The Capability Reality Check
What this does: Surfaces the gap between what your people think they can do and what they can actually do — in both directions. The expertise nobody knows about, and the confidence nobody has tested.
The business outcome: You find hidden capability you don’t need to hire for, and you find overconfidence before it causes damage. Both are high-leverage discoveries that don’t require spending money or changing the org chart.
Quickstart
Pick a team or an individual. Ask two questions: “What does he think he’s good at?” and “What is he actually good at?” Where those answers diverge, you’ve found a Frame gap. Gaps where the “self-model” is too small mean unused capability. Gaps where the self-model is too large mean unmanaged risk. For detailed instructions see “How to use it” below.
Template
| Person / Team | What they think they can do | What they can actually do | Gap: invisible capability (self-model too small) | Gap: overconfidence (self-model too large) |
|---|---|---|---|---|
| e.g., Senior Engineer | API design, system architecture | API design, system architecture, deployment automation (10 years at previous company) | Deployment automation — nobody knows, including him (he thinks of himself as “the API guy” here) | |
| e.g., VP of Sales | Sales strategy, team management, market analysis, regulatory compliance | Sales strategy, team management, market analysis | Regulatory compliance — he thinks he understands it from reading trade publications, but he doesn’t have the depth to evaluate actual risk |
The two gap columns are where the diagnostic value lives. The left column is free capability you’re not using. The right column is risk you’re not managing.
How to use it
Step 1: Pick your target. This works for individuals and for teams. For individuals, you’re asking what the person thinks his capabilities are versus what they actually are. For teams, you’re asking what the team collectively believes it can handle versus what it can actually handle.
Start with whoever you’re most worried about — either because you suspect they’re underperforming (might have invisible capability) or because you suspect they’re overreaching (might have overconfidence).
Step 2: Map the self-model. What does this person or team believe they’re capable of? You can get at this through direct conversation (“What would you say your strongest capabilities are?”), through observed behavior (what they volunteer for, what they avoid, what they claim expertise on), or through their track record of what they’ve attempted.
Don’t judge the self-model yet. Just map it.
Step 3: Map the actual capability. This is harder. You need evidence beyond the self-report. Sources:
- Resume and prior roles — capabilities built in previous jobs that aren’t being used here. The senior engineer with deployment expertise nobody knows about.
- Peer observation — what do the people who work with him actually see? Sometimes a person’s peers know more about his capability than he does.
- Performance in adjacent situations — has he ever handled something outside his normal role and done well? Or handled something inside his claimed expertise and done poorly?
- External assessment — formal evaluation, customer feedback, audit results.
Step 4: Identify the gaps.
Invisible capability (self-model too small): Things the person can do but doesn’t think are relevant, doesn’t know he has, or hasn’t been asked about. This is the most common form of Trapped Intelligence at the individual level. The intervention is awareness — making the capability visible to both the person and his manager. Sometimes a single conversation (“Did you know we need exactly what you built at your last company?”) releases years of trapped capability.
Overconfidence (self-model too large): Things the person believes he can handle but can’t. This is harder to surface because the person will resist the finding. Look for:
- Decisions in domains where the person has no formal training or deep experience, but speaks with confidence
- A pattern of delegating poorly in specific domains (which suggests he doesn’t understand the work well enough to evaluate it)
- Outcomes that surprised him — situations where his predictions were consistently wrong in a specific area
The intervention depends on the severity. Mild overconfidence responds to evidence — an audit, a failure, a trusted advisor who pushes back. Deep overconfidence doesn’t respond to evidence because the person can’t model why the evidence matters — the instrument that would detect the blind spot is the same instrument that has the blind spot. For deep overconfidence, the intervention is structural: add a gate, a co-decision-maker, or a review process that doesn’t depend on the person recognizing his own limits.
Step 5: Act on the findings.
For invisible capability: the cheapest, highest-leverage move in organizational development. Make it visible. Create the context where it gets used. Sometimes this means a new assignment. Sometimes it means a conversation. Sometimes it means restructuring a meeting so the person with the invisible expertise is in the room when the relevant decisions are made.
For overconfidence: the hardest intervention because it feels personal. Frame it structurally, not personally. “The position’s reach extends into regulatory territory, and we need someone with regulatory depth in the loop for those decisions” is structural. “You don’t understand regulatory risk” is personal. Same intervention, different framing, very different reception.
Where to learn more
Other tools:
- The Trapped Intelligence Diagnostic — The Capability Reality Check is a deep dive into Checkpoint 4 of this diagnostic. If you haven’t run the full diagnostic, start there.
- The Franz Ferdinand Check — If you found overconfidence, this maps where it creates blast-radius risk.
- The Team Trade-Off Matrix — If the invisible capability you found would change the team’s dynamics, this shows you how.
Vocabulary: Frame, Capacity, Realized, Trapped Intelligence, Cognitive Boundary