25 Coaching Techniques That Work in Real Sessions (and the Question Behind Each)
This is for the coach who is already trained and already has clients, and who still has sessions that drift. Forty good minutes, a client who leaves lighter, and nothing different by Thursday. The gap is almost never knowledge. It is that you reached for the wrong move at the wrong minute, or for nothing at all because the session was going fine.
These coaching techniques earn a place here only if each one passes three tests: you can run it in a normal session without announcing it, it works on a real person who is tired and slightly defensive, and it changes what happens after the call rather than only how the call feels. Anything needing a whiteboard or forty minutes of setup was left out.
Everything here was written and checked on 15 September 2026. There are no statistics, surveys or percentages in this article, because most of the numbers circulating in coaching content have no source you can click. Every script and example below is ours, written as an illustration rather than taken from a real client.
How were these 25 coaching techniques chosen?
They were chosen by session minute, not by school. Sorting techniques into cognitive, somatic, solution-focused and narrative buckets helps nobody at 10:04am when a client has gone quiet, so these are grouped by the moment you would reach for them.
Six moments cover almost every session, from the opening to the handoff, and each group below is one of them. A few techniques carry names and originators, and those are attributed; most are ordinary practice with no single owner. Each also had to have a failure mode worth naming.
Grouped by minute, not by school. All 25 techniques are sorted into the six moments of a session where you would reach for them, from the opening question to the handoff at the end.
What do you actually do in the first ten minutes of a coaching session?
You settle what this hour is for, in the client's words, before anything else. Most drifting sessions were lost in minute three, when the coach accepted "let's just see where it goes" as an agenda. Four jobs, in order: find out what changed, agree what the session is for, narrow to one thing, get permission to push.
1. Contracting the session outcome
Ask the client to name what they want true at the end of the hour, and write it down. Do it every session, including the ones where they arrive full of news.
Example: "By the time we stop, what would need to have happened for this to be worth your time?" The mistake is contracting and never referring back, which teaches the client it was a ritual.
2. The since-we-last-spoke scan
A short, factual sweep of what has happened, including what they said they would do and did not. Run it before contracting, because what actually happened often changes what the session should be about.
Example: "Quick version of the last two weeks. What moved, what stalled, what did you not get to?" Ask for all three or you hear only highlights. The mistake is letting the scan become the session, so cap it at five minutes out loud.
3. Agenda triage
When a client brings three topics, make them choose one out loud, with a reason. Coaching one thing properly beats touching three politely, and clients usually know which one is real.
Example: "If only one of those was different next month, which makes the other two easier?" That last clause does the work. The mistake is choosing for them because one topic interests you more.
4. Asking permission before you push
One line before you challenge, interrupt, or offer an observation nobody asked for. Use it the first time in a session that you are about to say something uncomfortable.
Example: "Can I say something that might land badly?" They almost always say yes, which is the point, because now they have chosen the challenge. The mistake is using it on every sentence, so the client ends up managing your comfort.
Which coaching frameworks give a session a shape when you lose the thread?
Three: GROW, OSKAR and CLEAR. Each is a sequence of stages you can drop into silently, and none of them need to be shown to the client. Treat them as recovery tools, not scripts. Noticing at minute twenty-five that the client has described reality for a quarter of an hour with no options on the table is using GROW correctly.
5. GROW
Goal, Reality, Options, Will. Developed in the 1980s by Sir John Whitmore with Graham Alexander and Alan Fine. Use it when a session has energy but no direction, especially around a decision.
Example: "get on top of the business" becomes a Goal of "decide by Friday whether to hire the operations person", then options, then a first move with a date. The mistake is spending most of the hour in Reality and rushing Will at the end.
6. OSKAR
Outcome, Scaling, Know-how, Affirm and action, Review. A solution-focused model set out by Mark McKergow and Paul Z. Jackson. Use it with clients stuck in problem analysis.
Example: instead of asking why the team meeting is dysfunctional, ask what a good one looks like, rate today, then ask what they know about the days it went better. The mistake is skipping Affirm, which is not flattery but raw material for the action step.
7. CLEAR
Contract, Listen, Explore, Action, Review. Developed by Peter Hawkins, and the most useful of the three for long relationships because contracting and review are explicit stages rather than assumptions.
Example: with a client six months in whose sessions have become pleasant chats, reopening the contract resets everything: "what are we doing here now, given you already solved what you hired me for?" The mistake is treating the contract as admin you did once.
How do you help a client say what they actually want?
You make the wanted thing concrete enough to notice when it arrives. Most clients bring a direction ("more balance", "grow the business") and no destination, and a direction cannot be planned, checked or finished. Analytical clients take to outcome framing and scaling; clients tired of thinking do better with the miracle question.
8. Outcome framing
Pin four things to the stated want: what specifically, by when, what proves it, and what is in their control. Use it whenever a goal cannot be falsified, which is most of them.
Example: "less stressed at work" becomes "I leave by six on at least three days, and I do not open email after eight". The mistake is negotiating the goal down to something trivial just to get a clean sentence.
9. The miracle question
Ask the client to describe tomorrow morning as if the problem had been solved overnight without their knowing. It comes from solution-focused brief therapy, developed by Steve de Shazer and Insoo Kim Berg, and works because describing a solved future is easier than designing one.
Example: "You wake up and do not know it happened yet. What is the first small thing you notice?" The mistake is accepting the grand answer ("I'd feel free") and moving on instead of asking what they would be doing at nine.
10. Scaling questions
Ask for a number from one to ten, then two follow-ups: why not lower, and what makes it one point higher. Also solution-focused in origin. Use it for progress, confidence, motivation, anything real but not countable.
Example: confidence sits at four. "Why not a two?" produces evidence of strength they would not volunteer. The mistake is asking what would make it a ten, because nobody plans a six-point jump and the client will invent an answer to be helpful.
11. The future-self letter
The client writes a short letter from themselves two years out, dated, describing an ordinary week in the present tense. Use it early, or with a client who can say exactly what they do not want and nothing about what they do.
Example instruction: "Write it as a Tuesday, not a highlight reel. What has stopped being your problem?" The mistake is doing it in session. Send it as between-session work, then read it back to them next time.
12. Values elicitation through peak moments
Instead of ranking a list of values, ask for three specific moments when work felt right, then mine them for what was present each time. Use it when a client is choosing between two acceptable options and no spreadsheet will settle it.
Example: a launch night, hard feedback delivered well, a Saturday teaching their kid to drive. What repeats might be "being trusted with something hard". The mistake is stopping at the label, since "autonomy" means nothing until they say what its absence feels like.
How do you get past the story a client keeps telling?
You change the position they are looking from, not the facts they are looking at. A repeated story is not stubbornness, it is a worn path, and harder questions along it only deepen it. These five are the riskiest here: at the right moment they open a session in one question, but used before trust exists they read as clever, and a client who feels handled stops telling you the true version.
13. Reframing
Offer a different reading of the same events, lightly, as an option rather than a correction. Use it when the interpretation is doing more damage than the event, and only after you have shown you understood the original reading.
Example: "they went around me to my boss" offered back as "they wanted a fast answer and did not know your process", said as a question. The mistake is reframing pain away, which is gaslighting with better manners.
14. Laddering down
Take the stated want and keep asking what having it would give them, until the answers stop changing. Use it when a goal feels borrowed, or when a client keeps achieving things and feeling nothing.
Example: the promotion gives money, which gives choices, which gives the ability to stop asking permission. The bottom rung is the one to coach. The mistake is running six rungs mechanically, when the tell is a pause and a change of voice.
15. Perspective chairs
Ask the client to describe the situation from a second position, usually the other person, then from a detached third. It comes from Gestalt practice, where it is done with actual chairs, and works on video with nothing but a change of tone.
Example: "Be your co-founder for a minute. What does he think this meeting is about?" Then ask what a sensible stranger sees that neither of them does. The mistake is rushing the switch, which turns it into a party trick.
16. Metaphor work
Ask what the situation is like, then stay inside whatever image they give you and ask questions in its language. Use it with clients who are articulate about everything except how they feel.
Example: the job is "like driving with the handbrake on", so ask what the handbrake is, who put it on, what happens if it comes off. The mistake is swapping their image for a better one of yours, which kills the material instantly.
17. Externalizing the problem
Give the problem a name and speak about it as a separate thing beside the client, not a trait inside them. The move comes from narrative practice. Use it for self-criticism, procrastination, and anything described as "just how I am".
Example: "How long has the Committee been giving you notes on your writing, and when is it quietest?" The mistake is using it on things the client should own. Externalize an inner critic, not a missed deadline that hurt a colleague.
The five riskiest moves sit in one group. Reframing, laddering, perspective chairs, metaphor and externalizing all work by moving the client's position, and all five read as tricks if you use them before trust exists.
What do you do when a client says "I don't know"?
You wait, and if waiting fails you remove the requirement to know. "I don't know" is nearly always one of three things: they need three more seconds, the true answer is socially expensive, or the question was badly built. Do not rescue, because a multiple-choice list teaches the client that going blank produces answers from you.
18. Ten seconds of silence
Say nothing after the question, longer than is comfortable, with a neutral face. Ten seconds is the working number and it feels like a minute. Use it after any question that deserves a real answer, especially on video.
Example: after "what are you actually afraid of here?", count to ten and let them arrive. The first answer is the prepared one, the second is the real one. The mistake is filling the gap with a rephrase, which is an escape hatch for both of you.
19. The hypothetical bypass
Accept the "I don't know" and ask the same question with the knowing requirement removed. Use it after silence has failed once, not as the first response.
Example: "Fair enough. If you did know, what would it be?" It works more often than it should, because the client usually knows and does not want to commit to saying it. The mistake is using it twice in a row, since a second failure means the block is real.
20. Immediacy
Describe what is happening in the conversation right now, with no interpretation attached, then hand it over. Use it when the content is fine but something in the room is off: energy dropped, answers got shorter, a topic keeps being deflected with a joke.
Example: "Every time we get near the money conversation, we end up back on the website. What do you make of that?" Then stop talking. The mistake is loading the observation with your theory, which is a diagnosis and sends the client into defending themselves.
21. The competing commitment map
Treat the failure to act as evidence of a second, hidden commitment doing its job perfectly. The approach comes from Robert Kegan and Lisa Lahey's immunity-to-change work. Use it on the goal that has not moved in months despite genuine wanting.
Example: the stated commitment is delegating, and what happens instead is taking work back at the last minute, because a second commitment says never be seen as sloppy. The mistake is calling that self-sabotage. It is protection, and deserves respect before renegotiation.
How do you turn an insight into something that actually happens?
You shrink the first action until it is almost embarrassing, then attach it to a specific time and an existing habit. Insight decays fast, and a client who leaves with a resolution instead of a scheduled action is relying on motivation that will not show up. These four run in sequence and take about six minutes.
22. The two-minute first action
Ask for the smallest version that can be done in about two minutes, and refuse anything bigger as the first move. Use it whenever the proposed action starts with "I'll sit down and properly...".
Example: "restructure the pricing" becomes "open a blank doc and write the three packages by name, nothing else". The mistake is letting them talk you back up to the big version, because two minutes usually becomes twenty and two hours becomes nothing.
23. If-then planning
Convert the action into a sentence with a trigger in it: if this happens, then I do this specific thing. The term implementation intention comes from psychologist Peter Gollwitzer. Use it for anything that has to survive a normal week.
Example: "I'll do my thinking time" becomes "if it is Tuesday and standup has finished, then I close the laptop and walk for twenty minutes". The mistake is writing the if-clause as a feeling, because "if I feel motivated" is a wish with formatting.
24. Accountability design
Agree four things out loud before the session ends: what exactly, by when, how you will both know it happened, and what the client wants you to do if it does not.
Example: "Draft to Maria by Thursday six o'clock, you message me the confirmation. If it has not gone, do you want me to ask on Friday, or wait for you to raise it?" Coaches skip that last question, and it is what makes follow-up feel agreed rather than parental.
25. Habit anchoring
Attach a new behavior to an existing routine that already happens without thought, rather than to a time of day. Use it for anything the client wants to repeat rather than complete.
Example: "after I put the coffee on, I write three lines while it brews". The old habit is the alarm clock. The mistake is anchoring to an unstable routine, so if mornings vary, use the end of the last meeting or the train pulling out.
How do you end a session so the work survives the week?
You get the takeaway back in the client's own words, write down what was agreed where you can both see it, and set one point of contact before the next session. The last four minutes decide more than the middle thirty.
Ask for the summary, never give it. "What are you taking from this?" produces the version they actually absorbed, which is often not the part you thought mattered. Then write it down: a short note they can read back, the action with its date, the if-then sentence in their words, kept apart from your clinical notes. Then agree one check-in between sessions, small enough that it is not a second appointment.
The last four minutes. Takeaway in the client's own words, the action with a date, the if-then sentence and one agreed check-in take about six minutes, and skipping them is why good sessions produce nothing.
Doing this in Coachful
Coachful is a coaching platform for solo coaches and small teams who deliver programs rather than only calls. It costs $29 a month on Lite for up to 5 clients, $49 on Solo for up to 20, and $99 on Pro for unlimited clients, with a 7-day trial that requires a card and charges nothing during the trial, and a 0% platform fee on what you sell. There is no free plan, it is not HIPAA-certified, and it is not a workout builder or a nutrition database. Public rating is 4.5 on Trustpilot from 21 reviews.
None of the 25 techniques need software. What software changes is whether the fourth one survives to the following Tuesday.
Goals and scaling. Client goals carry progress, milestones and a weekly reflection prompt, so a scaling question becomes a rating re-entered each week instead of a number you both forget. Habits and anchoring. Daily habits with morning and evening check-ins sit on the client home and in the mobile app, where an anchored behavior has to be visible to hold. The action. Programs carry weekly goals and daily tasks, so the two-minute action becomes a dated task rather than a line in your notebook.
The written summary. Session notes attach to the client record and the client hub, and a pre-session worksheet can be required before the next booking, which handles the future-self letter and the miracle-question homework without a separate email. Seeing who has gone quiet. The clients view shows task completion, check-in rate and who needs attention, so you find the stalled action before the session. Structure. Journeys and flows sequence the opening contract, the mid-point review and the closing session as steps, which is the CLEAR review stage made into a default. Preparation. Michelle, the built-in assistant, works from live practice data across clients, programs, sessions and payments and briefs you beforehand, so you walk in knowing what they said they would do.
The honest limit: software will not make a technique land. It only stops the good ones evaporating between Thursday and the next call. For the wider picture, see coaching practice management software or what Coachful includes.
What does a full 50-minute session look like with these techniques in it?
It looks like four or five techniques, not fifteen, with most of the hour spent on ordinary conversation. Here is one shape, as an example, for a client who has wanted to raise her rates for eight months.
Minutes 0 to 8. The scan, then contracting. One of three emails went out, and she wants to stop flinching at the pricing conversation. Minutes 8 to 18. Scaling. Confidence about the new rate is a three, and "why not a one?" surfaces that she raised prices two years ago and nobody left.
Minutes 18 to 32. The real work. She repeats the sentence she has used for three months, about not wanting to be greedy. You go quiet for ten seconds, then use immediacy: "every time we get close to a number, we end up on other people's feelings." That opens the competing commitment: she wants higher rates and equally wants never to price people out, which is why eight months of intention did nothing.
Minutes 32 to 48. Laddering down on the second commitment, a reframe offered lightly, then the close. Two-minute action: write the new number in a note on her phone today and tell nobody. If-then: if Monday's lead asks about price, she says the number and stops talking. Minutes 48 to 50. Her takeaway, in her words, and it is not the part you expected.
Five moves, not fifteen. The worked 50-minute session above uses the scan, contracting, scaling, silence and immediacy, and roughly twenty minutes of it has no technique in it at all.
What goes wrong when coaches learn new coaching techniques?
They use them too often, too early, and on the wrong clients. A technique learned last week has a gravitational pull, and the next four sessions all turn out to be about competing commitments.
- Technique stacking. Three moves in ten minutes, none given room. The client experiences it as being processed.
- Running the framework out loud. Announcing "so let's move to Options" turns a conversation into an interview.
- Insight as the finish line. The session where they say "oh, wow" feels like your best work, then nothing happens, because you both enjoyed it too much to spend six minutes on the action.
- Coaching the topic you find interesting. The tell is that sessions with different clients keep converging on one theme. That theme is yours.
- Challenge without the relationship. Reframing, immediacy and externalizing need credit in the account; in session two they read as clever.
The fix is a constraint: one new technique a month, used deliberately in about three sessions, with two lines written afterward. That is twelve a year, more than most coaches add in five.
How do you practice a new technique without experimenting on a paying client?
You run it on low-stakes material three times before it goes near a real session. An afternoon is enough to take a technique from read-about to usable.
Write it on one index card: the exact opening line, the two follow-ups, the failure mode. If it does not fit on a card you have not understood it. Then run it on yourself, out loud, on a real problem of your own. Then with a peer coach on a minor issue of theirs, asking afterwards how the timing felt. Then once with a long-standing client who likes you, where an awkward delivery costs nothing.
Peer practice is also what mentor coaching and supervision hours are for, and those hours have to be logged if you are working toward a credential. Our guide to ICF mentor coaching covers the requirement, and Coachful's coach development panel counts hours and clients against the ACC, PCC and MCC thresholds so the log is a by-product of the work.
So which coaching techniques should you learn first?
Four, in this order: contracting the session outcome, ten seconds of silence, scaling questions, and the two-minute action with an if-then sentence attached. Those cover the opening, the moment you would usually rescue, the middle when you need something measurable, and the close. A coach who does only those four well is ahead of most.
Add GROW next if your sessions drift, and immediacy if they stay polite instead. After that pick by symptom, not curiosity. Clients who cannot describe what they want need the miracle question and outcome framing; clients who describe it perfectly and never move need the competing commitment map; clients stuck in conflict need perspective chairs.
Two exceptions. If you coach inside organizations, learn CLEAR before GROW, because explicit contracting and review are what make three-way work with a sponsor manageable. If you coach groups, most of the challenge techniques need adapting before they are safe in front of an audience, and group coaching platforms covers that from the delivery side.
Frequently asked questions about coaching techniques
What is the difference between a coaching technique and a coaching model?
A model is the shape of the whole conversation; a technique is one move inside it. GROW, OSKAR and CLEAR sequence a session from start to finish, while scaling, silence, reframing and if-then planning drop into any of them when needed. You can coach well with one model and twenty techniques. Twenty models and no techniques gives you a very organized session with nothing happening in it.
How many coaching techniques should I use in one session?
Four or five, and most of the hour should be conversation with no technique in it. The worked session above uses five over fifty minutes and already feels full from the client's side. If you are reaching for a fourth move in ten minutes, usually you are uncomfortable, not the client stuck.
What coaching techniques work best for a first session with a new client?
Contracting, outcome framing and scaling. A first session has to produce a shared definition of success and a baseline you can both refer back to in three months. Save reframing, immediacy and externalizing for later, because challenge needs credit you have not earned. Peter Hawkins' CLEAR model suits a first session because contracting is an explicit stage in it.
Which coaching techniques help when a client is not doing the work?
The competing commitment map first, then accountability design. A client who wants the outcome and does not act is protecting something, and treating it as a discipline problem makes them defensive and makes you nag. Once it is named, rebuild the action smaller: a two-minute step, an if-then with a real trigger, and follow-up the client chooses.
Are there coaching techniques I should not use without training?
Anything that goes near trauma, and that boundary appears more often than coaches expect. Externalizing, perspective chairs and deep laddering can surface material that belongs in therapy. The signal is a response disproportionate to the topic, or the same distress whatever the subject. Stop, name what you notice, and refer. Knowing two or three therapists is part of being equipped.
What is the best coaching technique for building confidence?
Scaling with the "why not lower" follow-up, because it makes the client produce evidence of their own competence instead of hearing yours. Someone who rates confidence at three and explains why it is not a one has listed things they already handled. Pair it with a two-minute action, since confidence follows evidence.
Can I use these coaching techniques in group coaching?
Scaling, outcome framing, if-then planning and habit anchoring work in a group unchanged. Contracting has to happen at two levels, for the group and for each person. The challenge techniques are the problem: immediacy, externalizing and perspective work expose someone in front of peers, so keep those for breakouts, one-to-one slots, or where permission was asked in advance.
How do I track which techniques worked with which client?
Two lines in the session note within ten minutes of the call: what you used and what it produced. Patterns appear after about six weeks and are specific to the person, since a client who responds to scaling often goes flat on metaphor. In Coachful the note sits alongside their goals, habits and check-ins, so technique and result sit with the evidence.
Join Coachful now
Techniques are free and portable, and the reason they stop working is almost never the technique. It is that the two-minute action, the if-then sentence and the check-in never made it out of your notebook into the client's week.
Coachful puts the goal, the habit, the task, the check-in and the session note on one client record with your branding on it, from $29 a month on Lite and $49 on Solo for up to 20 clients, with a 7-day trial and a 0% platform fee. Run one existing client through a full cycle and see if the follow-through changes.
