Mid-Program Check-In Survey for Coaching Clients

By the time a coaching program ends, it’s too late to fix the thing that went wrong in week five. The client who quietly lost momentum, never asked their real question, or started doubting the process has already decided how they feel, and you only find out in the exit survey or, worse, when they don’t renew.

A coaching mid-program survey flips that. You ask at the halfway point, while there’s still runway to course-correct. This prompt builds a short, honest check-in that surfaces the stuck clients before they go quiet, tells you exactly what’s working, and gives you the two early-warning signals to watch. And by the end of this page you’ll understand why it works, so the next survey you build is even sharper.

When to use this

• Your program is long enough to have a clear midpoint (6 weeks, 12 weeks, a 6-month container).
• You’ve noticed clients sometimes fade in the back half and you want to catch it early.
• You’re about to design or rebuild your client-experience touchpoints.
• You want feedback that’s specific and honest, not the polite “it’s going great!” that tells you nothing.

The prompt

Copy this whole block into ChatGPT, Claude, or Gemini:

You are an expert client-experience designer for coaches. Your job is to write a short mid-program check-in survey that surfaces problems early enough to fix them, so clients stay engaged and finish the program.

Before writing, ask me up to 3 clarifying questions if anything below is unclear. Otherwise, proceed.

CONTEXT
- My program: {{PROGRAM_NAME}}
- Total length: {{PROGRAM_LENGTH}}
- My clients: {{CLIENT_TYPE}}
- The main outcome they're paying for: {{MAIN_OUTCOME}}
- How I deliver it: {{DELIVERY_FORMAT}}
- My tone: {{TONE}}

TASK
Design a mid-program check-in survey that I can send at the halfway point. It must:
1. Open with 1-2 warm sentences explaining why I'm asking and that honest answers help me serve them better.
2. Include 8-10 questions, mostly quick to answer, that reveal: progress toward {{MAIN_OUTCOME}}, confidence/momentum, what's working, what's getting in the way, and whether anything is at risk of making them drop off.
3. Use a mix of formats: 2-3 rating-scale questions (label the scale, e.g. 1-5), the rest short open-ended.
4. Include one direct question that uncovers hidden frustration before it becomes a quiet cancellation.
5. End with one forward-looking question about what they want from the second half.
6. Match my tone. No corporate jargon, no 'we value your feedback'.

CONSTRAINTS
- Keep it under 10 questions so it takes 3-4 minutes to complete.
- Phrase every question so it's easy to answer honestly without feeling judged.
- Do not invent client names, stats, or results.

After the survey, give me: (a) a 2-3 sentence email or message I can use to send it, and (b) a short note on which 2 answers I should watch most closely as early warning signs of a client dropping off.

How to customize it

Replace the six {{VARIABLES}} before you send it:

See it in action (full example)

Here is the exact prompt, filled in for a public-speaking coach. This is the whole input, nothing hidden:

You are an expert client-experience designer for coaches. Your job is to write a short mid-program check-in survey that surfaces problems early enough to fix them, so clients stay engaged and finish the program.

Before writing, ask me up to 3 clarifying questions if anything below is unclear. Otherwise, proceed.

CONTEXT
- My program: The 12-Week Confident Speaker Program
- Total length: 12 weeks
- My clients: mid-career professionals who fear public speaking
- The main outcome they're paying for: deliver a high-stakes talk without panic
- How I deliver it: weekly 1-on-1 calls plus a private practice community
- My tone: warm, encouraging, no-jargon

TASK
Design a mid-program check-in survey... [rest of the prompt unchanged]

And here is the complete output the model returns:

We’re at the halfway point, and I want to make the next six weeks count.

A few honest minutes here helps me adjust your coaching to what you actually need. There are no wrong answers, and “this part isn’t landing for me” is exactly the kind of thing I want to hear.

1. On a scale of 1-5, how confident do you feel today about delivering a high-stakes talk without panic? (1 = no change yet, 5 = noticeably more confident)

2. On a scale of 1-5, how much momentum do you feel right now in the program? (1 = stalled, 5 = on a roll)

3. What’s one thing in the first six weeks that’s genuinely helped you?

4. What’s the biggest thing still getting in the way, inside the program or in your week?

5. Is there anything we’ve covered that felt unclear, rushed, or like it wasn’t for you? (Be blunt, it helps.)

6. How are the weekly calls and the practice community working for you, separately? Is one more useful than the other?

7. On a scale of 1-5, how likely are you to finish all 12 weeks at your current energy level? (1 = honestly wobbling, 5 = fully in)

8. If anything has made you think about easing off or stepping away, what is it? (Even if it’s small, I’d rather know now.)

9. What would make the second half a clear win for you, the thing that would make you say “that was worth it”?

Thank you. I read every answer myself before our next call.

---

(a) Message to send it with:

“Hey [Name], we’ve hit the halfway mark of the Confident Speaker Program. Before our next call, would you take 3-4 minutes for a quick check-in? Your honest answers let me tune the next six weeks to exactly what you need. Here’s the link: [survey link].”

(b) The 2 answers to watch as drop-off warning signs:

• Question 7 (likelihood of finishing). Any score of 3 or below is a flashing light. Reach out personally within 24 hours, don’t wait for the next scheduled call.
• Question 8 (thoughts of easing off). Any non-empty answer here, however gentle the wording, means a quiet cancellation is forming. Address the specific reason they name directly and quickly.

That’s ready to drop into your survey tool. A 30-second tweak to sound like you and it’s live.

Why this works

Three LLM principles are doing the heavy lifting here. Learn them and every prompt you write gets better:

1. Role priming. The opening line (“You are an expert client-experience designer for coaches”) tells the model which slice of its knowledge to use. “Write me some survey questions” pulls the bland, generic average; “act as a client-experience designer” pulls the questions that are actually engineered to surface retention risk. Always assign a role.
2. Specificity in, specificity out. The model can only be as concrete as your context. Because you named the real outcome (“deliver a high-stakes talk without panic”) and the real delivery format (calls plus a community), the questions ask about those things by name instead of vague “how’s it going?” filler. Generic input produces a generic survey; specific input produces questions your client recognizes as theirs.
3. Constraints are quality control. Each rule removes a common failure mode. “Under 10 questions, 3-4 minutes” prevents the bloated survey nobody finishes. “Phrase so it’s easy to answer honestly without feeling judged” stops the leading, polite questions that only ever get “great!” back. And the closing ask for the two answers to watch turns raw feedback into an action plan. The “ask up to 3 clarifying questions first” line lets the model fill gaps by asking instead of guessing, which is the single biggest fix for generic AI output.

Do this now

1. Copy the prompt above into ChatGPT or Claude.
2. Replace the six variables with your real program details.
3. Send it. If it asks clarifying questions, answer them honestly.
4. Paste the questions into your survey tool, tweak the wording to sound like you, and send it to every client at their midpoint this week.

Pro tips

• Send it at the true midpoint, not the end. The whole value is having time left to act on what you learn. A survey at week 11 of 12 is just an exit interview.
• Keep question 8 in, even though it stings. The question that asks what’s making someone consider stepping away is the one that saves renewals. Polite surveys never reveal it.
• Act on the warning signs within 24 hours. A personal message the same day a client scores low on “likelihood to finish” is the difference between a save and a refund.
• Reuse answers as testimonials. The week-6 “what’s helped you” answers are often more specific and believable than anything you’ll get at the end. Ask permission and keep them.

Related

• Survey-Response Analyzer & Action-Plan Generator for Coaches
• Exit Survey & Offboarding Feedback Email for Coaches
• Post-Session Feedback Survey Builder for Coaches
• Re-Engagement Message Series for Clients Who Went Quiet