A client wrote this on her end-of-engagement form last March: "I think you pushed me harder than I wanted to be pushed, and I'm grateful for it." I read it standing at my kitchen counter at 6 AM, green tea going cold next to me, and I just sat with it for a while. That sentence taught me more about my coaching than anything I'd learned in a training. It told me something specific about how I show up, what I'm willing to risk in a session, and what that particular client needed from me even when she didn't know she needed it.
Most feedback forms never produce a sentence like that. They produce "very helpful" and "would recommend to others" and "I really enjoyed our work together." Which is nice. And tells you absolutely nothing.
Here's what I've figured out about building a feedback system that actually makes you better at the work, and how I use AI to find patterns I'd never see on my own.
The problem with asking if people liked it
Most coaching feedback forms are satisfaction surveys in disguise. They ask things like "How would you rate your overall coaching experience?" and "Would you recommend this coach to a colleague?" These questions produce polite answers because they invite polite answers. They're structured for reassurance, not information.
Think about it from the client's side. They've spent four or six months with you. There's a relationship. Maybe they genuinely like you. Maybe they feel grateful. Maybe they feel a little guilty about the sessions they cancelled. They're not going to write "honestly, I don't think the middle two months were very useful" on a form that reads like a hotel checkout survey.
The deeper problem is this: most of us are a little scared of real feedback. We should be. Good coaching means you sometimes sit in silence when someone wants you to rescue them. You ask the question that makes them go quiet in a way that isn't comfortable. You hold a mirror up and don't look away first. You never fully know how that landed until much later, if ever. A feedback form that only asks about satisfaction protects you from finding out.
I sent satisfaction-style forms for my first three years. I got uniformly positive responses. I filed them. I learned nothing. In hindsight, I was designing forms that told me what I wanted to hear.
Ask about experience and shift, not satisfaction
The design principle I eventually landed on: ask questions that invite the client to describe what happened to them, not how they'd rate what happened to them. There's a significant difference between "How satisfied were you?" and "What shifted for you, if anything?" The first is a judgment. The second is a reflection. Reflections produce sentences you can actually learn from.
I also split my feedback into two moments. A mid-engagement check-in (around session four or five of a typical engagement) and a full end-of-engagement form. The mid-engagement check-in exists because of something that happened about five years ago. I was coaching a director through a leadership transition, and the engagement ended earlier than I expected. He was gracious about it, said the timing wasn't right, thanked me. But when I replayed the previous few sessions honestly, I could see he'd been trying to signal something for weeks. Something about our work wasn't landing for him. He'd gone a little flat in sessions. His responses had gotten shorter. I hadn't picked up on it because I was focused on the process, on our stated goals, on the arc I had in my head for where we were going.
If I'd asked him four questions at the halfway mark, I would have caught it. I'm almost sure of it. So now I always ask.
The mid-engagement check-in
This one is short. Four questions. I send it as a simple email, not a formal form, because formality invites performance. I tell them it'll take five minutes and that I genuinely want to know.
---
Mid-Engagement Check-In (sent around session 4–5)
1. What's been most useful about our work so far? (Not "what did you like" but what's actually been useful to you in your day-to-day.)
2. Is there something you came in wanting to work on that we haven't gotten to yet? Or something that's emerged that feels more important than what we originally agreed on?
3. Was there a moment in our sessions where you wanted me to back off but I didn't?
4. Is there anything you'd want me to do differently in how I show up for our remaining sessions?
---
That third question is the one that matters most, and I want to explain why the wording is specific.
The question that actually gets honest answers
"What could I have done better?" doesn't work. It's too broad. It puts the client in the position of evaluating your performance, which most people are reluctant to do to your face (or even on a form they know you'll read). It invites a vague "nothing, it was great."
"Was there a moment where you wanted me to back off but I didn't?" works because it's specific, it's experiential, and it implicitly gives permission. It says: I know I push. I know that sometimes lands wrong. I want to know when. The question assumes the moment might exist, which makes it safe to describe.
I've gotten answers to this question that fundamentally changed how I coach. One client told me I had a habit of circling back to a point she'd already acknowledged, and it made her feel like I didn't trust her to hold it. She was right. I was doing that. Another told me there was a session where he started to get emotional and I moved to a question too quickly, and he wished I'd just let the silence sit. That one stung. And it was exactly what I needed to hear.
You won't always get something. Sometimes the answer is "no, I can't think of one." That's fine. But when you do get something, it's gold.
The end-of-engagement form
This one is fuller. I send it about a week after our last session, so there's a little distance.
---
End-of-Engagement Feedback Form
1. When you think back on our work together, what stands out most? (A moment, a conversation, a shift, anything.)
2. What changed for you over the course of our engagement, if anything? This can be something concrete (a behavior, a decision) or something harder to name.
3. Was there a moment where you wanted me to back off but I didn't? (Yes, I ask this one again.)
4. Was there something you wanted to bring to our sessions but didn't? What got in the way?
5. How did our work affect how you show up with your team, if at all?
6. Was there a point where our work felt stuck or less useful? What was happening for you at that time?
7. What would you want a future client to know about working with me?
8. Anything else you want to say that doesn't fit neatly into a question?
---
Question six is the second-most important one. "Was there a point where our work felt stuck?" gives language to something most clients feel at some stage but rarely name. Every engagement has a flat period. Knowing when it happens and what the client was experiencing during it tells you something real about your coaching posture.
Question seven is interesting because it reframes the client as an advisor rather than an evaluator. People give different, often more honest, information when they're advising someone else.
What I do with all of this (and where Margaret comes in)
For years I read each form, had a reaction, and filed it. I'd remember the vivid responses and forget the rest. Which meant I was learning from the most dramatic feedback, not from the patterns.
Now, every quarter, I pull all the feedback I've received (mid-engagement and end-of-engagement) into a single document. I anonymize it. Names out, identifying details out. Then I run it through Margaret with a prompt I've refined over about six iterations.
Here's the full prompt:
---
Quarterly Feedback Pattern Analysis Prompt
I'm an executive coach reviewing anonymized client feedback from the past quarter. Below are responses from [number] clients, including both mid-engagement check-ins and end-of-engagement forms. I'd like you to analyze this feedback and identify:
1. Recurring themes in what clients find most valuable about our work. What keeps showing up?
2. Recurring friction points or moments of discomfort. Where does my coaching tend to create tension, and is there a pattern in how clients describe it?
3. Any patterns in where engagements feel "stuck" or less useful. Is there a common timing or topic?
4. What kinds of clients seem to get the most out of working with me, based on how they describe their experience and outcomes?
5. Anything that surprised you, or that seems contradictory across different client responses.
Please be direct. I'm not looking for reassurance. I want to see what the data suggests about how I coach, including the parts that aren't working.
[Paste anonymized feedback below]
---
The first time I ran this, I was on my second cup of tea and not expecting much. What came back was uncomfortably specific. Margaret identified that three different clients had described a version of the same experience: feeling challenged by me in the first few sessions in a way that was productive, then hitting a middle period where they weren't sure what we were working toward, then a late-stage acceleration where things clicked. That mid-engagement dip was a pattern, not an anomaly. It told me I might be strong in the opening and closing of an engagement but less structured in the middle. That's not something I would have seen from reading individual forms.
The analysis also showed that my founder clients describe our work very differently than my mid-level leaders do. The founders talk about honesty and being challenged. The leaders talk about clarity and permission. Same coach, different experience. That changed how I frame the first session depending on who's sitting across from me.
Closing the loop
The feedback doesn't just sit in a quarterly review. It feeds back into how I set up new engagements. When I noticed the mid-engagement dip pattern, I started naming it explicitly in my working agreements: "Around session four or five, we may hit a point where the initial momentum fades and it's not clear what's next. That's normal. We'll check in about it." Just naming it reduced the friction.
When multiple clients mentioned wanting to bring something to sessions but not knowing if it was "in scope," I changed my intake questions to include: "Is there anything you'd want to talk about that might feel outside the boundaries of executive coaching? Leadership and life aren't as separate as the scoping document pretends."
The system feeds itself. Feedback shapes the form. The form shapes the feedback. Over time, the questions get sharper because you know more about what you actually need to hear.
I was on my walk yesterday, the usual loop through the park, thinking about why this matters to me as much as it does. It's not about optimization. It's not about getting better Net Promoter Scores from my coaching clients, which is a phrase that should never exist. It's that after nine years, I'm still genuinely unsure whether I'm better than I was at year five. Not because I haven't grown. Because I don't trust my own perception of my growth. The feedback form is the closest thing I have to an honest mirror.
The best coaches I know are the ones who are still curious about their own work after a decade. Still a little unsettled by it. Still willing to ask the question that might come back with an answer they don't love. The form is one small piece of that curiosity. Margaret helps me see the patterns. But the willingness to ask, and to sit with what comes back, that part's mine.