How the Assessment Scores Your Role
The full assessment's scoring engine is fully transparent. Here's exactly how 90 questions become a ranked list of 23 roles — step by step, verifiable with pencil and paper.
Two modes, one goal
There are two ways in, and they measure differently — on purpose:
The 2-minute express (free teaser)
A short read ranks your six role families— which is robust from few questions, because families aggregate signal. Then you pick the concrete flavour yourself: a category-specific “which of these is you?” question where each option is a real role. The result is an honest self-pick, not an inferred score. It's fast and directional — a rough top-3 — because the fine distinction between sibling roles is exactly what a 2-minute test can't infer, so we ask it directly.
The full assessment (the engine below)
90 questions, five weighted dimensions, all 27 units scored inferentially. This is what the rest of this page describes — and it's what validates or correctsyour express self-pick: you might think you're one role, and your full capacity-and-dimension profile might say another.
Three question sections
The 90 questions are organized into three sections with different purposes:
Section 1: Capacity (C1–C9)
Nine capacity questions assess your current situation — time availability, capital access, risk tolerance, technical background. Answers here can generate capacity flags: hard blocks (a role is structurally incompatible with your situation) or soft flags (possible but constrained). These flags appear on your results page alongside each role card.
Section 2: Assessment (Q1–Q78)
The core of the scoring engine. 76 questions map to one or more roles across five equally weighted dimensions. Each question uses a 7-point Likert scale (Strongly Disagree → Strongly Agree) or a Yes/No variant with a follow-up enjoyment rating. Every role has dedicated anchor questions — including capability items for roles like the Lawyer and the Talent Scout that older versions measured only indirectly.
Section 3: Founder Profile (F1, F3–F6)
Five questions specifically about entrepreneurial orientation — autonomy vs. scale ambition, appetite for investor overhead, lifestyle-business attitude. These fine-tune the Micro-Founder vs. Scaler distinction.
23 roles, 27 scored subtypes
The playbook defines 23 roles across 7 categories— that's the headline number you'll see everywhere. But four of those roles split into two subtypes each, because the two paths diverge enough (different daily work, monetization, and failure modes) that scoring them as one would be misleading:
- The Scientist → Tech Genius / Working Researcher
- The Founder → Micro-Founder / Scaler
- The Consultant → Strategy / Implementation
- The Freelancer → Strategy / Implementation
So the engine scores 27 distinct units (23 roles, with those four split in two). Your results rank all 27; on the overview they group back into the 23. A 7th category, the Charlatan, is an anti-type — the pattern to recognize and avoid — and is never scored, so it never appears in your results.
Five weighted dimensions
Every question is tagged to one of five dimensions. Scores within each dimension are normalized and then multiplied by the dimension's weight:
| Dimension | Code | Weight |
|---|---|---|
Pain Tolerance How much discomfort, uncertainty, and adversity you can sustain | PT | 20% |
Inspiration / Energy What activities energize vs. drain you | IE | 20% |
Dispositions Thinking style and lifestyle/scale preferences — how you naturally operate | DI | 20% |
Experience What you've actually done before | EX | 20% |
Skills Specific technical or professional competencies | SK | 20% |
All five dimensions carry equal weight (20%) — deliberately simple and easy to reason about. The differentiation between roles comes from the question weights instead: each question-to-role mapping carries a magnitude (10 = core anchor, 7 = strong, 5 = normal, 3 = adjacent signal), positive or negative.
The scoring formula
For each question mapped to a role, the contribution is calculated as:
effective_answer = direction==+1 ? raw : (8 − raw)
normalized = (effective_answer − 1) / 6 → range 0.0–1.0
contrib = normalized × question_weight
The direction field handles reversed questions — some questions are positively correlated with a role (+1), others negatively (−1, meaning a low score is a strong signal).
Per dimension per role, all contributions are summed and normalized by the maximum possible score for that dimension. Then each dimension score is weighted and summed to produce the role’s total on a fixed 0–100% scale — the share of that role’s maximum possible score your answers reached. There is no final rescaling: the top role is not forced to 100% (a strong match might land near 90%, a lukewarm one in the 60s), and because every role’s maximum is identical for everyone, the percentages are directly comparable across test-takers — the same answers always produce the same scores. The ranking is simply these scores sorted high to low.
Transparent breakdown on your results page
After completing the assessment, each role card on your results page has a collapsible accordion: “What shaped this result?” It shows the specific questions that most strongly contributed to that role's score, along with your answer and which dimension it fed into.
This is by design. The goal is that you can look at your top role's contributors and recognize yourself — or spot a misalignment that suggests a different interpretation.
Full audit CSV (admin)
Every intermediate calculation is logged in a per-submission audit CSV available to administrators — effective answer, normalized score, per-dimension sums, weighted totals. The entire scoring pipeline can be reproduced from the raw data.
See the scoring engine applied to your own answers.
Take the Free Assessment →
Dr. Bastian Brand, Ph.D. — author of The Hype Cycle Playbook, the framework behind the roletype assessment and this blog. About the author →