en
Create account
How it works Sample report Methodology Pricing FAQ Sign in Create account

How it works

Your first test takes about a minute.

No panel recruitment, no fieldwork, no waiting for a field institute.

  1. 1

    Pick an audience

    Audience scope

    de-consumercountry pack
    eu-west-consumermacro region
    eu-consumerEU average
  2. 2

    Paste your copy

    Message

    Neue Hafermilch: gleicher Geschmack, 30% weniger Verpackung.

    running · one LLM call per segment

  3. 3

    Read the split

    DE21 · real output

    appeal 55%
    trust 25%
    skepticism 10%
    indifference 5%
    concern 5%
    PDFPrint / Save as PDF
Samplio · Single message

Neue Hafermilch: gleicher Geschmack, 30% weniger Verpackung.

Full distribution · DE21

appeal 55%
trust 25%
skepticism 10%
indifference 5%
concern 5%

01

Single message

Every segment returns a full reaction distribution and a short verbatim “voice of the segment” — in the audience's own language.

  • One LLM call per segment
  • Five reaction categories, one distribution each
  • Simulated in the audience’s own language

See a sample report

Samplio · Variant comparison

Variant A

Variant B

Variant C

Verdict

No clear winner

The lead is not wider than the wording effect measured on this run.

Illustration · sample data

02

Variant comparison

2–5 variants run on the identical audience with the same seed. You get a ranking — and a winner only when the lead is wider than the wording effect measured on that run.

  • Same audience and seed for every variant
  • A ranking, with the wording effect measured on the run
  • “No clear winner” when the lead is inside that effect

Read the methodology

Samplio · Conditional survey

Q1

How appealing is this product?

Very Somewhat Neither Not really

Q2 · if appealing

What is the main reason?

Taste Less packaging Price Other

Path logic (in code)

Q1 all respondents if appealing Q2 reason other branches · up to 12 questions

Illustration · sample layout

03

Conditional survey

Up to 12 single-choice questions with skip logic. Who reaches which question is computed in code, never estimated by the model.

  • Up to 12 single-choice questions
  • Skip logic resolved in code, before the model is asked
  • Always stamped uncalibrated — directional reads

Create free account

Coverage

From one country to the EU-27.

EU-27 · eu-consumer Macro region · eu-west-consumer Country pack · de-consumer
  • Country packs29 EU markets, each simulated in the audience's own language (8 languages).
  • Macro regions & EU averageWestern, Northern, Southern, Central and Eastern Europe plus the EU-27 mean — documented, editable groupings for directional reads.
  • Composite runsA macro-region query runs each member country separately and combines the results population-weighted. The weighting math lives in code, not in the LLM.
  • country packs
    29country packs
    Simulation languages
    8Simulation languages
    Audience profiles
    35Audience profiles

Get started

Test your idea today.

Create free account See a sample report