The GROW model for sales coaching, with script

A coaching conversation differs from a pipeline review on one point: who does the thinking. In a review, the manager thinks and the rep reports. In coaching, the rep thinks and the manager asks. GROW is simply a way of keeping the questions in order, so that the conversation ends in a decision rather than in a vague sense of having talked things through.

Where does GROW come from, and why does it work in sales?

GROW was developed in the UK in the late 1980s by Sir John Whitmore, working with Graham Alexander and Alan Fine, and published in 1992 in Whitmore's Coaching for Performance, a book his own firm says has sold more than a million copies in over twenty languages. Exactly who coined the acronym is contested. Whitmore wrote in a later edition that Max Landsberg came up with it in conversation with Alexander, while Fine describes himself as a co-creator. None of that matters much to anyone who wants to use the model. What does matter is the pair of aims Whitmore set for coaching, awareness and responsibility: the rep should see their own situation more clearly and take ownership of what happens next.

The data backs the idea that structure beats time. Gartner studied 5,000 managers and found no significant link between how employees performed and how much time their manager spent developing them. What set the best managers apart, the ones Gartner calls connectors, was that they coached on the things they knew well and connected the employee to other people for everything else, rather than being permanently on hand. The manager who dispenses advice constantly, 'always on' in Gartner's typology, did not make the team any better. A good fifteen-minute GROW conversation is not a compromise, then. It is the format that actually works.

G: the goal, as a behaviour and not a number

The first mistake gets made right here: 'The goal is for you to close three deals this quarter.' That is a quota, not a coaching goal. A goal you can actually coach is a behaviour the rep can change in their next call: asking for the meeting on the first call rather than the third, raising budget before the demo, holding the price at the first sign of pushback. Ask 'Which call do you want us to work on today?' and follow with 'What would it sound like when it goes the way you want?'. A goal you can hear in one call is a goal you can measure in the next.

R: reality, with the call as evidence

Most 1-on-1s go wrong here in the opposite direction: reality gets described in feelings. 'The customer was difficult.' 'It felt good.' GROW wants observations, and in sales those live in the call itself. What did the customer say, word for word, when you asked for the meeting? At what point did price come up? Did you leave with a booked next step or a 'let's keep in touch'? If you have a scorecard, or scores from recorded or practised calls, this is the moment to bring them in. A manager who asks 'what happened?' rather than 'why did it go badly?' gets answers that can be acted on. And a manager who has already listened to the call gets to ask the one question nobody else can: 'What were you thinking right before you said that?'

O: the options, including practice

Only now do ideas enter the conversation, and they need to come from the rep. 'What could you do differently next time?' Then wait out the silence. Ask for three options before you evaluate any of them, because the first suggestion is nearly always the safe one. The manager's own ideas go last, offered as one option among several rather than as the correct answer. One option belongs on every list: practising. Run the same call again in a roleplay before it gets run for real on the next customer. That is the difference between having a plan and having a skill.

W: the will, with a date

The final step is what separates coaching from a pleasant chat. 'What do you choose to do, and when?' The answer needs a behaviour, an occasion and a time: 'I ask for the meeting on the first call in every call on Thursday, and we listen to two of them on Friday.' Follow up with 'On a scale of one to ten, how sure are you that you will do it?' Anything below eight means the goal is too big or something is standing in the way, and in that case shrinking the step beats pretending.

The script: a GROW conversation in fifteen minutes

1. Minutes 0 to 2, the frame. 'We have fifteen minutes. Which call or situation do you want to work on today?' Let the rep choose. Whatever the manager picks turns into an evaluation. Whatever the rep picks turns into coaching.
2. Minutes 2 to 4, the goal. 'What would it sound like when it goes the way you want? What does the customer say then?' Write the goal down as a single sentence describing a behaviour.
3. Minutes 4 to 8, reality. 'What happened in the last call, step by step? What did the customer say, word for word? What next step did you get?' Bring in scores, notes or a recording if you have them. No why-questions yet.
4. Minutes 8 to 12, options. 'What could you do differently? What else? And if you could choose freely?' Three options before any evaluation. Add 'practise it in a roleplay' if it is not already on the list.
5. Minutes 12 to 14, will. 'What do you choose? When do you do it for the first time? How sure are you, one to ten?' Under eight, shrink the step.
6. Minutes 14 to 15, the agreement. Play back what the rep said, in the rep's own words, book the follow-up and finish. No closing summary of what you think.

Three mistakes that turn GROW into an interrogation

The first is the leading question. 'Don't you think you should have asked about budget earlier?' is advice with a question mark on the end, and the rep hears it as such. The second is jumping to options before reality has been established, the classic move of managers who were good sellers themselves: they spot the problem inside twenty seconds and solve it. What the rep learns is that the manager solves things, not how to solve them. The third is ending without W. A conversation that closes on 'good, let's keep thinking' has produced precisely zero change in behaviour, however good it felt.

A word on time. In 2015 the Sales Management Association estimated that only around eight percent of a sales manager's working hours go to actual coaching, roughly 36 minutes per rep per week, while the managers in Highspot's 2025 survey put their own figure at 13 hours a week. That gap is the whole point: most of what gets called coaching is something else entirely. Fifteen structured minutes per rep per week is more than most reps get. Korn Ferry reported in May 2026 that organisations with strong coaching processes see 14 percent higher quota attainment, 15 percent higher win rates and nearly 20 percent lower voluntary turnover, and in Salesforce's State of Sales 2026, 75 percent of reps say they are more likely to hit target with a coach or mentor. Sweden starts from a better position than many markets, too. Gallup found in 2025 that 25 percent of Swedish employees are engaged at work, against a European average of 12 percent, and Gallup estimates that the manager accounts for at least 70 percent of the variance in engagement. A good fifteen minutes a week, then, is one of the biggest levers a sales manager holds.

GROW after a roleplay

The model works best when reality is already sitting on the table. After a practised call in Salesprep, the rep has seven scores, each with a written comment, and together they are the R step already done: 'Your objection handling got 5.8 and the comment says you went defensive when the customer said they already have a vendor. What happened there?' From there it is options, then will, then another roleplay on the same scenario before the rep tries it on a real customer. The Team plans put every rep's scores and trends in front of the manager in an admin view, so the fifteen minutes can start from data rather than from memory. Three calls are free when you create an account, which is enough to hold a first GROW conversation on real material as early as next week.