Performance Validation
Diagnostics
Two modelled estimates, built from figures you already publish. Every assumption is visible and overridable – open either calculator below and replace them with your own.
Metric one
Estimating the economic impact of suboptimal decision quality.
Revenue that live human judgement touches, multiplied by how often that judgement goes materially wrong, by what a wrong call costs, by how much of it is not caught before it lands.
Only the four percentages are assumptions, and they are deliberately conservative. Everything else comes off the P&L.
Modelled annual behavioural leakage
—
—
A 10% reduction is worth — a year.
An illustration of the model, not a commitment.
BAL = E × F × S × R
—
Decision-exposed revenue
—
—
Failure incidence
—
—
Loss severity
—
—
Realisation factor
—
—
Check the working
—
Each band is the revenue still in play after the factor above it. The last band is the leakage.
Worked example uses the FY2024 published accounts of a NASDAQ-listed brokerage. A modelled estimate, not a measured result. Homebridge Alliance does not claim to reduce this figure; the model sizes an exposure that already exists.
Metric two
What turnover costs, from four figures a CFO already has.
Four inputs a CFO already has: average salary, replacement multiple, headcount, attrition rate.
Most organisations have never seen it expressed as one number.
Modelled annual attrition leak
—
—
A 10% reduction is worth — a year.
An illustration of the model, not a commitment.
Productivity lost to disengagement
—
—
Cost of the exits
—
—
Cost of backfilling them
—
—
Check the working
—
A modelled estimate built from your own figures. It sizes a value pool that already exists in the organisation. Homebridge Alliance does not claim to reduce it.
The limit: Behavioural findings can be mapped to agreed client-owned risk, conduct and people indicators, but Homebridge Alliance does not yet claim to reduces misconduct, attrition, complaints, regulatory events or financial loss.
Both figures size exposures that already exist. Neither is a forecast.
What we measure
We evaluate our programs through objective, independent indicators mapped during discovery. All individual profiles and simulated exercises are designed around validated behavioral proxies, including standardized psychosocial hazard scores.
How
We take baseline measures at the start (T+0), track progress across the 77-day journey, and measure long-term change against verified post-programme baselines. The aim is to demonstrate a targeted 10% improvement in executive capability.
The benchmarks that matter
55%
of board time is now spent on risk and compliance.
Directors say that time would otherwise go to strategy, long-term direction.
58%
of directors say compliance and regulation have changed their board’s risk appetite.
2×
compliance roles have doubled since 2010, up 105% against 30% growth in total employment.
Order is the intention. Uncertainty is the condition. Human performance is the exposure. Growth is the opportunity.
A meta-analysis of 335 leadership training studies covering 26,573 leaders identified the design features that separate development which transfers into the workplace from development which does not.
Lacerenza, Reyes, Marlow, Joseph & Salas (2017), Journal of Applied Psychology, 102(12), 1686–1718. Transfer δ = .82, organisational results δ = .72, no evidence of publication bias.
Five factors · 67.4% of the variance
Five things coaching changed in these executives. The one involving their own manager came last.
And half reported no change at all in how that manager communicated back.
Executive coaching: It works! Consulting Psychology Journal: Practice and Research, 60(1), 78–90. 114 executives and 42 coaches in one global pharmaceutical company. Principal component analysis, α = .95, isolated five factors accounting for 67.4% of cumulative variance. 98% gained insight into how their actions are perceived and 91% reported better results with direct reports, but only about 50% saw improvement in how their supervisor communicated back.
LimitsSelf-reported. One company. No control group. Variance in a questionnaire, not in outcomes.
HBA works on the system, not only the leader
Self-motivated managers still stall
The most intrinsically driven managers in the unsupported group did no better, and on some measures worse.
Wanting to grow is not the same as being able to do it alone.
Do I need coaching to achieve a learning goal? Coaching, goal orientation, and goal framing: A randomized control trial study. Current Psychology, 44(16), 13574–13590. Randomised controlled trial, 202 managers across 13 organisations, 43 external coaches.
LimitsThe mastery finding sits just outside significance (p = .087). One country. Self-rated at a single point.
HBA builds the scaffold they are missing
686 executives · 300+ organisations
Asked in their own words what coaching changed, senior executives did not describe insight. They described running a team.
Executive leaders' perspectives on the impact of executive coaching on their leadership development. European Journal of Training and Development, 50(10), 127–145. 1,130 open-ended responses from 686 executives and 158 coaches across more than 300 organisations, predominantly Fortune 500, spanning North America, EMEA, Asia-Pacific and Latin America.
LimitsCoded descriptions, not measured outcomes. Lead author runs the firm studied.
Backed by Science, Delivered by Experience.
Leadership relationship quality is associated with people actually leaving, not merely intending to.
ρ = −.24 [−.29, −.19] · 42 samples · 28,637 employees · actual voluntary turnoverSurveying the forest: A meta-analysis of the antecedents of voluntary employee turnover. Personnel Psychology, 71(1), 23–65.
LimitsAssociation, not cause. North American samples.
Challenge sits here because it is what the four days actually do. No meta-analysis links it to attrition, so it is marked as practice rather than evidence.
Psychosocial work conditions predict future sickness absence, measured forward in time rather than at a single moment.
burnout OR 2.34 [1.59–3.45] · low job satisfaction OR 1.92 · low decision authority OR 1.49 · 20 prospective cohortsA meta-analysis of observational studies identifies predictors of sickness absence. Journal of Clinical Epidemiology, 60(11), 1105–1115.
LimitsJob conditions, not leader behaviour. Prospective, not experimental.
Lost productivity is not listed here. We could not find a meta-analysis linking these levers to a presenteeism outcome, so we do not claim one.
Safety climate is associated with the incident count itself, across the largest study of its kind.
safety climate → accidents and injuries ρ = .24 · leadership → burnout ρ = −.36 · 203 samples · 186,440 workersSafety at work: A meta-analytic investigation of the link between job demands, job resources, burnout, engagement, and safety outcomes. Journal of Applied Psychology, 96(1), 71–94.
LimitsIndustrial workforces. The read across to psychosocial risk is by analogy.
Where followers see their leader as ethical, self-reported deviance is lower.
workplace deviance ρ = −.34 · 8 samples · 1,807 people · overall review 134 samples, 54,920 peopleA meta-analytic review of ethical leadership outcomes and moderators. Journal of Business Ethics, 139(3), 517–536.
LimitsPerceptions of ethical leadership, not audited conduct.
Role ambiguity degrades performance. Role conflict, which is often treated as the same thing, does not.
role ambiguity → performance r = −.21 · role conflict → performance r = −.07, described as negligibleJackson and Schuler (1985) revisited: A meta-analysis of the relationships between role ambiguity, role conflict, and job performance. Journal of Management, 26(1), 155–169.
LimitsGeneral job performance, not cycle time. The literature is twenty years old.
This row rests on the least direct evidence of the eight. Structured discussion is known to improve what a group knows (Mesmer-Magnus and DeChurch, 2009, across 72 studies and 4,795 teams), and the operational indicator is agreed with you rather than assumed.
Trust inside a team predicts team performance, and it holds even after controlling for how the team performed before.
ρ = .30 · 112 studies · 7,763 teamsTrust and team performance: A meta-analysis of main effects, moderators, and covariates. Journal of Applied Psychology, 101(8), 1134–1150.
LimitsAssociation, not cause. The largest in this set.
Business units in the top quartile of engagement differ from those in the bottom quartile on every operational measure the business already tracks.
absenteeism 78% · turnover 51% (low-turnover firms) / 21% (high) · safety incidents 63% · quality defects 32% · productivity 18% · profitability 23%Q12 Meta-Analysis, 11th edition. 736 studies · 347 organisations · 183,806 business units · 3,354,784 employees · 90 countries.
LimitsGallup owns the instrument. A quartile gap, not an intervention effect.
How a manager receives challenge predicts whether people bring the next problem, measured at a later point in time rather than the same one.
managerial openness predicts subsequent employee voice · psychological safety → voice ρ = .31 · leader relations → psychological safety ρ = .44 (30 samples, 10,180 people)Leadership behavior and employee voice: Is the door really open? Academy of Management Journal, 50(4), 869–884. With Personnel Psychology, 70(1), 113–165.
LimitsVoice measured as behaviour, not concerns logged in a system.
The only randomised trial showing that training managers changes a hard outcome.
cluster randomised trial · 88 managers · about 4,000 employees · significant reduction in work-related sick leave, p = 0.049 · reported return of £9.98 per £1Workplace mental health training for managers and its effect on sick leave in employees: A cluster randomised controlled trial. The Lancet Psychiatry.
LimitsOne trial. One emergency service. p = .049.
And the finding that runs the other way, which we publish alongside it.
participatory organisational interventions: mental health SMD = −0.04 · work performance SMD = 0.01 · 8 cluster randomised trials · null across all outcomesEffects of participatory organizational interventions on mental health and work performance. Journal of Occupational Health, 68(1).
LimitsParticipatory interventions, not leadership development. Mostly Japanese workplaces.
We measures the behavioural and organisational conditions that shape how decisions are made and work gets done.
We establish baselines, track change and bring established measures into relationship — testing what the data reveals, what it obscures and where prevailing assumptions warrant challenge.
We distinguish measured change from inferred impact. Where commercial, regulatory or people outcomes are examined, we report the strength of the evidence and do not claim causation where it cannot be established.
Homebridge Alliance complements formal audit, risk, compliance and internal audit functions. It does not replace them, and nothing here is a sign-off.
