your thinking against the world before the real one ever sees it. Not just whether it will land, but why.

Our approach

Every way to listen, in one simulation.

The methods research has always relied on — recreated as a living population you can question again and again, in a night rather than a quarter.

n = 12,438 verbatims
01

Qualitative feedback

Open-ended interviews at population scale. Read the verbatims and the reasoning behind every answer — not just the score.

μ 62.4 · σ 11.2
02

Quantitative feedback

Structured measures across the whole population — intent, preference, and sentiment, cut by any segment that matters.

7 personas · live
03

Focus groups

Convene any room you can imagine. Watch a segment deliberate, react, and shift its mind as the conversation unfolds.

Validation

In a blind back-test, Simulant reproduced a national attitudes study — six months of fieldwork — in a single night. Where our results diverged from the survey, follow-up found Simulant was closer to the truth.

Read the validation report
The instrument

Interrogate the simulation.

Pose a question, run it against the population, and read the answer by any cut that matters — with the verbatims and the reasoning behind every number.

Spring Launch — Hero Video A/B
Export
Recall
72/100
+8 vs benchmark
Sentiment
78/100
Positive skew
Intent
54/100
Mid-funnel
Clarity
81/100
Strong
Creative
hero_v2_final.mp4
00:24 / 00:30
Hook
0:00–0:05
Peak
0:22
Drop
0:11
Sentiment by segment
Where it resonates
PosNeuNeg
  • Gen Z · Urban84% positive
  • Millennials · Suburban71% positive
  • Parents 35–5058% positive
  • Tech early-adopters79% positive
Run metadata
  • RanMar 12, 2026 · 14:22
  • Paneln=800 · 4 segments
  • Sources5 active
  • Verbatims800 collected
Actions
4 / 4
Maya, 24 · Gen Z Urban
Positive

The hook landed instantly — felt like it was made for my feed, not against it.

Jordan, 38 · Millennial Suburban
Neutral

I get the vibe but I'm not sure what they're actually selling until 0:18.

Priya, 44 · Parent
Negative

Music is great but the text is too small on my phone, I'd scroll past.

Devon, 29 · Tech early-adopter
Positive

Love the product shot at 0:22 — that's where I'd tap to learn more.

Use cases

What you'll put it to work on.

One instrument, pointed at the decisions that can't be A/B tested in the real world.

Creative testing

Pre-test ads, hooks, and hero assets before the spend.

Value proposition testing

Find the message that lands — and the reason it does.

Product launches

Rehearse the market's reaction before you ship.

Campaign strategy

Model reach, resonance, and the blind spots ahead.

Competitive positioning

See where you win, where you lose, where you're ignored.

Brand perception tracking

Watch how you're seen move over time and segment.

Crisis response

Stress-test a statement against the room before you send it.

Narrative tracking

Follow how a story spreads, mutates, and takes hold.

The record

Every agent is grounded in evidence.

Simulant is built on the official record — not vibes. Every response traces back to the signals behind it, so the readout survives the room, the audit, and the front page.

Census

National population, demographics, and geography

Official
Electoral studies

Voting behaviour and political attitudes

Longitudinal
Social cohesion

Trust, values, and community sentiment

Annual
Household panels

17,000+ people tracked over two decades

Longitudinal
World Values Survey

Global attitudes, values, and beliefs

Global
Media & comms

Channel reach and media consumption

National
The oldest question

What will people do?

We're a small team at the frontier of multi-agent simulation, behavioural modeling, and applied AI — building the instrument that answers it.