Life Coach Questionnaire for Clients: Build and Analyse It With AI

A good life coach questionnaire for clients has about twenty questions across three parts: their situation, a scored wheel of life areas, and their goals and constraints. With written consent, AI can summarise the answers into one page of themes, gaps and questions for the first session. You still read the originals, and the client corrects the summary.

Searching for this topic mostly returns question lists to copy. Lists are the easy part. The harder parts are scoring the wheel so it means something, getting consent for an AI step, keeping answers private, and checking that the summary has not smoothed out the one sentence that mattered. Get any of them wrong and a tidy one-page summary can mislead you before you have even met the client.

Follow me on Instagram@sagnikteaches

Before you write a question: decide what each one is for

Every question should earn its place by changing something you do. A useful test: write next to each question what you will do with the answer. If the honest note is "nothing, I just like knowing", cut it or move it to the first session, where a conversation does that job better.

Connect on LinkedInSagnik Bhattacharya

The questionnaire below supports four decisions: whether you are the right coach for this person; what to focus the first session on; what practical limits (time, money, energy) the programme has to fit; and whether anything needs a referral to a different kind of professional first.

Subscribe on YouTube@sagnikteaches

The twenty-question template

Send it a few days before the first session through a form tool, with a note that it takes fifteen to twenty minutes and that skipping any question is fine. Use plain, short questions with free-text boxes for the open ones.

Part A: where you are now (questions 1 to 6)

1. What made you decide to look for a coach now rather than six months ago?
2. In a sentence or two, what is going on in your life at the moment?
3. What do you do for work, and how do you feel about it on an ordinary Tuesday?
4. Who is part of your daily life (partner, family, friends, colleagues) and who supports you?
5. Have you worked with a coach, mentor or counsellor before? What helped and what did not?
6. Are you currently getting professional support for your health or wellbeing that I should know about? (Optional)

Part B: the wheel (questions 7 to 14)

Ask the client to rate each area from 1 (very unsatisfied) to 10 (very satisfied), and add one sentence on what a higher score would look like.

7. Career or work
8. Money
9. Health and energy
10. Relationships (partner, family, friends)
11. Personal growth and learning
12. Fun and recreation
13. Home and physical environment
14. Sense of purpose or contribution

Part C: goals and practicalities (questions 15 to 20)

15. If our work together went really well, what would be different in six months?
16. Which one area from the wheel would you most like to change first, and why?
17. What have you already tried, and what got in the way?
18. How much time a week can you realistically give to this? Be honest.
19. What would make you stop halfway through a programme?
20. Is there anything you want me to know, or avoid, that I have not asked about?

Eight areas is a common wheel size, but you can change the labels to suit your niche. A career coach might split work into satisfaction, progress and skills; a wellbeing coach might add rest. Keep the count between six and ten so the picture stays readable.

Scoring the wheel so the numbers say something

A wheel of eight scores is a snapshot, not a diagnosis. Four simple calculations turn it into something you can use in the session.

1. Average score: add the eight and divide by eight. It gives a feel for overall satisfaction but hides the pattern.
2. Lowest two areas: candidates for focus, but not automatically. Ask the client which low score hurts most.
3. Spread: highest minus lowest. A spread of 6 or more means one area is dragging; a spread of 2 or less often means a general flatness, which is a different conversation.
4. Gap between score and priority: compare the lowest area with their answer to question 16. If they differ, the mismatch is the first thing to explore.

A caution on the numbers: people rate on different scales. One client's 6 is another's 9. Use scores to start the conversation and for the client's own re-scores at week twelve, not to compare clients with each other.

Rewriting weak questions before you send them

Many downloadable questionnaires are full of questions that produce short, polite, useless answers. Rewrite them once and the AI summary improves because the input is richer. A few before and after pairs:

The last row matters for scope. Coaches who ask for life histories drift into territory that belongs to therapists, and they collect data they then have to protect. Keep the form to the present and the future, and let the client choose how much past to bring.

Consent before the AI step

Intake answers can be personal: relationships, money, health and worries. If you plan to paste them into an AI tool, tell the client first and get a yes you can show later. Put this above the form:

How I use your answers
I read every answer myself before we meet. If you agree, I also use an AI tool to
produce a one-page summary of themes, which I check against your original words and
share with you at the start of our first session so you can correct it.

- Tool: [name of tool and plan], set so that your answers are not used to train models
- What is entered: your answers with your name removed (I use a code)
- Who sees the summary: me, and you
- How long I keep answers: [period], then they are deleted
- You can say no. If you tick "no AI", I will read your answers myself and nothing will be
  entered into any tool.

[ ] Yes, you may use an AI tool on my answers    [ ] No, please do not

Two settings matter here. ChatGPT's and Claude's business plans do not train on your content by default; on their personal plans, training is governed by a switch in privacy settings, and ChatGPT's switch starts on for personal accounts, so check it before you promise anything to a client. And if the client leaves the box blank, treat that as a no.

The summary prompt

Remove the client's name, employer and any other identifying detail from the pasted text, and replace it with a code. Then use a prompt that restricts the model to what the client wrote.

You are helping a life coach prepare for a first session. You are not a therapist and
you do not diagnose or give advice. Use only the client's answers below.

Produce, in under 350 words:
1. Situation in three sentences, using the client's own words where possible.
2. Wheel summary: list scores lowest to highest, the average, and the spread.
3. Three themes that appear in more than one answer, each with the answer numbers it
   comes from.
4. Mismatches or tensions between answers (for example a low score they did not choose
   to focus on).
5. Practical limits: time, money, support, as stated.
6. Five open questions I could ask in the session, phrased neutrally.
7. Flags: quote any sentence that mentions self-harm, abuse, crisis, medication or a
   professional diagnosis, exactly as written, without interpreting it. If none, write
   "No flagged phrases".
If something is not answered, write "not answered". Do not guess.

CLIENT CODE: C-014
ANSWERS: ...

The numbers in point 2 are where a model is most likely to slip, so calculate the average and spread yourself in a spreadsheet and compare. Models can also invent a "theme" from a single remark. Insisting on answer numbers for each theme makes false ones easy to spot.

One sample client, start to finish

Say client C-014 is a project manager in her late thirties, an illustrative composite and not a real person. She completes the form and ticks yes to AI. Her wheel scores are: career 4, money 6, health and energy 3, relationships 7, personal growth 5, fun 3, home 7, purpose 4. The average is 39 divided by 8, which is 4.9. The spread is 4 (7 minus 3). Her lowest two areas are health and energy (3) and fun (3).

In question 16 she says she most wants to change "career, I hate Sundays". Here is the first part of an illustrative summary the tool returns after the prompt above:

Situation: C-014 describes a demanding job with long hours and a sense of drifting
("I keep saying next year"). She has a supportive partner and a close friend, and has
tried a mentor before, which she found "too vague".

Wheel: lowest to highest: health and energy 3, fun 3, career 4, purpose 4,
personal growth 5, money 6, relationships 7, home 7. Average 4.9, spread 4.

Themes: (1) exhaustion and no time for herself (answers 2, 9, 12, 18); (2) wanting
concrete steps not talk (answers 5, 17); (3) career dissatisfaction linked to Sundays
(answers 3, 16).

Tension: she picks career as the focus, but her two lowest scores are health and fun.

Practical limits: three hours a week, evenings only; modest budget for tools.

Flags: No flagged phrases.

What you would fix: answer 5 said the mentor was "too vague", so a summary line about "wanting structure" is supported, but "wanting a step-by-step programme" would not be. Check each theme against its answer numbers. Also, the tension is the best part of the output: you now walk into the session with one honest question ("You said career, and your lowest scores are energy and fun; which of those would change your Sundays?") that you might otherwise have found in minute thirty.

Checking the summary with the client

Open the first session by showing her the summary and asking: "What is wrong, missing or surprising?" Her corrections are the real intake. Say she replies that "exhaustion" is overstated and that the real problem is that nobody at work notices what she does. A model that read only her scores could never have found that, which is the argument for treating the summary as a draft to be corrected, not an assessment.

A second client: the flat wheel

Not every wheel has a dip. Say client C-021, an illustrative composite, scores 6, 6, 5, 6, 6, 5, 6, 6. The average is 5.75 and the spread is only 1. A tool asked to "find the lowest areas" will dutifully name two areas that differ from the rest by a single point, which means nothing. A flat wheel usually points to a different conversation: the client may be coasting, or may be underrating everything to be modest. The summary should say so plainly, and the useful content moves to the free text. Suppose his answer to question 15 is "I want to feel like I chose my life, not inherited it". That sentence, not the grid, is the session. Add a line to your prompt for this case: "If the spread is 2 or less, say that the wheel does not show a clear priority and give more weight to the free-text answers." Without it, the model produces confident nonsense about a priority that does not exist.

Adapting the template to your niche

The twenty questions are a base. Keep the structure and swap the wording.

• Career coach: replace the wheel's eight areas with work-specific ones, such as role satisfaction, progress, pay, skills, manager relationship, hours, values fit and energy. Add a question about the last time they felt proud at work.
• Relationship or family coach: keep the wheel but add a question on what the other person would say about the situation, and a safety screen at the start with a referral route.
• Wellbeing or habits coach: add questions on sleep, movement and routines, and state at the top that you do not give medical advice. Anything on health should go through the same careful storage as the rest; see personal trainer intake forms and health data for the habits worth borrowing.
• Executive or business coach: replace the life wheel with a leadership wheel and ask for a stakeholder list. The AI summary can then group themes by stakeholder, with the same consent rules.

A referral script for answers that point beyond coaching

Sometimes the questionnaire shows something you are not the right person for: a client describing hopelessness, a crisis, or a health problem they are managing alone. Decide your response before the form goes out, and write it down. A short version: "Thank you for being so open. What you describe sounds like something a doctor or a counsellor is better placed to support, and I would not want to get in the way of that. Here are some places to start. If you would like to work with me alongside that, we can talk about timing." Keep a list of referral options that suit the clients you serve. If your answer has a flagged phrase in it, a person calls the client the same day, and AI plays no part in that decision beyond quoting the phrase you asked it to quote.

Tracking change with the second wheel

Back to C-014. After twelve weeks she re-scores: career 6, money 6, health and energy 5, relationships 7, personal growth 7, fun 5, home 7, purpose 6. The new total is 49, so the average is 6.1, up from 4.9. Look at the movement rather than the average: health and energy up 2, fun up 2, career up 2, purpose up 2, growth up 2, and money, relationships and home unchanged. You can show the client a simple before and after table and ask which changes she attributes to the work and which to something else. Ask the AI only to line up the two sets of numbers and her comments side by side; the interpretation belongs to the two of you. Keep an eye on a trap: a rise in the average on its own can mean she has become more generous in scoring rather than that anything has improved, which is why her comments sit next to the numbers.

Where intake summaries drift from what the client wrote

• The summary sounds more certain than the client: the model turns "sometimes I wonder" into "she wants to leave her job". Compare verbs.
• Averages that do not add up: recompute every figure.
• A flag missed or invented: read the free-text boxes yourself for anything sensitive, every time.
• Over-coaching in the prompt: if you ask the model for advice or an action plan, you start the session with its opinion in your head. Keep the prompt to summary and questions.
• Identifying details left in: employer names, a child's school, a rare job title. Redact before pasting.

One more habit: keep the corrected summary, not the raw model output, as the record. The corrected version carries the client's own voice, and it is the one you will want to reread before session six.

Storing answers and deleting them

Treat the completed form as sensitive. Keep it in one place with access limited to you, protected by two-step sign-in, and never in an email thread. Do not leave answers sitting in a chat history: use a temporary or no-history mode where the tool offers one, and delete the conversation after copying the summary into your notes. ChatGPT's Temporary Chat stays out of your history and out of training, though OpenAI says it may keep those chats for up to 30 days for safety; Claude's incognito chats are not used to improve its models. Neither mode replaces taking the name out first. Write your retention rule down (for example, "twelve months after the last session") and put it in the consent text. If you serve clients in the EU, the GDPR gives them rights to see and delete their data, so keep a simple way to find and remove one client's file.

A quick sum on the time saved

Say you take on six new clients a month. Reading twenty answers closely and drafting your own prep notes takes you about 35 minutes each. With the AI summary, you still read the originals, but your prep falls to roughly 15 minutes: five for the check, ten for choosing questions. That is about 20 minutes saved per client, 2 hours a month. Not dramatic, and honest: the gain is a more consistent starting point and a faster response to the client, not a replacement for your attention. If you only see two new clients a month, skip the AI step and just use the template.

Using the answers after session one

The questionnaire becomes useful again at the end of the programme. Ask the client to re-score the wheel, compare the two sets side by side, and note which areas moved and which did not. Keep the original summary, the corrections and the final scores together in the client's file. For turning goals into exercises, see how coaches use AI to build workbooks and programmes, for the weekly rhythm after session one, AI session notes and follow-ups for coaches, and for the step before this one, qualifying discovery-call bookings. If the human side worries you, how business coaches use AI without losing the personal touch is worth a read, and if you want your own questionnaire reviewed and automated end to end, that is what my AI implementation consultation covers.

Questions coaches ask about AI and client questionnaires

Do I need client consent before using AI on their answers?

Yes, and ask in plain words before they fill in the form: which tool, what it does with the text, who sees the summary, and how long answers are kept. Make the AI step optional, with a manual route for clients who say no.

How long should a coaching intake questionnaire be?

Twenty questions that take about fifteen to twenty minutes is a workable limit. Longer forms tend to be half-finished, and the extra detail is better gathered in the first session. Every question should change something you will do or say.

Can AI spot mental-health problems from the answers?

No, and you should not ask it to. Coaching is not therapy. Add a screening question about current support and a clear referral script, then let a person decide. A summary tool may highlight words you gave it as flags, not diagnose anything.

Where should I store completed questionnaires?

In a private, access-controlled place you already trust, such as your coaching platform or an encrypted drive, with a set retention period. Avoid leaving answers in a chat history. Use a client code in prompts and keep the code-to-name list elsewhere.

Should the client see the AI summary?

Offering it is good practice and builds trust. Share the summary at the start of the session, ask what is wrong or missing, and treat their corrections as the final word. It also catches errors before they shape your questions.

Further reads

• Should Coaches Build an AI Version of Their Method for Clients? — Whether to turn your method into an AI tool.
• Are AI Tools Built for Coaches Worth Paying For Over ChatGPT? — Which paid coaching tools beat a general assistant.
• Wedding Planner Client Questionnaire: Answers to a Brief With AI — Another questionnaire-to-brief workflow.
• Automate Client Onboarding With AI: Forms, Contracts, and Emails — Send the questionnaire and consent automatically.
• How to Turn Sales Call Recordings Into Coaching Notes With AI — Coaching notes from recorded calls.
• Personal Chef Client Intake Form: Allergies, Diets and Menus With AI — A short intake form for a client-facing service.
• Life Coaching Rates During Automation: How to Reprice Packages — How to reprice life coaching packages once AI handles prep and admin: hours-based versus milestone-based structures, with a worked example.
• AI 360 Feedback for Coaches: Collect, Theme and Report — A coach's 360 process from question set to debrief: anonymity thresholds, AI theme coding with a human check, a report clients trust, and what tools cost.
• AI for Life Coaches: Between-Session Support Within the ICF Code — Reminders, reflection prompts and homework recaps with AI, tested against the ICF Code of Ethics, plus what never goes into a chatbot.
• Best AI Model for Life Coaching: What Session Notes and Privacy Need — How a life coach should choose between ChatGPT, Claude and Gemini for session prep and follow-ups, with privacy settings, ICF Code limits and a test routine.
• AI Tools and AI Development: The Complete 2026 Guide — the AI hub, including every tutorial in the AI-for-business series.

Sources: OpenAI's business-data page and data controls FAQ, Claude's pricing page and Anthropic's model-training help article (checked 4 November 2026), and widely used coaching practice (the wheel-of-life exercise and goal-setting questions), described generically.