Savanna

Manual

What this is, how it works, why it works, and what you - or your team - actually get out of it.

What this is

Savanna is a self-report personality assessment that maps your trait pattern onto eight archetypes named for African animals (Lion, Elephant, Cheetah, Giraffe, Hyena, Meerkat, Rhino, Eagle). The questions are grounded in well-validated psychology frameworks - the Big Five, HEXACO, DISC, Belbin team roles, and CliftonStrengths - plus descriptive items drawn from ADHD and autism cognitive-style research.

The archetypes themselves are a framing layer, not a validated taxonomy. Use the result as a conversation tool, not a verdict. See Research for the full bibliography and limits.

How it works

  1. Take the assessment. 48 questions across five categories: personality, work ethic, delivery habits, neurodiversity-style cognitive items, and story scenarios. Mix of multiple choice, 1-5 Likert, and scenarios. Auto-saves as you go.
  2. Each answer carries weights. Every option you choose contributes weights toward one or more archetypes. For scaled items, agreement above the midpoint adds; disagreement below subtracts.
  3. Scores are normalised. Per-archetype totals are divided by the maximum positive contribution achievable across the questionnaire and scaled to 0-100. This is the normalised score you see on your report.
  4. Primary and secondary picked. Highest score is your primary archetype; second highest is the secondary. The gap between them controls how confident the read is - we label dominance as strong, clear, or blended.
  5. Insights are generated. The full report describes strengths, growth edges, work / delivery style, cognitive style, and the archetypes you tend to pair with vs. clash with.
  6. Cognitive-style spectrum scored. The neurodiversity items also feed three dimensional reads in parallel: ADHD-style, autism-style, and anxiety-style cognition (0-100 each). These are descriptive only - they detect self-reported preferences correlated with each style, never diagnose.
  7. Applied judgement scored. The 10 scenario items also produce a 0-100 SJT-style score - your alignment with consensus best-practice from the leadership and coaching literature on workplace situations. This is aptitude in the SJT sense, not raw cognitive ability.

Why a spectrum and not a label

The strongest empirical work on ADHD, autism, and anxiety treats them as continuous dimensions in the population, not categorical bins. Twin studies (Constantino & Todd, 2003; Ronald & Hoekstra, 2011), large-N epidemiology (Polderman et al., 2014), and the AQ literature (Baron-Cohen et al., 2001; Wakabayashi et al., 2006) all converge on this point.

So the report doesn't tell you "you are ADHD" or "you are not autistic." It tells you where on each dimension your self-reported preferences sit relative to the maximum positive score available. A high read on a spectrum is a starting point for self-awareness or a clinician conversation - never a clinical conclusion.

The spectrum items draw inspiration from validated screeners - the AQ (Autism-Spectrum Quotient, Baron-Cohen et al., 2001), the ASRS (Adult ADHD Self-Report Scale, Kessler et al., 2005), and trait-anxiety inventories (Spielberger, 1983) - but Savanna is not those instruments and is not a substitute for them.

Why it works

Four reasons.

  1. The trait foundations are solid. The Big Five model has held up under decades of factor-analytic and cross-cultural replication. HEXACO and ADHD/autism cognitive-style research add specific dimensions (Honesty-Humility, monotropism, hyperfocus) that the Big Five doesn't capture cleanly on its own. The questionnaire is built so each item moves a specific, defensible dimension.
  2. The format encourages honesty. Mixing scaled items with story scenarios surfaces real-world defaults rather than aspirational self-image. People will rate themselves "open to feedback" on a Likert; they reveal more by picking a behaviour in a scenario.
  3. The applied-judgement layer is grounded in SJT research. Situational judgement tests predict job performance with validity comparable to structured interviews (McDaniel et al., 2007; Christian et al., 2010), at lower adverse-impact cost than pure cognitive-ability testing. The 10 scenario items use the same method - consensus-best-practice keys against realistic workplace situations.
  4. The team layer is grounded in team-composition research. Meta-analyses (Bell, 2007; LePine et al., 2011; Mathieu et al., 2014) show that complementarity of roles predicts team performance better than mean trait scores. The cohesion / tension / diversity / coverage outputs are designed accordingly.

For individuals

The benefits an individual gets from a careful read of their result:

  • Self-awareness with structure. A vocabulary for what you are naturally good at and where the same instincts become liabilities under stress.
  • Concrete next steps. "Address first" surfaces the highest-leverage development priority for your archetype - the one thing to change this quarter rather than a vague growth list.
  • How others should work with you. The communication section gives your manager and teammates a script for briefing you, giving you feedback, and addressing conflict in a way that lands. Share it.
  • Stress early-warning system. Named, specific behaviours that show up when your archetype is overloaded. Easier to spot in yourself with practice; faster to fix when you have a vocabulary.
  • Career direction. Role-fit notes describe environments where you'll thrive vs. struggle. Useful when choosing roles, projects, or work environments.
  • Cognitive-style read without diagnosis. If you've ever wondered whether your focus pattern, sensory load tolerance, or routine-need is "normal," the cognitive-style spectrum gives you a descriptive read - not a label, but enough to ask better questions of yourself or a clinician.

For teams and businesses

What a company gets when teams take it together:

  • Team composition mapping. See the archetype mix across a team in one view - which strengths are present, which functions aren't covered.
  • Pre-mortem on conflict. The pair-by-pair analysis flags pairings likely to generate friction. Not "these people don't work" - rather "these two need explicit working agreements about X."
  • Each member, separately and together. The team page shows every member's individual contribution (archetype, role fit, communication tips, stress signals) before the team-level synthesis. Plan with both reads, not just the aggregate.
  • Organogram + structure fit. Build the team's reporting hierarchy and Savanna will tell you whether it plays to the team's archetypes (high structure fit = pair-friendly lines; low fit = friction-prone lines that need explicit working agreements).
  • Suggested team structure. Given the team's composition, Savanna proposes which members fit a visible lead shape, which fit a deputy / chief-of-staff shape, and which lead best as principal ICs. Discussion-prompt, not directive.
  • Invite by link or QR. Each invite generates a one-use link and a printable QR code. Useful for in-person onboarding, kickoffs, and remote-first teams.
  • Onboarding accelerator. A new hire who can articulate their archetype in week one starts conversations with their team three sprints earlier than they otherwise would.
  • 1:1 and coaching scaffolding. Managers get a vocabulary for trait-based feedback that doesn't drift into judgement. The "watch for", "stress signals" and "for your manager" sections are particularly useful here.
  • L&D programmes. The framework gives a shared language for development conversations across a programme cohort - cohorts can compare archetypes and discuss complementarity.

What this is not: a hiring or promotion tool. Personality inventories carry well-documented adverse-impact risks if used in selection. Use Savanna for development, self-awareness, and team conversations, not gating decisions. See Limits for more.

How to read your report

Four things to keep in mind:

  1. The gap matters. If your top two archetype scores are 5 points apart, treat the result as blended - the secondary will surface often. If 20+ points apart, it's a strong read of one archetype.
  2. The growth edges aren't insults. They are the same traits that make the archetype effective, viewed under stress or scale. A Lion's "decisiveness" and "steamrolling quieter voices" are the same trait observed in different lights.
  3. Cognitive-style notes are descriptive. "ADHD-style" or "autistic-spectrum-style" mean the pattern is consistent with that style - not that you have or don't have any condition. If anything in the read is striking to you, that's a signal to investigate, not a diagnosis.
  4. Applied judgement is SJT, not IQ. The score reflects alignment with consensus best-practice on management scenarios. It does not measure general cognitive ability and should not be used as a hiring filter.

Where to go next