
Why You Should Never Coach Your Team Members
Coaching your team creates a downward spiral of insecurity and lost ownership. Give them data instead, let them name their own gaps, then train on what they missed.
Why is coaching your team members a mistake?
Coaching puts you in a position of correcting people, which makes them feel insecure and stops them from taking ownership of their own work.
Paul Veth built this insight while running a customer satisfaction pilot at a big phone company, managing a team of 26 people. His first instinct was not to fix anyone's behavior. It was to ask a question most managers skip entirely: do I actually know why my team members work here? Not the assumption. The real answer, spoken out loud by them. Once someone names their own reason for working somewhere, it gets integrated into their system automatically. That's the foundation everything else is built on. If you skip that step and jump straight to coaching, you're building corrections on top of a relationship you never actually established.
What should you do before evaluating your team's performance?
Go on a one-on-one walk with each team member, talk about your own personal life first, and ask why they chose to work for your company.
The walk is the mechanism, not small talk for its own sake. Paul Veth describes sharing a bit about his own personal life first, which naturally invites the other person to open up about theirs. From there, asking why someone chose the company stops feeling like an interrogation and starts feeling like a real conversation. A lot of people assume customer service employees are only in it for the money and never name it. According to Paul Veth, that doesn't matter. What matters is that they say it out loud, because that's the moment it becomes real to them.
How do you evaluate a team member without coaching them?
Give them data first, sit next to them rather than across from them, and ask what they think they can do better before you say anything.
Data changes the dynamic completely. When people have concrete numbers about their own work, they stop needing you to tell them what's wrong, and they start taking responsibility for it themselves. Paul Veth's method is simple: sit with them, not against them, and ask two questions. What can you do better? What is going well? If they name something you also noticed, you confirm it and connect it to the company's vision and strategy so they feel seen. If they don't name something you noticed, you write it down and stay quiet. You don't tell them. That's the part most managers get wrong.
What do you do with the gaps your team doesn't name themselves?
You don't coach those gaps directly. You build a training around them and repeat it regularly until the team names those gaps themselves.
This is the step that separates this approach from ordinary management. If you take the list of things your team members didn't mention and confront them with it directly, they feel insecure and they stop taking ownership. Instead, turn those gaps into training content, delivered daily, weekly, or monthly depending on your rhythm. People absorb it without feeling singled out. At the next evaluation, they often name the exact thing you trained on, as if it were their own discovery. Paul Veth points out that this is the whole difference: training is about getting better, coaching feels like being told you're bad at your job.
What results can this framework actually produce?
Applying this approach with a 26-person team took Net Promoter Score from plus 21 to plus 43 within six weeks, alongside a measurable rise in team trust.
Six weeks is not a long runway to move a customer satisfaction metric that much. Paul Veth measured the shift with NPS and watched it move from plus 21 to plus 43 in that window, while trust within the team also rose noticeably. The mechanism behind it wasn't more pressure or more oversight. It was people feeling seen, given real data, and trained on their actual gaps instead of being corrected for them. Other managers often ask how there's time left over for anything else. The answer is that quality time replaces constant firefighting.
Frequently Asked Questions
Why shouldn't managers coach their own team members?
Coaching puts the manager in a corrective role, which makes team members feel insecure and stops them from taking ownership of their work. Paul Veth found that once you fall into that trap, it becomes a downward spiral. Data, training, and self-discovery work better than direct correction.
How do you start building trust with a team before evaluations?
Go on a one-on-one walk with each person, share something about your own personal life first, and ask why they chose to work at the company. According to Paul Veth, once someone says their real reason out loud, it becomes integrated into how they approach their job.
What role does data play in team evaluations?
Data lets team members see their own performance and take responsibility for it, instead of relying on the manager to point things out. In an evaluation, you sit with them and ask what they can do better and what is going well, using the data as the shared reference point.
What do you do with weaknesses a team member doesn't mention themselves?
You write them down without saying anything in the moment, then build a training program around those specific points. Repeating that training regularly helps team members absorb the lesson without feeling singled out, and they often name the gap themselves at the next evaluation.
What results has this no-coaching approach produced?
Paul Veth applied this framework to a 26-person team and saw Net Promoter Score rise from plus 21 to plus 43 within six weeks. Team trust also increased, and the dynamic shifted from managers against employees to a team growing together.
Related articles
Discussion
The idea that coaching your team actually destroys ownership is pretty counterintuitive, especially when most leadership advice pushes the opposite. Have you ever noticed the moment you started coaching someone on your team, they stopped thinking for themselves? Share what happened.