There’s a question that keeps coming up when I talk to engineering directors and HR leads about coaching for their mid-career women: “Can you prove it works?”
Fair question. Here’s the answer.
The problem you already have
As we established in Why experienced women leave engineering – and what it costs [link], the problem is large, structural and expensive: experienced women leave the profession far faster than men, and replacing each one costs a multiple of her salary. This article is about the next question – does coaching actually keep them?
What the strongest evidence says about coaching
I want to be specific about what “strongest” means. Not industry surveys. Not self-reported data from coaching bodies. Peer-reviewed, independently conducted research.
A 2023 meta-analysis in the Academy of Management Learning & Education reviewed 37 randomised controlled trials of workplace coaching – the same methodology used to test whether a new drug works. Across 2,528 participants, coaching produced a statistically significant effect size of g = .59 on leadership and personal outcomes. That’s moderate and replicable. Not a miracle. Not nothing. A measurable, evidence-based improvement comparable to other established workplace interventions.
A study published in PNAS – the Proceedings of the National Academy of Sciences, one of the most prestigious scientific journals in the world – tested a coaching intervention specifically with early-career women in STEM. Randomised design, baseline measures, control group. The coached women showed a 12.68% improvement in capability measures compared to what their scores would have been without the intervention. And the company’s own managers independently rated the coached women’s performance higher. That’s not self-reported. That’s observed.
A separate meta-analysis in Frontiers in Psychology (2023) found coaching has a larger effect on behavioural outcomes – how people act and decide – than on stable personality traits. It doesn’t change who people are. It changes what they do. For a retention problem driven by stalled progression and lack of recognition, that’s exactly what you need.
Where the manager fits
Gallup’s 2025 State of the Global Workplace report puts it plainly: managers account for 70% of the variance in team engagement. Female managers’ own engagement fell 7 points last year – the sharpest fall of any group.
Managers who receive coaching-skills training show half the disengagement of those who don’t, and a 28% performance uplift. The people who determine whether your best women stay and progress are themselves unsupported. Coaching all sides – the women, the women managers and their managers – addresses the full equation.
The arithmetic
I’m not going to publish programme pricing in an article. But I will give you this: a coaching intervention for a mid-career woman typically costs less than 10% of what it costs to replace her when she leaves. If the programme prevents even one departure, the return covers the investment several times over.
A detailed business case – with worked examples and ROI calculations for different scenarios – is available as a separate resource. If you’d like a copy, get in touch.
One Page Flyer: The Business Case for Coaching Mid-Career Women in STEM
The question worth asking
You already invest in recruiting women into STEM. The evidence says coaching is one of the most effective ways to keep them – and to make sure the ones who stay can actually progress.
The cost of doing nothing is measurable. So is the return on doing something about it.
Sally Purdie CEng FIMMM – Chartered Engineer | STEM Retention Coach
Sources:
Gallup (2025). State of the Global Workplace
Academy of Management Learning & Education (2023). Meta-analysis of 37 coaching RCTs
Proceedings of the National Academy of Sciences (2022). Preventing soft skill decay among early-career women in STEM
Nicolau et al. (2023). The effects of executive coaching on behaviors, attitudes, and personal characteristics. Frontiers in Psychology, 14:1089797
