Almost every behavioral archetype model circulating in HR comes from the same place: a management book, a consultant who popularized them, a taxonomy published decades ago. They get applied across different companies, contexts and countries, and the centroids — the “average profile” of each archetype — are assumed to be universal.
They’re not.
When we started building Talen.to’s archetypes module, the most important decision wasn’t which archetypes to include. It was how to define the centroids. And the answer was: with data. With real assessments from real people in real roles, not with aspirational literature.
This post explains how the model works, the 10 archetypes, the classification algorithm, and why in-house calibration matters more than it looks.
Why archetypes, and not just OCEAN
OCEAN, in its six extended dimensions, is a rich representation of a person’s profile. But “rich” isn’t the same as “communicable.” A manager who reads “your candidate is O=72, C=58, E=44, A=66, EE=70, SR=51” doesn’t know what to do with that.
Archetypes are a translation layer. They reduce the multidimensional space to a namable pattern: “she’s an Analytical with a secondary Specialist.” That sentence doesn’t replace the OCEAN detail, but it makes it operable.
The design question was: in what space do we define the archetypes? Directly in the 6 OCEAN dimensions, or in a derived set?
The 7 traits of the model
We chose a derived 7-trait space computed from the full assessment. Each trait combines signals from several OCEAN items and Talento Index responses. They don’t replace OCEAN — they’re computed from it — but they offer vocabulary closer to management language:
| Trait | What it captures |
|---|---|
| Autonomy | Need for external direction vs ability to operate without supervision |
| Social Capacity | Comfort and effectiveness in interpersonal interactions |
| Pace | Preferred speed of processing and action |
| Conformity | Adherence to existing processes, norms and structures |
| Energy | Activation level, drive and persistence in action |
| Logic | Preference for structured analysis, data and deductive reasoning |
| Ingenuity | Ability to generate novel or unconventional solutions |
Each trait is scaled 1-5. Every person, on completing the assessment, gets a point in R^7 space.
The 10 archetypes
Each archetype is defined by a centroid in that R^7 space. Here are the ten, with the logic each one represents:
1. Connector
Builds bridges between people. Their social capital and influence are their strength. Shines in matrixed, multi-stakeholder contexts.
Dominant traits: high Social Capacity, medium-high Energy. Typical roles: account manager, HRBP, community manager. Risk: may avoid necessary conflicts.
2. Analytical
Thinks in data and method. Decomposes problems, validates with evidence, reaches defensible conclusions. Especially useful where others tend to jump to a solution without understanding the problem.
Dominant traits: high Logic, medium-high Conformity, relatively low Social Capacity. Typical roles: data analyst, process analyst, researcher. Risk: can stall decisions through over-analysis.
3. Entrepreneur
Turns ideas into action without waiting for permission. Most comfortable in ambiguity and resource scarcity. Ideal for opening markets, launching new products or running autonomous units.
Dominant traits: very high Autonomy, low Conformity, high Energy and Ingenuity. Typical roles: founder, business developer, intrapreneur. Risk: may neglect operations and delegate poorly.
4. Collaborator
Adds value when there’s a clear framework and a solid team around them. Reliable as an executor with frequent feedback, but performs worse in contexts where they have to set direction.
Dominant traits: low Autonomy, medium-high Social Capacity, near-neutral profile. Typical roles: junior analyst, functional support, team executor. Risk: can lose focus in ambiguous contexts.
5. Mentor
Multiplies team impact by developing others. Their success is measured by how the people around them grow.
Dominant traits: high Logic and Social Capacity, high Conformity, medium-high Autonomy. Typical roles: tech lead, senior tutor, head of training. Risk: may withhold information as a subtle form of power.
6. Specialist
Solves problems no one else can. Unique technical depth in a concrete domain.
Dominant traits: very high Logic, high Autonomy, low Social Capacity. Typical roles: subject matter expert, researcher, senior engineer. Risk: inflexibility outside their domain, opaque communication.
7. Executor
Turns plans into measurable results, on time and on budget. The profile the team needs when strategy is clear and operational discipline is missing.
Dominant traits: high Pace, high Conformity, high Energy, medium-high Autonomy. Typical roles: project manager, operations lead, delivery manager. Risk: may ignore the strategic “why.”
8. Strategist
Sees the forest before the trees. Connects business signals, people and context into long-range decisions.
Dominant traits: high Logic, high Ingenuity, high Autonomy. Typical roles: general manager, VP of strategy, senior consultant. Risk: disconnect from day-to-day execution.
9. Adapter
Their superpower is not getting locked into one way of working. Performs well in ambiguous roles, organizational transitions, or projects where context changes faster than it gets documented.
Dominant traits: balanced profile, no extreme peaks, low Conformity. Typical roles: external consultant, technical firefighter, transitional role. Risk: lack of a clear role identity.
10. Operator
Keeps day-to-day operations running. Respects processes, doesn’t deviate from standard, delivers predictable results.
Dominant traits: high Conformity, low Ingenuity, low Social Capacity. Typical roles: operations, customer service, administration. Risk: resistance to process change.
How classification works: Euclidean distance in 7D
The algorithm is deliberately simple, and that’s a decision, not a limitation. When you work with empirical calibrations, transparent models are auditable. A random forest or a neural net would maybe gain a little accuracy, at the cost of explainability. In a regulated domain like hiring, that trade-off doesn’t pay off.
The procedure:
- Compute the 7 trait scores for the candidate from the assessment.
- Compute the Euclidean distance between that point and each of the 10 centroids in R^7.
- Convert distance to matchScore with the formula
(1 - dist / √112) × 100, clamped between 0 and 100. (√112 is the maximum possible distance in R^7 with a 1-5 range.) - Rank the 10 archetypes by matchScore. The top one is the primary.
Confidence: gap between primary and secondary
If the primary wins by a wide margin, the classification is robust. If it wins by a small margin, there’s genuine ambiguity — the person lives between two archetypes — and that’s information the manager should see.
| Gap between primary and secondary | Confidence | Interpretation |
|---|---|---|
| > 15 points | High | Clearly dominant archetype |
| 8 to 15 points | Medium | Visible blend: the secondary is reported alongside the primary |
| < 8 points | Low | Mixed profile: look at topTraits to understand better |
The candidate report shows the primary, the secondary (if confidence isn’t high) and the 3 topTraits closest to the primary archetype’s centroid.
How the framework was built
Here’s the point that differentiates this model from the rest. The centroids weren’t invented by reading a book. They were computed from real assessments.
Our people science team ran the calibration exercise with complete field assessments — profiles covering development, operations, sales, training and management at contemporary LATAM companies. Each assessment was already tagged with qualitative observations from managers and peers about real behavior in the role.
The process:
- Exploratory clustering over the points in R^7, looking for natural groupings.
- Qualitative labeling of the resulting clusters against the archetype descriptions. Does this cluster match “Analytical”? Does this one match “Connector”?
- Manual centroid adjustment based on the geometric center of each cluster plus theoretical review. Theoretical review matters because some theoretically valid archetypes (Strategist) had little representation in the initial sample and we had to extrapolate with judgment.
- Validation against a held-out set of assessments: does the algorithm correctly assign people we already knew the archetype for?
The result lives as the source of truth for the module, with a versioned changelog of every recalibration.
Why it matters that the centroids are in-house
This is the part that usually gets under-communicated. And it’s where the model becomes an asset, not just a feature.
”Off-the-shelf” archetypes are calibrated on populations that aren’t yours
Most archetype taxonomies circulating in HR come from North American samples, pre-2000 blue-collar/white-collar populations, or academic psychometrics groups. They’re valid as a starting point. They’re not valid as a benchmark for a software company in Latin America in 2026.
Cultural norms around assertiveness, conformity and energy vary across countries. The role distribution in a software scaleup is radically different from a traditional manufacturer’s. Applying centroids built for another reality introduces systematic bias.
In-house centroids enable continuous recalibration
Every new candidate who enters the platform adds a point to the space. When a company accumulates enough assessments, its local centroids start to diverge from the global ones. That’s information, not noise.
The platform has per-company calibration tools: /calibration/assessments, /calibration/recalculate-all. After N months of use, an organization can recalibrate its archetypes against its real population. Managers hire better when archetypes are tuned to the company’s reality.
In-house centroids are defensible under audit
In 2024, the EU AI Act partially entered into force, classifying automated hiring tools as high-risk systems. New York’s Local Law 144 requires annual audits of any AEDT (Automated Employment Decision Tool).
Having documented centroids — with calibration date, authors, sample and procedure — is what separates an auditable system from a black box. When the audit comes, we can show how they were computed and what was done to validate them. No magic: just a JSON with coordinates and a changelog.
What it’s used for in practice
The archetype doesn’t make the hiring decision. That would be a serious error. Hiring is decided by the bundle: OCEAN fit + Values fit + Competency fit + manager context. The archetype is a narrative overlay that makes the profile operable.
Concrete use cases we see:
| Case | Use of the archetype |
|---|---|
| Pre-interview | The hiring manager sees “Analytical with secondary Specialist” and prepares questions that validate technical depth, not team leadership |
| Team composition | A team with 4 Operators and 0 Adapters has a rigidity risk against change. Only visible if the model is calibrated |
| Internal mobility | An Executor performing well in delivery can transition to Mentor if the topTraits align. Visible in the report |
| Development plan | A Specialist on a management track needs to develop Social Capacity specifically, not “leadership in general” |
| Candidate conversation | Sharing the archetype report with the candidate — where appropriate — improves the quality of the candidate’s own decisions |
What the model doesn’t do (and it matters to say so)
Any archetype model has limits. Ours:
- It doesn’t predict specific performance in a concrete role. That’s what the full fit score (OCEAN + Values + Competencies) does, not the archetype.
- It isn’t static. People move through R^7 space over time. Today’s archetype isn’t five-years-from-now’s.
- It doesn’t replace the manager’s judgment. It’s input, not decision.
- It isn’t universal. Calibrated with Latin American data + software/operations. Applying it to another population requires recalibration, and the platform supports it.
Saying this explicitly is part of operating responsibly in HR Tech under regulations that get stricter every year.
How it fits with the rest
Archetypes are the communication layer of the scoring engine. Underneath are extended OCEAN, company-configurable Values, Competencies in dual mode (soft + technical multi-evaluator) and a FitCalculator that combines everything into an overallFit.
If you want to see how the pieces connect: the hub post has the map. If you want the multi-evaluator competency deep dive: Multi-evaluator competency assessment. The module in the product: /features/archetypes.
Implement this with us
If your organization is using an archetype model from a 2008 management book, or a personality profile without local calibration, you’re leaving precision on the table. We help you run the model on your population, calibrate the centroids with your own data after N months, and connect archetypes with decisions on hiring, internal mobility and team composition.
Book a 15-minute demo and I’ll show you a real case.
Questions? Email me at clara@talen.to.
Related OCEAN+ profiles
Discover which personality dimensions to look for in each role.
Related Articles
Algorithmic transparency in HR Tech: why your scoring engine should be auditable
Most assessment tools are black boxes. Why that's bad practice, what regulation requires (EU AI Act, NYC AEDT), and what an auditable, configurable scoring engine looks like.
Multi-evaluator competency assessment: when the psychometric test isn't enough
A psychometric assessment tells you who the person is. A multi-evaluator competency assessment tells you how good they are at what the role actually demands. When you need one, when you need the other, and how they combine.
Why generic assessments fail at culture fit (and what to do about it)
Traditional assessments evaluate the candidate against a general population. Talen.to evaluates them against your role, your company, and your cultural context. Extended OCEAN, cultural factors, 10 archetypes calibrated with real data.