What the instrument is, and what it is not yet

Current validation status

This instrument has not undergone formal psychometric validation. It is not a validated psychometric instrument, a clinical assessment, or a performance measure. Treat scores as a structured conversation input, not a measurement of individuals.

Pilot v0.2 has not completed psychometric validation. We have not established test-retest reliability, factor structure, or predictive validity at the standard required to call it a validated psychometric instrument. We are transparent about this status with every organization that uses it, and it is not to be represented otherwise in any customer-facing or internal material.

Binding Constraint: the primary interpretation

The core output of a diagnostic cycle is the Binding Constraint — the single condition, among the nine, that the evidence indicates is most limiting the team right now. This is deliberately not an average productivity score. Averaging across nine conditions would hide the one condition that actually matters for action, and BrainEng does not compute or present such an average as a headline figure.

The Binding Constraint is selected using condition means, dispersion (how divided the team's experience is), and the gap between leader and team perception, all defined in a versioned scoring configuration so that results stay interpretable over time. It is a prioritization signal for where to focus the next six-week experiment, not a certified measurement of organizational health.

Order of repair: what the engine does before it ranks

Selection is deterministic and applies a fixed precedence. It is not a search for the lowest score:

1. Fear gate. A materially adverse Fear-Load reading is selected first, overriding a lower raw mean on any other condition.
2. Direction gate. If Fear-Load clears, low Expectation Precision is examined next.
3. AI leverage, in parallel. Low AI Leverage Architecture is surfaced early as a separate mechanical-load opportunity. It is never an automatic override of the gates above.
4. Ranked binding constraint. Otherwise the constraint is ranked on level, between-person dispersion, leader-team gap and the strategic result stated at launch.
5. Human confirmation. A leader must confirm the reading with the team before an experiment can be designed or activated.
6. Reassessment. The next constraint comes from a fresh diagnostic, never from the previous reading.

Confidence statements attached to a recommendation describe only how stable this prioritization ordering is under the provisional Pilot v0.2 decision rules. They are not statements about psychometric validity, measurement accuracy, or causality.

Current decision thresholds

Provisional, configurable rules for Pilot v0.2 (1–5 scale):

• Fear gate: Fear-Load mean ≤ 3.0 is addressed first.
• Direction gate: Expectation Precision mean ≤ 3.2 is examined next.
• AI leverage opportunity: AI Leverage mean ≤ 3.0.
• Disputed reading: between-participant spread ≥ 0.9.
• Material leader gap: ≥ 0.7 points.
• Clear separation for confidence: ≥ 0.3.

Crosswalk: the book, this app, and Appendix G

The underlying book and this product are related but not interchangeable. Stated precisely, so nothing is implied that is not true:

• Governing model. Chapters 4–12 define the nine conditions, and the nine-condition pass in Chapter 14 is the governing diagnostic. The nine conditions used throughout this app are those conditions and nothing else — there is no eight-factor model here.
• What the app measures. Brainload Engineering Diagnostic — Pilot v0.2 is a separate 27-item pilot operationalization of those nine conditions, three items per condition, answered on a 1–5 agreement scale. It is a pilot instrument, not the book.
• Appendix G is a different instrument. Appendix G is a ten-item, 0–4 behaviorally anchored collective-state instrument. It differs from Pilot v0.2 in item count, scale, and anchoring approach.
• No equivalence claim. This app does not claim psychometric equivalence with Appendix G, and it does not silently implement Appendix G's Team Flow Propensity Index. Pilot v0.2 scores must not be read as Appendix G scores, or converted to them.
• Team Flow Propensity in-app. Any Team Flow Propensity view in this product remains a separate, exploratory, non-probabilistic and non-predictive lens, as described in the section above. It is not Appendix G's TFPI.
• Chapter 15 maps to the working loop. The Chapter 15 implementation discipline is realised in-app as Intervention Studio → Six-Week Flow Loop → Impact Review → a fresh diagnostic. A fresh diagnostic, not the previous reading, selects the next constraint.

Every claim above is bounded by the validation status stated at the top of this page and by the pilot disclosure shown wherever a score, mirror, brief or report is displayed:

This instrument has not undergone formal psychometric validation. It is not a validated psychometric instrument, a clinical assessment, or a performance measure. Treat scores as a structured conversation input, not a measurement of individuals.

Team Flow Propensity is a separate, exploratory construct

Some reports surface a construct we call Team Flow Propensity. This is exploratory and distinct from the Binding Constraint. It must never be described as a probability, a prediction, or a validated predictor of performance or outcomes. It is one additional lens for a team's own conversation, not a scored forecast.

What we do not claim

• We do not claim a validated return on investment or percentage of capacity recovered.
• We do not claim a guaranteed time to resolve a constraint. Six weeks is the design of the experiment cycle, not a guarantee of resolution.
• We do not collect passive telemetry, load monitoring, or any form of continuous behavioral surveillance.
• We do not produce individual performance rankings or automated classifications of people.

Ongoing work

As more organizations complete cycles, we intend to study reliability and construct validity properly, and we will update this page and the instrument version when that work changes what can be claimed.