04 · Advanced Coaching Techniques (the GROW Model)

Level 2's coaching fundamentals taught you to ask more and tell less. This module gives that instinct a repeatable structure: GROW (Goal, Reality, Options, Will), the most widely used coaching model in professional practice, adapted for a working manager who has fifteen minutes, not a certified-coach hour. The value of a structure isn't ritual — it's that without one, coaching conversations drift into either advice-giving (you solve it for them) or vague sympathy (you validate their feelings and nothing changes). GROW forces the conversation through four stages in order, and skipping a stage is the single most common way a coaching conversation quietly turns back into you giving instructions.

1. The GROW Worksheet

GROW COACHING WORKSHEET

GOAL — what does the person actually want from this conversation,
in their words, stated as an outcome not a topic?
  Topic (too vague): "My workload"
  Goal (specific): "Decide whether to push back on the new project"
  ______________________________________________

REALITY — what's actually true right now, gathered through
questions, not asserted by you?
  What have they already tried?
  What's working, even partially?
  What are they assuming that might not be true?
  ______________________________________________

OPTIONS — generated by them, not you. Resist filling silence.
  What are ALL the options, including ones they'd reject? (list ≥3)
  What would they do if resources weren't a constraint?
  What would they advise a friend in this exact spot?
  ______________________________________________

WILL — converts one option into a commitment.
  Which option, specifically?
  First concrete step, and by when?
  How will they know it worked?
  What could get in the way, and what will they do about it?
  ______________________________________________

The Reality and Options stages are where most managers cheat: they rush through Reality because they think they already know it (you usually know your version of it, not theirs), and they shortcut Options by suggesting the first plausible one instead of making the person generate three.

2. Why the order matters

Each stage disciplines the one after it. Skip Goal and you coach the wrong problem entirely — often the presenting complaint isn't the actual thing they want resolved. Skip Reality and Options gets generated against a fictional situation. Skip Options and Will becomes a decision you made that they merely agreed to, which evaporates the moment it gets hard.

3. Worked example: a full GROW conversation

Tomas manages Bea, a senior analyst who opened with "I think I need to talk about my workload," which is a topic, not a goal.

GOAL

Tomas: Before we get into it — what do you want to be true by the end of this conversation? Not the topic, the outcome.

Bea: I guess... I want to know if it's reasonable to ask to drop one of my three projects, or if I just need to manage my time better.

Tomas: Good, that's a real goal — a decision, not a feeling. Let's work toward that.

REALITY

Tomas: Walk me through the three projects. What's actually on your plate on each one this week?

Bea: Reconciliation automation, the vendor onboarding, and covering for Sam's project while he's out.

Tomas: Which of those did you actually choose to take on versus have handed to you?

Bea: Sam's coverage was handed to me two weeks ago when he went on leave. I didn't say no because it felt like a two-week thing.

Tomas: And is it still a two-week thing?

Bea: ...No, actually. He's out for two more months, I just found out yesterday. I hadn't put that together with everything else.

OPTIONS

Tomas: Given that it's now a two-month commitment, not two weeks — what are your options? Give me at least three, including ones you'd reject.

Bea: One, I keep all three and just work more hours. Two, I hand Sam's coverage to someone else. Three, I pause vendor onboarding for two months since it's the least time-sensitive. Four — I guess — I go back to whoever assigned Sam's coverage and say the scope changed.

Tomas: What stops you from doing the fourth one?

Bea: It feels like admitting I can't handle it.

Tomas: Is that true, or is that a two-week commitment quietly becoming an eight-week one without anyone re-deciding it?

Bea: ...The second one. Nobody re-decided it.

WILL

Tomas: So which option, specifically?

Bea: I'll go back to whoever assigned Sam's coverage — that's actually you — and ask to either extend it formally with something dropped, or hand it off, now that it's two months not two.

Tomas: Good. Let's do that right now instead of scheduling another meeting — what would you drop, vendor onboarding or Sam's coverage?

Bea: Vendor onboarding pauses. It has no deadline this quarter.

Tomas: Done. I'll tell the vendor lead it's paused with my name on it, not yours. Anything that could stop this from sticking?

Bea: Old habit of just absorbing it quietly again next time.

Tomas: Then the actual commitment is smaller than "manage your workload" — it's "the next time a temporary ask changes scope, you flag it to me within a week instead of two months in." Deal?

Bea: Deal.

Notice Tomas never told Bea what to drop. He asked the question that surfaced the real fact (two weeks became two months, unnoticed), and once she saw it, the decision was close to obvious to her — which is why it will stick.

4. Coaching questions that keep you honest to the model

• Goal: "What would make this conversation worth having?"
• Reality: "What's the part of this you haven't said yet?"
• Options: "What would you tell a friend in exactly this position?"
• Will: "On a scale of one to ten, how likely are you to actually do this?" — if the answer is below 8, the option or the step is wrong, not the person's willpower; go back to Options.

5. When GROW isn't the right tool

GROW assumes the person has agency over the outcome. If someone is bringing you a problem that is genuinely outside their control — a layoff decision made above both of you, a policy they can't change — running GROW on it will feel hollow and manipulative, because you're pretending they have options they don't. Name that plainly instead: "This one isn't yours to solve, and I don't want to pretend otherwise — what I can do is X" is more respectful than forcing a coaching frame onto a situation that calls for direct information or support.

How It Actually Works

At the advanced level, GROW's ordering constraint is best understood through a decision-theory lens: exploring Options before Reality is fully specified is optimizing over the wrong search space. Reality-gathering is, formally, reducing uncertainty about the actual state of the problem before committing computational (or conversational) effort to searching for solutions — the coaching equivalent of specification before design. Skip it, and Options-generation searches a space defined by the coach or coachee's assumed version of Reality, which routinely diverges from the actual constraints in ways that only surface once a plan built on the wrong assumptions fails to execute — at which point the cost of having skipped Reality is paid in wasted implementation effort rather than the much cheaper cost of a few extra minutes of questions upfront.

Why "coaching questions that keep you honest to the model" function as a bias-correction mechanism, not a script. A coach under time pressure or personal investment in a particular outcome is vulnerable to a specific, well-documented failure: unconsciously steering questions toward a predetermined answer (leading questions), which collapses the self-generation mechanism from the Level 2 coaching module back into disguised advice-giving — the coachee still "arrives" at an answer, but it's the coach's answer wearing the coachee's words, so it doesn't get the retention or ownership benefit of a genuinely self-generated one. The honesty-check questions exist because coaches, like anyone, are subject to confirmation bias and can't reliably self-detect it from inside a conversation they're actively invested in.

Why GROW isn't the right tool for some situations, and recognizing that is itself part of the skill. GROW assumes the coachee has the requisite capability and information to generate a viable answer if guided well — it is a retrieval-and-construction process, not a knowledge-transfer one. When the actual gap is a missing fact or skill the coachee has no way to derive (a technical constraint they've never encountered, a policy they don't know exists), no amount of well-sequenced questioning will produce it, because there's nothing there to retrieve — this is the situation where direct instruction is the mechanistically correct choice, and insisting on coaching regardless wastes time searching for an answer that isn't present in the coachee's existing knowledge.

Exercise

Pick a real, current issue a direct report has raised with you and run the full GROW worksheet in section 1 as an actual conversation this week, in order, resisting the urge to skip to Options. Afterward, write down the moment (if any) where you were tempted to just tell them the answer, and what question you asked instead.