A note on bias: I’m a coach, so I have an interest in this answer. I’ve tried to report the research as it is, limits included, and every source is linked below.
Why this question is harder than it looks
“Coaching” covers a lot: a manager’s weekly check-ins, a sports coach, a life coach, an executive coach, a coaching app. The word isn’t protected, so anyone can use it. That makes “does coaching work?” a bit like “does exercise work?”. It depends on what you mean and how you measure it.
The best evidence comes from randomised controlled trials, where people are randomly assigned to coaching or to a comparison group, so that differences afterwards can be put down to the coaching. Most of these trials are about workplace and executive coaching, with professional coaches.
What the randomised trials say
The largest RCT-only review. Erik de Haan and Viktor Nilsson analysed 37 randomised controlled trials published between 1994 and 2021, covering 2,528 people. Coaching had a moderate positive effect across leadership and personal outcomes (g = 0.59). They also found signs of publication bias, and the effects were larger when people rated themselves than when others observed them. The length of the coaching didn’t make much difference (de Haan & Nilsson, 2023).
For a sense of scale: an effect of around 0.5 to 0.6 means the average coached person ends up better off than roughly 70% of the comparison group on the measured outcome. That’s a meaningful effect for an intervention of a few months.
Effects on different outcomes. Tim Theeboom, Bianca Beersma and Annelies van Vianen combined studies of coaching in organisations. They found positive effects on all five outcome types they looked at: performance and skills, well-being, coping, work attitudes, and goal-directed self-regulation. Effects ranged from g = 0.43 for coping to g = 0.74 for goal-directed self-regulation (Theeboom, Beersma & van Vianen, 2014). The strongest effect was on goal-directed self-regulation: setting goals and steering toward them.
Executives specifically. In a randomised study with executives in a public health agency, Anthony Grant and colleagues found that coaching increased goal attainment and resilience and improved workplace well-being, and reduced depression and stress, compared with a control group (Grant, Curtayne & Burton, 2009).
Life coaching outside work. A randomised study of a 10-week life coaching programme with adults from the community found gains in well-being and hope compared with a waiting-list group, and people pursued their goals more actively (Green, Oades & Grant, 2006).
What drives the results
Carolin Graßmann, Franziska Schölmerich and Carsten Schermuly pooled 27 samples covering 3,563 coaching processes. They looked at the working alliance: the bond between coach and client, and their agreement on goals and on how to work. A better alliance went with better outcomes, with a moderate and consistent relationship (r = 0.41). It also went with fewer unwanted side effects. The link held regardless of the type of client, the coach’s experience, the number of sessions or who rated the outcome (Graßmann, Schölmerich & Schermuly, 2020).
A large study of more than 1,800 coaching relationships, led by Erik de Haan, found the same: the working alliance and the client’s belief in their own ability to change predicted results. Matching coach and client personalities did not (de Haan et al., 2016).
In plain terms: who you work with, and how safe and honest the relationship is, matters as much as any method. That matches my own small survey, where 18 of 20 people said chemistry in the first conversation would decide which coach they chose.
Where ontological coaching fits
My own training is in ontological coaching, which works with three connected domains: language (the stories we tell), emotions and moods, and the body. It’s older as a practice than as a research topic. Randomised trials have only recently started, including a pilot trial with nursing students and a multi-centre trial with practising nurses from a research group in Singapore (Pereira et al., 2024).
Each of its domains has strong support in mainstream psychology. Naming feelings reduces the brain’s alarm response (Lieberman et al., 2007). How you phrase your inner talk changes how you handle stress (Kross et al., 2014). Mood colours judgement (Forgas, 1995). Emotion lives in the body as well as the mind (Niedenthal, 2007). So I’d describe the evidence for ontological coaching as promising and early. The research page goes through it in detail.
What about ROI?
The most quoted figures come from the ICF Global Coaching Client Study (2009), commissioned by the International Coaching Federation and carried out by PricewaterhouseCoopers with the Association Resource Centre. It surveyed over 2,100 coaching clients in 64 countries. Among companies that could calculate a return, the median reported ROI was seven times the investment. Among individuals, 68% said they at least made back their investment, and those who saw a financial gain reported an average of 3.44 times what they spent.
Read these with care. The numbers are clients’ own estimates, and the coaching industry’s main professional body commissioned the study back in 2009. They tell you that many clients felt coaching paid for itself. They don’t prove it.
For an individual, I find a different question more useful: what would it cost you if nothing changed in the next 12 months? Time, health, relationships, the business you’re not building, the person you’re not becoming. Coaching is worth it if it changes that cost, and you’re the one who can judge that.
What coaching doesn’t do
Coaching is for people who are functioning and want to change something. It’s not treatment for depression, anxiety disorders, trauma or addiction. A good coach will tell you when something belongs with a doctor or therapist, and will refer you. If you’re not sure which you need, ask. A good coach would rather say “this isn’t for me” than take you on.
How to use this when you choose a coach
- Have a real first conversation. Given how much the relationship predicts results, treat the first call as the most important test. Notice whether you feel safe enough to say the true thing.
- Ask how they work. A method you can understand is a good sign. So is a coach who can tell you what you’ll leave with.
- Check accreditation. Bodies such as the ICF and the EMCC accredit coaches on training, logged practice hours, an assessment of their skills, and ongoing education.
- Agree how you’ll know it’s working. Decide early what would count as progress, and look at it together along the way.
How to choose a coach goes into this in more depth.