The hidden cost of doing nothing.
Every organisation carries operational risk from unaddressed non-technical skills gaps. Most never price it. This tool estimates the annual cost you are already paying, using effect ranges drawn from peer-reviewed literature on safety incidents, turnover and error-driven loss.
Adjust the inputs on the left. The model runs live. The assumptions are visible at the foot of the page — nothing is hidden.
Applying the meta-analytic range for well-designed team training (Cohen's d ≈ 0.58, McEwan et al. 2017) to the cost drivers above gives a conservative estimate of what a properly implemented, supported programme could shift in a year. The figure assumes a supportive transfer environment. Without one, reductions are materially smaller.
Every figure above is derived from published ranges in the peer-reviewed literature. The model is deliberately transparent rather than a black box — every assumption can be challenged, adjusted or replaced with organisation-specific data.
Incident cost
Incident frequency by context draws on Pimenta et al. (2025) — 19–62% of ICU admissions with at least one safety incident — scaled down for lower-risk environments. The NTS-attribution share uses Uramatsu et al. (2017), where 47% of fatal medical accidents traced to NTS failures; applied as a 30–50% range to typical operational incidents. Pimenta et al. 2025; Uramatsu et al. 2017; Ibiam & Harrop 2022
Turnover and team climate
Roughly 20–25% of voluntary turnover is associated with poor manager relationships and team climate — a conservative blend of figures from meta-analytic turnover research. Replacement cost is estimated at 75–150% of annual salary for knowledge roles, depending on seniority. Blended from standard HR practice ranges and Alayande et al. 2024
Coordination loss
Team-level coordination failures cost an estimated 4–8% of productive time in operational settings, scaling with complexity. Derived from the variance-explained figures in Salas et al. (2008) and Weaver et al. (2014), who found team training accounts for 12–20% of variance in team outcomes. Salas et al. 2008; Weaver et al. 2014
Discretionary effort
Disengagement associated with low psychological safety and weak team climate is applied as a 5–10% discount on payroll, reflecting the performance premium engaged teams deliver. Lower bound aligns with conservative engagement-research estimates; upper bound with Gallup-style figures. Ito et al. 2021; Vu et al. 2025; Ansari et al. 2019
Expected reduction
The reduction figure applies the mid-range of the team-training meta-analytic effect (Cohen's d = 0.58, McEwan et al. 2017) to the aggregate cost — translated to roughly 20–25% reduction when the transfer environment is supportive. In hostile transfer environments — blame culture, hierarchy, compliance orientation — the reduction drops toward zero. McEwan et al. 2017; Hughes et al. 2016; Bunch 2007
What this model does not include
Catastrophic-tail events — the Piper Alpha or Deepwater Horizon class of outcome — are excluded because their frequency in most organisations is too low to annualise meaningfully. Reputational damage, regulatory exposure and strategic opportunity cost are also excluded. The model is therefore an undercount, not an overcount, of true exposure. Ibiam & Harrop 2022
How to use this figure
This calculator produces an estimate, not a promise. It is designed to move a conversation from "should we spend on training" to "how much risk are we already carrying." The figures are most useful as the starting point for a deeper co-designed assessment using the organisation's own incident, turnover and engagement data. Treat the output as directional — the order of magnitude is more reliable than the specific number.
Human in the System
Be Better Than Yesterday