Solution-Focused Coaching Questions: Ten Useful Purposes

Solution-focused coaching questions help a client describe a useful difference, examine relevant experience and choose what to do next. Their value depends on the conversation around them. An elegant question that ignores the answer, a real constraint or the client's refusal can make the work less useful.

This guide offers ten original groupings of questions, with their purpose, follow-up and limits. They are practical coaching examples, not an official ten-part taxonomy or a therapy protocol. A complete fictional case follows a client who rejects a hypothetical framing, obtains limited help rather than presumed weekly support, completes a writing task and corrects the resulting record.

Understand the approach and its evidence limits

Solution-focused brief therapy developed through the work of Steve de Shazer, Insoo Kim Berg and colleagues at the Brief Family Therapy Center in Milwaukee. The SFBTA's current manual index identifies its third edition from 2025. The manual describes clinical conversations involving preferred futures, exceptions, scales and relationship perspectives. It emphasizes connected dialogue rather than isolated questions and recognizes the client's problem situation.

Coaching and psychotherapy have different purposes and competence requirements. These examples concern ordinary choices and practical action. A list of questions is not training to assess or treat a condition. The 2025 ICF Core Competencies support distinguishing professions, listening and adapting to the client's needs. They do not validate this question bank or endorse a software product.

A 2021 randomized analogue study by Neipp and colleagues compared four computerized question conditions in 246 psychology undergraduates. Scaling produced more listed action steps than miracle or exception questions, without significantly outperforming the problem-focused condition on that outcome. Miracle questions reduced negative affect more than problem questions; the exception comparison was borderline, reported at p = .050. Solution-focused conditions did not show superiority for positive affect, self-efficacy or goal approach.

These were immediate responses in a mostly female, non-clinical convenience sample, not long-term results from actual coaching conversations. The original study's methods, results and limitations do not establish that asking more questions causes better life or work outcomes. Keep method description separate from claims about effectiveness.

Use ten question purposes flexibly

1. Preferred-future questions: define the useful difference

An original prompt is: “At the end of the week, what would let you say this was a useful change?” Follow the answer into something the client could recognize. “Less chaotic” might mean one agreed responsibility, a clearer request or time to finish a particular task.

Ask which part matters most now and which parts depend on other people. A preferred future can contain disagreement or uncertainty. It does not need to describe a perfect household, organization or emotional state.

If the client first needs to explain an event or its effects, listen. Do not force future language to prevent discussion of something important. The active listening guide helps connect the next question to what the client has actually said.

2. Miracle-style questions: try an optional imagined scene

Try a modest invitation: “Would imagining one easier version of Thursday help us describe what you want?” If the client agrees, ask what they would see, do or hear in that scene. It may reveal a concrete preference behind an abstract goal.

Keep the framing optional. A client may dislike imaginary change, supernatural language or the suggestion that constraints disappear overnight. A direct question about what needs to be different can serve the same immediate coaching purpose.

Translate any imagined cooperation into a real dependency. A scene where a partner offers help is not evidence that the partner has agreed. Retain the difference between an illustration and an arrangement.

3. Exception questions: investigate a relevant occasion

Ask: “Was there a recent occasion when even one part of this was easier?” If the client identifies one, explore the circumstances, their own contribution and what others did. A useful exception is specific enough to examine.

Ask whether those conditions exist now. An approach may have worked with more time, a different workload or another person's help. The repeatable element could be a preparation step, while the support still needs obtaining.

Do not insist that a success must exist. “I cannot remember one” is a valid answer. Move to preferences, support or necessary information rather than treating the client as resistant.

4. Scaling questions: unpack a personal judgment

Ask whether a scale would help, then define its purpose and endpoints together. For readiness to make a request, one endpoint could mean no workable request yet and the other a clear request that allows a genuine refusal. This is different from rating how much the whole problem is solved.

An original follow-up is: “What is already present in the number you chose, and what is missing?” That may reveal wording, resources, permission or uncertainty. Ask for the meaning instead of pushing the number upward.

The score is a subjective description, not a diagnosis or a ratio measurement. Moving from 3 to 4 does not establish a one-third improvement. A lower number may reflect clearer understanding, and a client can decline numerical ratings altogether.

5. Small-step questions: choose a proportionate action

Ask: “Which part would be worth doing within the time you actually have?” Connect it to the client's purpose. A small action might be a bounded request, checking a condition or explicitly deciding not to proceed.

Specify who owns the action and what would count as completing it. Sending a request is different from receiving an answer, obtaining agreement and carrying out the work. Do not collapse those stages into one task called “sort it out.”

Small does not necessarily mean immediate. If a necessary person is unavailable, record that condition. A genuine delay or a chosen stop can be more useful than assigning busywork to keep momentum looking positive.

6. Coping questions: notice support without demanding endurance

An original prompt is: “Which things are helping a little, and which demands need changing?” It can reveal practical support or a boundary the client wants to retain. Acknowledge the answer in its actual context.

Someone may not want praise for continuing under unreasonable demands. Accept that correction. Ask whether they want to discuss reducing, redistributing or stopping something rather than celebrating persistence.

If the conversation reveals harm or a need outside the coaching agreement, address that need within your role. A question about strengths must not become a reason to overlook protection, necessary professional support or the client's wish to stop.

7. Resource questions: establish what is available

Ask: “What would make this possible, and which of those resources are already agreed?” Resources include time, practical knowledge, access, authority, money and another person's help. Separate a potential resource from a confirmed one.

A useful follow-up is: “What would you ask, and what could you do if the answer is no?” That makes room for a limited offer or another route without pressuring the client or the person being asked.

Count receiving work. A colleague reading a document or a partner taking on a household task has their own capacity. Calling an action small for the client does not make it free for somebody else.

8. Relationship questions: form a hypothesis, then check it

Ask: “If the other person found this arrangement clearer, what might they notice?” This can translate a private aim into an observable difference. It can also reveal whose experience the client has not yet considered.

The answer remains the client's hypothesis. Imagining another person's relief does not establish their feelings, agreement or availability. Where appropriate, a direct question can provide better evidence.

Do not imply that the client controls the relationship by choosing the right words. Clearer communication can coexist with another person's refusal, disappointment or continuing concern.

9. Past-success questions: identify a transferable element

Ask: “What did a different task teach you that might be useful here?” A client may identify asking earlier, separating two decisions or preparing a short first version. Follow that specific element rather than attaching a broad strengths label.

Examine where the analogy stops. A method that helped with a personal project may not satisfy a workplace's approval or safety requirements. The previous result does not confer authority in a new setting.

Past success and exceptions overlap. Use whichever wording helps the client, without asking both versions solely to complete a framework. The conversation should progress because an answer is useful, not because ten categories are available.

10. Difference questions: review what happened honestly

Ask: “What changed, what stayed the same, and what would you choose now?” Invite useful, unwelcome and unchanged results. Do not require a positive account as the price of being seen as cooperative.

Compare the chosen action with the actual result. A completed draft, an agreement on one date and a lasting change in a relationship are different findings. Record the narrow result the evidence supports.

The client can keep, revise or end the experiment. A review question should leave those options open. For an overall conversation structure, see the GROW guide, while retaining the client's own purpose and language.

Follow Rosa's answers through a complete case

All people and outcomes below are fictional. On Monday 5 July 2027, Rosa and coach Nina spend forty-five minutes considering how Rosa could finish a 90–110-word personal album caption about her windowsill herb garden. She wants two quiet twenty-minute writing periods that week, without turning the task into another continuing obligation. Nina spends fifteen minutes preparing and ten minutes writing session notes.

Accept a refusal and clarify the desired difference

Nina: Would it help to imagine a week where the difficulty had disappeared?

Rosa: No. The household work would still exist. I want one small piece of writing, not an imaginary quiet week.

Nina: Then what would make this particular week useful?

Rosa: Two twenty-minute periods and a caption I want to keep. I do not want a new weekly schedule.

Nina: Has any part been easier on a recent occasion?

Rosa: Leaving my phone in the hallway helped me stay with a short task. That is something I can repeat myself.

Nina does not return to the rejected hypothetical framing. When she asks what has helped Rosa cope with the busy week, Rosa says she does not want praise for always carrying on. Nina acknowledges that and asks which demand, if any, Rosa wants to alter for this experiment.

Rosa identifies the shared grocery order scheduled for Tuesday morning. She could ask partner Tariq to handle it on one date. Nina asks what Tariq might find clearer. Rosa guesses that he would prefer a precise request, then corrects herself: “That is what I think. I need his answer, not another assumption about him.”

They use a readiness scale with 0 meaning no workable request and 10 meaning a clear request that preserves a real no. Rosa chooses 3: she knows the task, but has not checked Tariq's availability. The number is a local description for the conversation, not a validated assessment.

Obtain the actual limited agreement

Rosa spends five minutes composing and sending this request on 5 July: “Could you handle our grocery order for twenty minutes on Tuesday 6 July, 10:00–10:20 Europe/London, and again on Thursday 8 July at that time, so I can write? I am asking about those two dates, not an ongoing arrangement. Please say if either does not fit.”

Tariq has reserved forty minutes for the work and communication he is willing to undertake in this finite experiment. He spends ten minutes reading and replying: “I can do the grocery order on Tuesday 6 July, 10:00–10:20. I cannot cover Thursday. I am accepting Tuesday only. We still need a separate discussion about our regular household responsibilities; one order does not settle that.”

Rosa spends ten minutes reading the answer, choosing another Thursday arrangement and sending her acceptance: “Thank you. I accept Tuesday only. I will use twenty minutes currently set aside for optional browsing on Thursday, without moving another task to you. I understand the regular household question is still open.” Tariq spends five minutes reading that acceptance. No additional acknowledgment is claimed.

The resulting plan is not the original request for two periods of partner help. It combines one accepted task transfer with Rosa's choice to drop an optional activity. Tariq's twenty minutes are counted as his work, and Rosa does not assume she has gained recurring support.

Complete the task without expanding its meaning

On 6 July, Rosa writes from 10:00 to 10:20 while Tariq performs the grocery order during his accepted twenty-minute window. Her phone remains outside the writing space. On 8 July, she uses the independently chosen 10:00–10:20 period for the second attempt. No message handling or household task is described inside either twenty-minute writing interval.

She spends another ten minutes refining the caption, then five sending Nina the caption and a selected conversation record. The final original caption reads:

“The basil began as a small pot beside a kitchen window. I kept moving it, looking for the place where the leaves stayed bright. Some mornings I remembered to turn it before making tea; other mornings I noticed only when the stems leaned toward the glass. This photograph keeps one ordinary stage of that experiment. The plant is neither a finished project nor proof that I have become an expert gardener. I like the uneven leaves and the empty space beside them. They remind me that a small thing can receive attention without becoming a whole new routine.”

The caption stays in Rosa's personal album. It is not published or submitted as evidence of gardening expertise. The writing task has a complete output; it does not require another person to approve its personal meaning.

Correct the actual coaching record

Rosa's selected record nevertheless says Tariq has agreed to a weekly quiet-time arrangement. On 8 July, Nina spends ten minutes reading the submission and five sending a comment in Rosa's private client conversation: “The agreement you described covers Tuesday 6 July only. Your record says weekly, which changes its scope. Please correct that distinction and keep Thursday's independently chosen time separate. The caption can remain as you wrote it.”

Rosa spends ten minutes reading, correcting and replying: “I have changed the record to one Tuesday grocery order. Thursday came from dropping my own optional browsing. I will not treat the two writing periods as Tariq agreeing to recurring cover. The wider household question remains open.” Nina spends five minutes reading that answer. These are actual handling periods, not extra feedback assumed to happen inside the original session.

At their twenty-minute review on 9 July, Rosa chooses 4 on the same readiness scale because she now has experience of making a request and receiving a partial no. Neither person interprets that as a measured improvement in wellbeing or a ratio increase in confidence. Rosa says she is pleased with the caption and still uncertain about future household arrangements.

She chooses not to ask for another writing commitment now. Nina records that ending in five minutes of notes. There is no new weekly task, a presumed agreement to discuss chores on a particular date or an assertion that the relationship issue has been resolved.

Organize selected preparation and real feedback in Coachful

Coachful is our #1 recommendation for organizing the selected worksheet, client conversation and follow-through in an ongoing coaching relationship. This is an editorial workflow recommendation, not evidence that the platform makes solution-focused questions more effective. Keep the client free to decline a prompt or keep private material out of the submission.

Use Forms & worksheets for an original selected conversation record, explaining the actual response audience. Internal Coach review is private to the authorized coach team. Nina's returned comment in the case is a separate message in Rosa's private client conversation, reached from the client profile's message route. Reading, correcting and replying are counted separately.

Unshared journal pages remain private, while clients can explicitly share particular pages through supported specialty-module routes. Confirm the chosen record and visibility rather than promising that every journal page can never be seen by a coach. The Coachful features support the working arrangement; submitting a record does not create Tariq's consent or an automatic weekly commitment.

Account for the whole finite experiment

The whole known account is 295 person-minutes against 330 reserved, leaving thirty-five: fifteen each for Nina and Rosa and five for Tariq. Rosa writes for forty minutes and refines for ten; Tariq's twenty-minute contribution remains visible. The cancelled browsing is an opportunity choice, not a second twenty-minute work item or a cash saving.

The result supports a narrow statement: a client used selected questions, obtained one real agreement, completed a caption and corrected an overstatement. It does not isolate the effect of the questions from her own work, Tariq's contribution, ordinary circumstances or the coaching relationship.

Frequently asked questions

Should I ask all ten types in one session?

No. Follow the client's purpose and answer. Several groupings overlap, and a well-chosen follow-up may be more useful than introducing another question. Rosa's case does not require ten separate techniques.

What if a client rejects an imagined future?

Accept the refusal and ask whether a direct description of the desired difference would help. Do not treat dislike of a framing as evidence of resistance or lack of motivation.

Must the client find an exception or give a scale score?

No. Invite either only where useful. A missing example or a preference for words can lead to a different conversation about support, constraints or next information.

Does solution-focused mean avoiding the problem?

No. Listen to the situation, its effects and necessary action. Exploring a useful difference must not prevent discussion of harm, structural conditions or needs outside coaching.

Can these questions establish that coaching caused progress?

No. Record the actual action, support and result, with remaining uncertainty. A completed task or a changed subjective rating does not establish causation, treatment benefit or a durable change in someone else's behaviour.