A concept test is the fastest way to get a directional read from your target market. You give each persona a description of your concept — a product idea, a packaging redesign, a campaign tagline, or a pricing proposal — and each one responds as they genuinely would: in their own voice, with their own concerns and motivations.
Use a concept test when you need to answer questions like:
- Will this new product resonate with our target segment?
- Which of two messaging directions lands better?
- Does this price point feel acceptable, or does it create friction?
- How will consumers react to a reformulation or repackaging?
How it works
When you create a concept test, every persona in your group independently reads your prompt_question. Boses generates a structured response from each persona containing:
| Field | Description |
|---|
| Reaction | A candid response in the persona’s own voice — how they’d actually talk about this with a friend |
| Sentiment | A single label: Positive, Neutral, or Negative |
| Key themes | The 2–4 ideas that stood out most to this persona |
| Notable quote | A presentation-ready line you can pull directly into a deck |
After all personas have responded, Boses generates an aggregate result across the group that includes:
- Sentiment distribution — the breakdown of positive, neutral, and negative responses as percentages
- Top themes — the ideas that surfaced most frequently across all personas
- Strategic recommendations — concrete next steps based on what the group told you
Running a concept test
Create the simulation
Send a POST request to /simulations with simulation_type: "concept_test" and your prompt_question. The simulation starts immediately as a background task.curl -X POST https://api.temujintechnologies.com/api/v1/projects/<PROJECT_ID>/simulations \
-H "Authorization: Bearer <TOKEN>" \
-H "Content-Type: application/json" \
-d '{
"persona_group_id": "<GROUP_ID>",
"simulation_type": "concept_test",
"prompt_question": "We are launching a premium instant noodle range called Mie Emas at 18,000 IDR per pack — roughly 3× the price of Indomie. The packaging uses matte black with gold typography and positions it as a weekday treat for busy professionals. How would you respond to seeing this in your local Alfamart?"
}'
The response includes the simulation id. Save it as SIM_ID. Poll for completion
Concept tests run in the background. Poll the simulation endpoint until status is "complete". The progress field shows how many personas have responded so far.curl https://api.temujintechnologies.com/api/v1/projects/<PROJECT_ID>/simulations/<SIM_ID> \
-H "Authorization: Bearer <TOKEN>"
Example response while running:{
"id": "sim_01j...",
"simulation_type": "concept_test",
"status": "running",
"progress": {
"completed": 3,
"total": 8
}
}
Fetch the results
Once status is "complete", retrieve per-persona responses and the aggregate report.curl https://api.temujintechnologies.com/api/v1/projects/<PROJECT_ID>/simulations/<SIM_ID>/results \
-H "Authorization: Bearer <TOKEN>"
Example result (abbreviated):{
"individual": [
{
"persona_name": "Rina Kusuma",
"sentiment": "Positive",
"reaction": "Ooh, this feels like it's finally for me — not just the usual budget option. I'd try it at least once to see if it's worth it for a late night at the office.",
"key_themes": ["premium positioning", "treat occasion", "professional identity"],
"notable_quote": "It feels like it's finally for me."
}
],
"aggregate": {
"sentiment_distribution": {
"positive": 62,
"neutral": 25,
"negative": 13
},
"top_themes": [
"premium self-treat occasion",
"price sensitivity vs. aspiration",
"packaging as social signal",
"skepticism about taste vs. price ratio"
],
"recommendations": "The concept resonates strongly with aspirational consumers who associate premium packaging with personal reward. Lead with the 'treat yourself on a busy day' positioning rather than nutritional claims. Consider a trial 2-pack at 30,000 IDR to reduce first-purchase risk."
}
}
Reading the aggregate result
The aggregate result synthesises what the whole group told you. Focus on three things:
Sentiment distribution — Is the majority positive, mixed, or divided? A split result (e.g., 45% positive / 40% negative) is as informative as a clear winner: it tells you the concept polarises, which segments are drawn in, and which are not.
Top themes — These are the ideas your personas kept coming back to, regardless of sentiment. A theme like “price sensitivity” appearing in both positive and negative responses tells you the price is front of mind — but not necessarily a dealbreaker.
Recommendations — These are concrete and actionable. They synthesise the themes and sentiment into directional guidance for your next step: refine the concept, run a focus group to probe a specific theme, or proceed with confidence.
Tips
The richer your prompt_question, the richer the persona responses. Include the price point, describe the packaging or visual treatment, name the channel (e.g., “in your local 7-Eleven”), and mention any competitors you are positioning against. Vague prompts produce generic reactions.
After a concept test, run a focus group to let personas discuss the concept with each other. Group dynamics will surface tensions and trade-offs that individual responses miss.
Use a briefing to upload a product image, campaign mockup, or PDF deck before creating the simulation. Boses will incorporate the briefing content into every persona’s context.