Business
GROW coaching model: stages, questions and a worked example

Stefan Weiss
The GROW coaching model structures a conversation around Goals, Reality, Options and Will. In practical terms: clarify what the person wants, examine the current situation, explore possible ways forward and agree what they intend to do next. “Will” is also described as the way forward.
The value is the thinking each part encourages, not filling four boxes. A conversation may return to the goal after the situation becomes clearer, or revisit the options when a proposed action turns out to be unrealistic.
The Open University’s introduction to GROW presents it as a framework for structuring sessions and describes flexible use of the associated skills. Performance Consultants’ model overview sets out the same four areas. This article offers an independent explanation and original fictional examples, rather than reproducing a training worksheet or branded diagram.
What each stage is trying to achieve
A useful session leaves the person clearer about a chosen issue and a realistic next step. That may be a small experiment or a question to investigate, rather than a dramatic solution.
Goals: choose the outcome of this conversation
Start by distinguishing a wider aspiration from what can be usefully explored now. “Become a better manager” is a broad direction. “Work out how to delegate next week’s project update without taking it back” gives this session a more specific purpose.
Check that the goal belongs to the person being coached. A sponsor may have a concern about delivery, but that does not establish what the individual wants from the conversation. Make differing expectations visible rather than quietly treating one as the other.
Original questions you could adapt include: “What would make this conversation worth your time?” and “Which part of the situation would you most like to handle differently?” Ask one question, listen to the answer and clarify what it means before moving on.
Do not rush to make the goal precise at the expense of understanding it. If the person starts with a vague concern, that can be the opening of the exploration. Refine the focus as the conversation develops.
Reality: examine a specific situation
Move from general statements to events and conditions that can be discussed. “My team never takes ownership” is an interpretation. A recent handoff, the instructions given, the response and what happened afterwards provide a more useful starting point.
Ask about the person’s own actions as well as the surrounding conditions. What information did they provide? What did they expect? What did another person actually say, and what are they inferring from it? Avoid treating an account as complete merely because it is confidently told.
You might ask: “Which recent occasion best shows the problem?” or “What did you notice before deciding to step in?” These questions can reveal constraints, assumptions or exceptions that a broad judgement hides.
The purpose is not to cross-examine the person or prove them wrong. It is to build a sufficiently clear account to explore useful choices. Keep unknown motives and missing information labelled as unknown.
Options: widen the choices before choosing
Explore several plausible ways forward before evaluating one. Otherwise, a familiar suggestion can become the plan simply because it was the first thing mentioned.
Options can include changing the task, the timing, the support available or the way progress is checked. They can also include collecting information before making a larger commitment. Not every problem needs a more ambitious action plan.
Try asking: “What is another approach you could try?” or “Which part of this is within your influence?” If the person requests advice, clarify that shift. Offering an idea is different from helping them generate their own options, and it should not silently become an instruction.
Then compare the choices against the person’s goal and actual constraints. An attractive option that requires authority or resources they do not have may need a preliminary conversation rather than immediate action.
Will: turn a chosen option into an intended next step
Make the proposed action concrete enough to revisit. What will the person try, in which situation, and what support or clarification do they need? Agree how they want to review it.
A useful question is: “What is the smallest useful step you can take before we next meet?” Another is: “What might prevent that, and what would you do then?” These are invitations to check feasibility, not a demand to promise more.
Record the commitment accurately. “I might try this” is still an option. “I will try it in Tuesday’s handoff” is a stated intention. Neither sentence proves that the action later happened or that it worked.
If the person does not want to commit to an action yet, explore why. They may need information, a different goal or a more appropriate form of support. A realistic unresolved question is preferable to an artificial promise made to finish the model.
A worked GROW conversation
The following fictional exchange illustrates the structure. It does not describe a customer, a real coaching session or a tested product output.
A manager, Riley, says that preparing the weekly project update takes too much time because they keep rewriting a colleague’s draft. Riley wants to delegate the task but worries that a poor update will reach senior stakeholders.
Goal: identify the immediate focus
Riley’s first goal is “stop doing everyone’s work”. After discussion, the session focus becomes: “Decide how to hand over next week’s update with a clear review boundary.”
This does not settle whether the wider concern is accurate. It chooses one situation in which Riley can examine the pattern and try something different.
Reality: examine the last handoff
Riley describes the most recent update. They asked for “the usual report”, did not identify the intended reader and began rewriting before explaining which parts needed improvement. The colleague had not been told what would make a draft acceptable.
That account suggests a possible issue with the handoff, alongside Riley’s concern about quality. It does not establish that unclear instructions are the only cause or that the colleague lacks skill. The next conversation with the colleague may reveal other factors.
Options: explore more than simply stepping back
Riley considers three approaches: provide a short description of the reader and required content; review one early outline rather than a complete draft; or keep ownership this week while arranging a separate learning session.
They discuss the trade-offs. An early outline may reveal misunderstandings before much work is done. A clearer brief may reduce ambiguity. Keeping the work may meet an immediate deadline while leaving the delegation problem unchanged.
These are options to evaluate, not a universal ranking of the best delegation technique. The appropriate choice depends on the task, the colleague’s needs and the time available.
Will: agree one manageable attempt
Riley chooses to send a brief on Monday identifying the reader, the required information and the review point. They intend to review the outline on Tuesday before the colleague develops the full update.
The review question for the next session is specific: did Riley communicate the requirements and hold the agreed review boundary? A further question is whether that approach helped the colleague produce a usable draft. Those are related questions, but they need different evidence.

What to review at the next session
Begin with what the person says they tried, what happened and what they learned. Avoid starting with a verdict about whether they succeeded as a manager or whether coaching “worked”.
For Riley, first establish whether the brief and outline review took place. If they did, ask what the colleague understood, which parts of the draft still needed work and how Riley responded. If they did not, explore the obstacle instead of assuming a lack of motivation.
The result may justify adapting the goal or choosing another option. A single attempt can provide useful information without proving a general pattern. Keep the original intention and the later account distinct so the learning is visible.
Use saved conversations as evidence, with appropriate agreement
When a person has agreed to recording and review, earlier conversations can help you return to the goal in their own words. Weeve’s coaching-progress workflow supports reviewing stated goals, reported attempts and open questions across saved conversations.
Ask for the relevant passages and inspect their sources. A useful review request separates what the person committed to, what they later reported doing and what remains uncertain. Check attribution and the scope reviewed before drawing a conclusion.
The genuine source-reader image here uses approved synthetic pilot-planning content. It demonstrates inspecting an underlying passage; it is not a GROW session, an evaluation of Riley or evidence of improved coaching outcomes.
Keep records proportionate to the agreement. A short agreed action note may be enough. If a participant does not want a conversation recorded, use another agreed way to retain the next step. A framework does not create permission to capture or share a session.

Common ways the model becomes less useful
Treating the stages as a compulsory script
A person may discover the real issue while discussing the current situation. Returning to the goal can be more useful than insisting that the “Goals” stage has already been completed. The framework supports the conversation; it does not replace listening.
Asking questions that contain the preferred answer
“Wouldn’t a better handoff solve this?” sounds like a question but embeds an explanation and a solution. Ask what the person notices about the handoff, then explore alternatives. If you want to offer a hypothesis, label it as one and invite disagreement.
Choosing an action before examining the situation
An action can sound productive while addressing the wrong problem. In the fictional example, telling Riley simply to stop rewriting would miss the unclear expectations in the last handoff. Examine enough reality to understand what the chosen step is intended to change.
Confusing commitment with completion
A tidy action note records an intention. It does not establish that the action happened, met the goal or caused an improvement. At the next review, look for the person’s account and other appropriate evidence rather than recycling the original promise as proof.
When GROW needs something alongside it
A question-based conversation cannot supply missing technical knowledge, authority, resources or a required organisational decision. If the person needs a demonstration, specialist advice or a change to the work itself, identify that need rather than repeatedly asking them to generate an answer they cannot yet provide.
CIPD’s coaching and mentoring guidance treats coaching as one development approach among others, rather than a solution for every learning need. Consider whether a different or combined approach suits the situation.
For a useful first attempt, choose one real issue, keep the four areas in mind and listen for what the answers change. Finish with a realistic next step or a clearly identified unknown. Then review what happened before deciding what to do next.


