What Is the GROW Model in Coaching? Stages, Origins, and Questions
What is the GROW model in coaching?
The GROW model is a four-stage coaching conversation: Goal, Reality, Options, and Will (sometimes Way Forward). This guide explains each stage, gives example GROW model questions for each one, lays out where the model came from (including who gets credit, which is disputed), what it cannot do, and how to run it on yourself.
The GROW model is a way of structuring a coaching conversation in four stages: Goal (what do you want), Reality (what is true right now), Options (what could you do), and Will, which some versions call Way Forward (what will you actually do). The person being coached does most of the thinking; the coach mostly asks questions in that order instead of handing over advice. It is usually credited to Sir John Whitmore and his colleagues, became widely known through Whitmore's 1992 book Coaching for Performance, and is one of the most commonly taught frameworks in workplace coaching.
That is the short answer. The rest of this page covers what each stage is for, where the model came from (the credit is genuinely disputed, and we say so rather than picking a side), a long section of example GROW model questions you can adapt, how a conversation tends to move through the stages, what the model leaves out, and how to use it on yourself when there is no coach in the room.
The four stages and what each one is for
Goal comes first because a conversation with no target tends to wander. The aim of this stage is to turn "I want to talk about my job" into something the person is actually pulling toward: an outcome for the conversation itself, and often a larger outcome behind it. A coach at this stage is mostly listening for the difference between a topic and a goal.
Reality is the honest look at where things stand now. What has happened, what has been tried, what is getting in the way, what is already working. The point is not to build a case for why the goal is impossible; it is to give the next two stages something true to stand on, since options generated against a wrong picture of the situation solve the wrong problem.
Options is where the person widens the field: what could they do, including things they would normally dismiss. Generating possibilities and judging them are different mental jobs, and the stage works best when the first is allowed to finish before the second begins.
Will is where one option becomes a commitment. The word is doing deliberate work: the question has moved from what could be done to what this person will do, by when, and what might stop them. Some versions of the model write the last letter as Way Forward, and others split the O into Options and Obstacles; there is no single official wording, which is part of the history below.
Where the GROW model came from, and why the credit is disputed
The common account is that GROW grew out of the Inner Game approach developed by the tennis coach W. Timothy Gallwey, whose book The Inner Game of Tennis appeared in 1974. Gallwey noticed that telling players what to fix often did not produce lasting change, and found that asking questions that helped players notice what they were already doing worked better. GROW is generally described as carrying that question-led approach into business coaching, and is usually dated to the late 1980s and early 1990s.
Who deserves credit for the model itself is less settled than most summaries suggest. The most widely cited published version is Whitmore's Coaching for Performance, first edition 1992, and the model is most often attributed to Whitmore together with colleagues Graham Alexander and Alan Fine. But accounts differ on the details. In a 2009 article, Whitmore is recorded as saying that Max Landsberg coined the name GROW in a conversation with Graham Alexander, and that the model had already been in use for some time before it had that name. Landsberg published it in the first edition of The Tao of Coaching in 1996. And in his 2010 book You Already Know How to Be Great, Alan Fine claimed to have co-developed the model with Whitmore and Alexander.
We have not found a primary source that settles who did what, and this page does not try to. What the sources agree on is the shape of the model and the fact that it was in circulation among coaches before the 1992 book made it famous. If you need to cite an origin, a safe wording is that the model is associated with Whitmore, Alexander, and Fine, was popularised in print by Whitmore in 1992, and that the details of its authorship are contested.
Explore: the inner game
How a GROW conversation actually goes
A GROW conversation starts with a person who has something on their mind and a coach who is trying not to solve it. The coach opens with the Goal: what would make this time well spent, and what would the person like to be different afterwards. Once there is something concrete to aim at, the conversation turns to Reality, where the coach's job is mostly to ask and to reflect back what they hear, without hurrying to a fix.
Options usually feels like the shortest stage and often decides the outcome. The coach asks for more possibilities than the person first offers, holds back their own suggestions, and avoids calling any idea good or bad while the list is still growing. Then Will: the person picks something, says when they will do it, and names what could get in the way and what they would do if it did.
The stages are a sequence on paper and a loop in practice. A person describing their Reality may realise the Goal they stated was not the real one, and a conversation can reasonably go back to it. The order is a default that stops you skipping steps, rather than a rule against revisiting one.
GROW model questions: how to choose them
The questions are where the model succeeds or fails, which is why this is the part most people searching for the model are really after. A few principles hold across all four stages. Prefer open questions that cannot be answered yes or no. Ask one question at a time and leave the silence afterwards. Watch for the question that is really advice with a question mark on the end, such as "Don't you think you should talk to your manager?"; it tells the person what to conclude instead of inviting them to work it out. And treat the lists below as a starting vocabulary, not a script: the best next question is nearly always one that follows from the answer you just heard.
Explore: socratic questioning · calibrated questions
GROW model questions for the Goal stage
These questions are trying to turn a topic into a target. "What would you like to get out of this conversation?" sets the goal for the session. "If this goes well, what is different when we finish?" makes it concrete. "And if you get that, what does it give you?" often surfaces the larger goal behind the immediate one. "How will you know you have got there?" tests whether the goal is specific enough to recognise. "Is this something you want, or something you feel you should want?" separates a goal the person owns from one they inherited. If the answer comes back vague, reflect it back and ask the person to sharpen it before moving on.
Explore: goal setting theory · the miracle question
GROW model questions for the Reality stage
These questions build an honest picture without turning into blame or a lecture. "What is happening right now?" is the plain opener. "What have you tried so far, and what happened?" brings out history and prior effort. "What is working, even a little?" prevents the picture from going all negative. "On a scale of one to ten, where are you relative to your goal, and what makes it that number rather than a lower one?" is a common way to surface strengths and progress. "What is getting in the way, and which of those things are in your control?" separates obstacles from circumstances. "What have you not said yet?" leaves room for the thing the person has been circling. Keep your own interpretation out of it for as long as you can; Reality is the person's account, not yours.
GROW model questions for the Options stage
These questions are aimed at quantity first and judgment later. "What could you do?" is the baseline. "What else?", asked repeatedly, does a lot of work in this stage, because the first answer is often the obvious one. "What would you do if you had no constraints of time or money?" loosens the self-editing. "What would someone you respect do here?" gives the person a different seat to think from. "Who could help, and what could they do?" brings in resources. Only once the list has stopped growing: "Which of these appeals to you, and what makes it appeal?" and "What would each cost you, and what would each get you?" If you have an idea yourself, ask permission before offering it ("Would you like a suggestion?"), and offer it as one more option rather than the answer.
GROW model questions for the Will stage
These questions convert a preference into a plan. "What will you do?" is the core, and it is deliberately different from "what could you do." "When exactly will you do it?" turns intention into a date. "What might get in the way, and what will you do if it does?" is a form of if-then planning; research on implementation intentions, notably by Peter Gollwitzer, finds that specifying when and where you will act tends to help people follow through, though that research is about plans in general, not about GROW. "How committed are you, from one to ten?" is a common check, and if the answer is low, the better move is usually to ask what would make it higher or what the person would need to change about the plan, instead of pushing. "Who will you tell, and how will you check in?" adds a light form of accountability. A well-run session often ends by asking what the person has noticed or decided that they did not know when they sat down.
Explore: woop method
A short illustration
The following is invented to show the flow, not a transcript of a real session. Someone says they are exhausted by their workload. Goal: "What would you like from this conversation?" They answer that they want a workload they can sustain, and when asked what that looks like, say two evenings a week with no work. Reality: "What does a typical week look like now? What has you working in the evenings?" They realise most of it is replying to messages they could batch. Options: "What could you do about that? What else?" They list batching replies, telling their team their response hours, asking a colleague to cover Thursdays, and, after another "what else?", saying no to one recurring meeting. Will: "Which of those will you do, and by when?" They choose to block out two evenings and tell their team on Monday, and say the likely obstacle is guilt, so they plan to ask a colleague to hold them to it. Notice what the coach did not do: suggest any of the options.
What the GROW model does not do well
The first limit is a question of scope. Jonathan Passmore and Stefan Cantore argued in 2012 that a goal-directed model like GROW may suit goal-oriented areas such as sport or business better than careers conversations, fit questions, or life-coaching conversations where existential or transpersonal approaches may help more. Whitmore, Kauffman and David (2013) responded that GROW has evolved to include transpersonal goals. Both positions are in the published record, and a reader can reasonably weigh them differently.
The second limit is that GROW is a structure rather than a skill. Whitmore's own consultancy records him saying that any dictator can use GROW, and draws the conclusion that the framework is meant to be used alongside coaching skills rather than alone. Four letters on a card will not stop someone running the stages as an interrogation, or asking leading questions while following the order perfectly.
The third limit is the evidence. A meta-analysis of workplace coaching by Theeboom, Beersma and van Vianen, covering 18 studies, found significant positive effects of coaching on outcomes such as performance and skills, well-being, coping, work attitudes, and goal-directed self-regulation. That is evidence about coaching in general, not about GROW specifically. Goal-setting research by Edwin Locke and Gary Latham supports a related idea, that specific, difficult goals tend to produce better performance than vague ones when certain conditions are met, and that too is evidence about a mechanism the Goal stage relies on, not about the model as a package. In our search we did not find a trial comparing GROW with other conversation structures; that is a statement about what we found, not proof that none exists.
There are also several differing versions of the model, with different descriptions of what belongs in each stage, and published extensions that add steps. One example is the GROWS model, proposed by Sasi Panchal and Pamela Riddell in 2020 to help people start and sustain behaviour change, which the authors tested with four coaching clients. That is a small sample, and we mention it as an example of how people extend GROW rather than as proof that the extension works.
Using the GROW model on yourself
GROW can be run alone, on paper, and many people do. Write the four headings and answer one stage at a time, using the questions above. Writing the Goal as a full sentence, then asking yourself how you would know you had reached it, does more than it looks like it should. In Reality, aim for facts and observations before interpretations. In Options, set a number, such as ten possibilities, and keep going past the first three. In Will, write the action, the date, and the most likely obstacle with what you will do about it.
The trade-off is that a coach is a second mind, and you are not. Alone, you will tend to ask yourself the questions you already expect, accept the first answer, and skip the stage that makes you uncomfortable. Some people compensate by reading their answers aloud to someone, or by returning to the page a few days later and asking what they notice. Others prefer to talk the structure through with another person or an AI coaching tool, which can ask the next question when you would not. If you are specifically looking for software that runs GROW, our separate guide on that question lays out what exists and what the options are.
Where IX Coach fits
This page is about the model, not about us, and it should be useful whether or not you ever open our product. For readers who want to try a conversation instead of a worksheet: IX Coach is an AI coaching system you can talk to directly, designed around reflection, goal clarity, and the move from insight to action, and it is disclosed as an AI coach rather than a person. GROW is one of the coaching methods published on this site as readable pages you can open and check, including a breakdown of the stages and the questions inside each. Whether a structured four-stage conversation or a more open one suits you is something only trying both can answer.
Explore: the grow model · the grow coaching model
Frequently asked questions
What does GROW stand for? Goal, Reality, Options, and Will. Some versions write the last letter as Way Forward, and some expand the O to include Obstacles.
Who created the GROW model? It is most often associated with Sir John Whitmore, Graham Alexander, and Alan Fine, and was published by Whitmore in Coaching for Performance in 1992. The accounts of who did what differ, as described above, so credit is better treated as contested than settled.
Is the GROW model only for managers? No. It is a conversation structure that has been used in workplace coaching, and it can be used on any goal or problem a person wants to work through, including by yourself. Its critics argue it fits goal-oriented topics better than open-ended ones.
Do you have to follow the stages in order? The order is the default because each stage gives the next something to build on, but conversations often revisit an earlier stage, and doing so is normal.
Is there proof that GROW works? There is evidence that coaching in general has positive effects, and that specific goals help performance, but we did not find controlled research isolating GROW itself from other approaches.
What is the GROW model in coaching?
The GROW model is a four-stage coaching conversation: Goal, Reality, Options, and Will (sometimes Way Forward). This guide explains each stage, gives example GROW model questions for each one, lays out where the model came from (including who gets credit, which is disputed), what it cannot do, and how to run it on yourself.
What does GROW stand for?
Goal, Reality, Options, and Will. Some versions write the last letter as Way Forward, and some expand the O to include Obstacles.
Who created the GROW model?
It is most often associated with Sir John Whitmore, Graham Alexander, and Alan Fine, and was published by Whitmore in Coaching for Performance in 1992. The accounts of who did what differ, as described above, so credit is better treated as contested than settled.
Is the GROW model only for managers?
No. It is a conversation structure that has been used in workplace coaching, and it can be used on any goal or problem a person wants to work through, including by yourself. Its critics argue it fits goal-oriented topics better than open-ended ones.
Do you have to follow the stages in order?
The order is the default because each stage gives the next something to build on, but conversations often revisit an earlier stage, and doing so is normal.
Is there proof that GROW works?
There is evidence that coaching in general has positive effects, and that specific goals help performance, but we did not find controlled research isolating GROW itself from other approaches.
Research
- Whitmore, J., (1992), Coaching for Performance, First edition, 1992 — The book through which the GROW model became widely known in print; its authorship history is described in the GROW model entry on Wikipedia.
- GROW model, Wikipedia — A summary of the competing accounts of who named and developed GROW (Whitmore, Landsberg, Alexander, Fine), the Inner Game connection, and the Passmore and Cantore criticism with Whitmore, Kauffman and David's response; a secondary source, so the primary texts it cites are worth reading directly.
- The GROW Model for coaching, Performance Consultants — The consultancy Whitmore co-founded; it records his remark that any dictator can use GROW and describes the four stages.
- Theeboom, T., Beersma, B., & van Vianen, A. E. M., (2014), Does coaching work? A meta-analysis on the effects of coaching on individual level outcomes in an organizational context, The Journal of Positive Psychology, 9(1), 1-18 — Evidence that workplace coaching in general is associated with positive outcomes; it does not test GROW against other approaches.
- Locke, E. A., & Latham, G. P., (2002), Building a practically useful theory of goal setting and task motivation: A 35-year odyssey, American Psychologist, 57(9), 705-717 — Review of goal-setting research behind the idea that specific, challenging goals tend to outperform vague ones under certain conditions; it concerns goals in general, not GROW.
- Gollwitzer, P. M., (1999), Implementation intentions: Strong effects of simple plans, American Psychologist, 54(7), 493-503 — Background on if-then planning, the mechanism behind asking when and where you will act and what you will do if an obstacle appears.
- Panchal, S., & Riddell, P., (2020), The GROWS model: Extending the GROW coaching model to support behavioural change, The Coaching Psychologist, 16(2), 12-25 — One published extension of GROW, tested with four coaching clients; useful as an example of how the model gets adapted.
- Gallwey, W. T., (1974), The Inner Game of Tennis — The book generally described as the starting point of the Inner Game approach that GROW is said to build on.
Practice this with IX Coach
Keep reading
- The Best GROW Model Coaching App: What Actually Exists
What is the best GROW model coaching app?
- What Is AI Coaching? A 2026 Guide to Technology, Ethics, and Behavior Change
What is AI coaching?
- GROW Model vs SMART Goals and Other Coaching Models
GROW model vs other coaching models: which should you actually use?
- Kegan's Stages vs. Spiral Dynamics: Which Developmental Map Is More Useful?
Kegan's stages vs spiral dynamics: which developmental map is more useful?
- Team Coaching vs. Group Coaching vs. Individual Coaching: What Each One Actually Is
What is team coaching, and how is it different from group or individual coaching?
- The Best Tool for Applying the AQAL Model to Your Own Growth
What's the best tool for actually applying the AQAL model to your own growth?
- Which Tool Actually Assesses Your Skill Level Against the Dreyfus Model?
Which skill level assessment tool actually uses the Dreyfus model, and is it worth using?
- Goal-Setting Coaching, Including the Part the Field Leaves Out
What is goal-setting coaching and does setting goals actually work?