At a high level:
- Synthetic In‑Depth Interviews (IDI) and Synthetic Focus Groups (FG) help you understand how people reason
- Synthetic Surveys (Synthesis) help you understand how earlier learning holds up across perspectives
They answer different questions, and they should be interpreted differently.
When you’ll see each type of report
- You’ll see IDI and FG reports immediately after running these studies
- You’ll see a Synthesis report only after you’ve run one or more IDIs and/or FGs that share the same Decision To Support (ie, belonging to the same lineage)
IDI and FG reports are inputs into synthesis, not alternatives to it.
*This guide focuses on IDI and FG outputs.
Synthetic Survey (Synthesis) works differently, creating insight by aggregating signals across multiple studies in a lineage. We will cover that separately in a subsequent post.
Reading IDI and Focus Group Results
What IDI and FG reports are for
IDI and FG studies are designed to surface:
- individual reasoning
- motivations and hesitations
- trade-offs people make
- contradictions and tensions
They are qualitative by design. You should read these reports as exploratory and directional, focused on how people think something.
Local Language & Translation
All IDI and Focus Group studies in Project Brainstorm are conducted in the local market language.
This is intentional. Running studies in the local language helps preserve the way people naturally think, express trade-offs, and frame everyday constraints, even in simulation.
In practice, this allows the simulation to reflect:
- culturally specific ways of reasoning
- everyday phrasing and priorities
- how decisions are actually discussed in real-life contexts
Translations to English are provided to make the output accessible. But the original language is important because it adds an additional layer of psycho‑cultural realism that is often lost when simulations are conducted directly in English.
Key sections to focus on
(1) Transcript – how people actually think
At the foundation of every study is the transcript.
- In an IDI, this is a one-to-one conversation between Lisa, the moderator, and a single persona
- In a Focus Group, this becomes a multi-person discussion where Lisa and up to 6 personas interact
The transcript shows how a persona reacts, reasons, and revises their own thinking in real time. This is where insight actually starts.
For example:
- initial reaction: “it seems quite cute”
- followed by reflection: “it’s not that simple”
- followed by constraint: “time, space, parents”
This progression, from reaction to reflection to constraint, is often the most important signal in the study.
(2) Executive Summary
Start here for a compact overview of what mattered most and why it matters to the decision.
This section is a synthesis of what emerged, not a replacement for reading the reasoning beneath it.
(3) Reasoning Chains & Mental Models (IDI)
This shows how participants arrived at their views.
- not just what they said
- but the logic and trade-offs behind it
This is where you see how decisions are actually formed, not just the final position.
(4) Themes & Motivations
Themes describe recurring patterns. Motivations explain why those patterns exist.
- psychological drivers
- social context
- real-world constraints
This layer connects observable reactions to underlying structures, which lets us see what actually drives behaviour.
(5) Hesitations, Overrides & Trade-offs
This is where complexity shows up.
Examples:
- attraction vs practicality
- curiosity vs commitment
- interest vs feasibility
These are not flaws, they are signal. Most real decisions live here.
(6) Contradictions, Tensions & Opinion Shifts
Pay close attention here. Tension often tells you more about decision risk than agreement does.
For example:
- “It’s cute” vs “I wouldn’t actually keep it”
- “I’d look at it” vs “I wouldn’t commit to it”
These phrases indicate gaps between perception and action, often the most important risk signal.
In Focus Groups, you may also see opinion shifts and norm enforcement, where individuals adjust or reinforce their views in response to others.
- Opinion shifts occur when exposure to peer reasoning causes someone to reconsider or soften their original stance
- Norm enforcement occurs when the group collectively pushes toward what feels more “realistic”, “responsible”, or socially acceptable
For example, in a study you might see:
- initial reactions leaning toward interest in a topic
- but as other raise practical concerns (maybe space, responsibility, parents, etc), the group quickly converges toward feasibility constraints as the dominant frame
This is not noise, it is signal.
It shows:
- which views are fragile vs reinforced
- how social context reshapes individual thinking
- where real-world decisions are likely to land when discussed in families or peer groups
(7) Signal Strength labels
- Strong: surfaced consistently and grounded in transcripts
- Indicative: directional, but not stable yet
- Exploratory: early or narrow signals to treat cautiously
These labels tell you what is stable versus what still needs more studies within the lineage.
(8) Ideas That Failed Under Scrutiny
This section highlights ideas, assumptions, or claims that looked promising early on but did not hold up under deeper reasoning or group discussion.
For example:
- a concept that seems appealing at first glance
- but breaks down when practical constraints are introduced
- or is challenged by other personas
This is especially valuable because it shows not just what works, but what fails when tested against reality.
In Focus Groups, these often emerge when:
- one persona proposes an idea
- others question feasibility
- the group collectively rejects or weakens it
(9) Supporting Voices
This section anchors findings in verbatim excerpts from the transcript.
These are not random quotes, they are:
- selected to represent key themes
- tied to specific personas
- used to ground interpretation in observable responses
Always read this to sanity-check whether the interpretation matches the raw voice.
(10) Risks, Limits & Knowns
Every study clearly defines what it does not tell you.
This may include:
- limited persona coverage
- missing stakeholder perspectives (eg, parents vs children)
- untested alternatives
- constraints of the simulation
This section tells you what still needs to be validated before acting.
Examples might include:
- findings that reflect simulated responses, not real-world prevalence
- certain perspectives (eg, parents) may not be directly represented
(11 - 13) Grounding, Sources & Context
These sections explain how the study is grounded:
Grounding Coverage Summary
- shows whether all key inputs (Problem, Decision, Market, Audience, etc) were defined
- confirms the study had sufficient grounding before execution
Sources & Citations
- clarifies that findings are simulation outputs, not external evidence
- explains how grounding sources were used (for realism, not proof)
Context Sources (NOT Evidence)
- shows what background narratives personas may have been exposed to
- does not validate findings
Together, these sections support transparency, not argumentation.
(14) Persona Summaries
This section provides a condensed overview of each persona, including their background, context, and decision-making tendencies.
Each persona summary is not the full profile, but a readable synthesis designed to help you quickly understand where their responses are coming from.
Every persona is associated with a unique Persona ID, which links back to the full persona profile in your My Personas library. There, you can access a more detailed background, including deeper contextual attributes and the underlying model configuration used to generate that persona.
This linkage is important. It allows you to:
- trace responses back to specific profiles
- revisit and reuse personas across studies
- understand how different personas contribute to patterns in the lineage
(15) Decision Guidance & What to Explore Next
The final section connects everything back to the decision.
It typically includes:
- what patterns matter most for the decision
- what still needs testing
- what would invalidate current thinking
Finally, “What to explore next” is especially important. This section defines the next step in the lineage, not a final answer.
The sections below move from interpretation to decision support. They will help you understand what holds, what fails, and what to do next.
How not to read IDI/FG results
- Do not treat them as representative samples
- Do not count quotes or themes as prevalence
- Do not assume “Strong” means “proven in market”
IDI and FG outputs help you understand the landscape of reasoning, not finalize decisions.
How to use results together
A healthy Project Brainstorm workflow looks like this:
1. Use IDIs – more than one -- to explore and refine individual reasoning
2. Use FGs to see which views hold up socially, repeating when new tensions appear
3. Use Synthesis to summarize directional patterns across the lineage
4. Decide what needs real‑world validation
5. Iterate, iterate, iterate
At every step, findings are meant to inform judgment, not replace it.
What comes next
This guide focuses on IDI and FG outputs.
As we mentioned earlier, Synthetic Survey (Synthesis) works differently. It aggregates signals across multiple studies in a lineage. We will cover that separately.
A final reminder
All Project Brainstorm outputs are:
- synthetic
- exploratory
- designed to support early‑stage decision‑making
They help you see the market more clearly, not make decisions for you.
---------
We’ll continue publishing additional guides and FAQs on lineages, synthesis, personas, and interpretation to help you get the most out of the platform. Use the platform to explore, iterate, and let the market population surprise you -- before committing to action.
Remember: Project Brainstorm is an experimental beta. It is designed to help you see the market more clearly, not to replace judgement or downstream validation.
If you’d like to learn more about Project Brainstorm or share feedback on the beta, you can reach us at contact@projectbrainstorm.xyz