Build a brand from a blank page — the 7-step workflow you can run in Claude
Go from a blank page to a full brand book with seven skills, each feeding the next.
Showcase · Skills· Market & user research
Run customer interviews before you have customers
Turn $25–65k, 6–12-week market studies into continuous, affordable insight. Four free Agent Skills build synthetic personas, run interviews, find cohort patterns, and turn findings into roadmap actions. Learn the method and run it in Claude.
Free to try. One click loads the skill into Claude or ChatGPT, with no account and no setup.
What is synthetic-persona research — and can you trust it?
Traditional market research is expensive and slow: a proper study runs $25–65k and 6–12 weeks, which means most teams do it rarely, late, or never. Synthetic-persona research replaces the human panel with AI personas built to simulate real customer responses — so you can test a concept, a message, or a price point this afternoon instead of next quarter.
Can you trust it? With honest limits, yes. Personas built on the three-layer architecture hit 76–85% accuracy against human panels — strong enough to narrow options, validate direction, and kill bad ideas cheaply before you commit real research budget. The right framing is supplement, not replace: synthetic research for continuous, fast, cheap iteration; human research to confirm the high-stakes calls.
The framework behind it
The skill isn’t magic. It encodes a proven framework anyone can learn. Here’s the method itself, so you understand what it’s doing for you.
The three-layer persona architecture
synthetic-persona-creatorA synthetic persona is only as useful as it is realistic, and realism comes from building it in three layers. Layer 1 — identity foundation: demographics (age, income, location, occupation, household). Layer 2 — psychographics: values, motivations, anxieties, decision drivers. Layer 3 — context: the situation and constraints that shape how this person actually behaves. A persona built on all three responds like a real segment; one built on demographics alone is a caricature.
The payoff is measurable: personas built this way respond to surveys and interviews with 76–85% accuracy versus human panels — enough to test concepts, messaging, and pricing before you spend on expensive human studies. They supplement human research, they don’t replace it — but they turn research from an episodic $25–65k, 6–12-week event into a continuous, always-on feedback loop.
The skills that get you there
Use any one on its own, or chain them. Each hands its output to the next. Every one is free in the public library.
Synthetic Persona Creator
- synthetic-persona-creator
Builds AI personas from a three-layer architecture (identity, psychographics, context) that simulate real customer responses at 76–85% accuracy vs. human panels.
Synthetic Interview Conductor
- synthetic-interview-conductor
Runs qualitative interviews against those personas — concept, message, and pricing tests — without recruiting a single human.
Persona Cohort Analyzer
- persona-cohort-analyzer
Finds the statistical patterns across a cohort of personas, so you see segment-level signal, not just anecdotes.
Research-to-Action Bridge
- research-to-action-bridge
Turns the insight into concrete roadmap and campaign actions — closing the gap between “interesting finding” and “what we do next.”
When you’d reach for it
Any time you need customer signal faster or cheaper than a human study allows — especially early, when you don’t yet have customers to ask. It’s for narrowing options and validating direction before you spend on the expensive stuff.
- Pre-launch: testing concepts, messaging, and pricing before you have users
- Seed-stage founders who can’t afford a $50k research study
- Product teams who want an always-on feedback loop between real studies
- Marketers validating campaign angles across demographic segments
What this research suite does
It lets you do seed-stage market research before you have a market: build persona cohorts, run qualitative interviews against them, find the statistical patterns, and turn the insight into roadmap actions — all without recruiting a single human panel.
The chain runs creator (build the personas) → interview conductor (ask them) → cohort analyzer (find the patterns) → research-to-action bridge (decide what to do). It’s novel enough to earn attention on its own, and cheap enough to run continuously.
Try asking it
Real prompts, and what the skill hands back.
You ask
“Create a cohort of 20 personas representing small-clinic office managers, then interview them on whether they’d pay $99/mo for automated eligibility checks.”
What comes back
A three-layer persona cohort plus interview responses on price sensitivity and purchase intent — signal you can act on before a human study.
You ask
“Here are the interview responses. Find the patterns by segment and tell me what to change on the roadmap.”
What comes back
A cohort analysis surfacing segment-level patterns, bridged into concrete roadmap and messaging actions.
What makes it credible
Three-layer personas
Identity, psychographics, and context — realism comes from all three, not demographics alone.
76–85% accuracy
Against human panels — enough to narrow options and kill bad ideas cheaply.
Continuous, not episodic
Turns a $25–65k, 6–12-week study into an always-on loop you can run any afternoon.
Supplement, not replace
Synthetic research for fast iteration; human research to confirm the high-stakes calls.
Ends in action
The chain doesn’t stop at insight — it bridges to concrete roadmap and campaign decisions.
How the four chain together
The suite runs creator → interview conductor → cohort analyzer → research-to-action bridge — from building the personas to deciding what to do with what they tell you. Run the whole loop, or just the piece you need (e.g. the creator to seed personas for a brainstorm).
It pairs naturally with the brand workflow — real audience personas sharpen the synthetic ones, and the research feeds back into positioning.
Who it’s for
- Founders validating a market before they have customers
- Product teams who want continuous signal between expensive studies
- Marketers testing messaging and pricing across segments, fast
Do it yourself — run synthetic research
You can build personas and run a research pass entirely in Claude:
Point Claude at the right skills
MCP server- agentman_skills
Already connected your library to Claude? Tell it: “Use the synthetic-persona-creator skill (and the others above) from the agentman_skills MCP server.”
Not connected yet? The Try in Claude / ChatGPT button loads the first one from the public library, with no setup. Or connect your library.
Load the first skill
Click “Try in Claude / ChatGPT” above, or — if you’ve connected your library — tell Claude: “Use the
synthetic-persona-creatorskill from theagentman_skillsMCP server.” Add the others by slug as you go.Build a cohort
Run
synthetic-persona-creatorwith your target segments. Give it real detail across all three layers — identity, psychographics, context — for realistic personas.Interview them
Run
synthetic-interview-conductorto test a concept, message, or price against the cohort — the questions you’d ask a human panel.Find the patterns
Feed the responses to
persona-cohort-analyzerfor segment-level signal instead of one-off anecdotes.Decide what to do
Run
research-to-action-bridgeto turn the findings into concrete roadmap or campaign actions.Confirm the big calls
Treat it as a supplement — use synthetic research to narrow and validate, then confirm the high-stakes decisions with real users.
That’s a market-research function you’d normally outsource — now a continuous loop you run yourself.
Try Synthetic Persona Creator right now
One click loads the skill into Claude or ChatGPT. No account, no setup.
FAQ
Synthetic Persona Creator: common questions
Are synthetic personas accurate enough to trust?
Built on the three-layer architecture, they respond at 76–85% accuracy versus human panels — strong enough to narrow options, validate direction, and kill bad ideas cheaply. The right use is to supplement human research, not replace it: synthetic for fast iteration, human to confirm high-stakes calls.
Do I need all four skills?
No. Run just the creator to seed personas for a brainstorm, or the interview conductor for a quick concept test. Chained, they form a full research loop from persona to roadmap action.
How is this cheaper than real research?
A traditional study runs $25–65k over 6–12 weeks. Synthetic personas run in an afternoon at a fraction of the cost, so you can research continuously instead of once a year — and save the expensive human studies for the decisions that truly need them.
Do I need an account to try it?
No. Click “Try in Claude / ChatGPT” above and the first skill loads from our public library — no setup. Clone the skills to customize and share them.