Sales Coaching Framework: Choose a Model, Then Run It Every Week

A rep can get three kinds of coaching in one week: a manager asking GROW questions about a lost deal, a pipeline review that drifts into call advice, and a quick note after a call to "ask better questions." Each conversation makes sense on its own. Together, they rarely change what the rep does on the next call, and nobody can say whether the coaching worked.

A sales coaching framework fixes that by deciding three things for every manager on the team: which kind of coaching conversation is happening, which model runs it, and what evidence shows the behavior changed. This guide compares four models that fit sales, gives you five questions for choosing between them, and lays out a weekly loop that ends with a check on a real buyer call.

TL;DR

• A sales coaching framework decides what kind of coaching conversation is happening, which model runs it, and what evidence shows change.
• Keep skill coaching, deal coaching, and pipeline reviews in separate time slots. The urgent deal usually crowds out the skill work.
• Use GROW for skill conversations, Did-Doing-Do for weekly deal check-ins, and AOR with a fixed rhythm for pipeline and activity reviews.
• Run skill coaching as a four-step loop: pick a call moment, name one behavior and its pass condition, rehearse it, and check a named later call.
• Roll it out by agreeing on definitions and calibrating managers on shared calls before anyone scores reps.

What a sales coaching framework decides

A sales coaching framework is a shared answer to three questions every manager on the team uses the same way:

1. What kind of coaching conversation is this? Building a skill, moving one deal, or reviewing the pipeline.
2. How does the conversation run? The model the manager follows, such as GROW or Did-Doing-Do.
3. What evidence shows the coaching worked? Usually, the same behavior showing up on a later buyer call.

A framework is different from a template. A template is the page a manager fills in for one session. The framework decides which template to use, how often, and what counts as progress. If you need the session page itself, the sales coaching template has a 30-minute 1:1 agenda, a call coaching form, and a four-week plan. This guide covers the decisions that sit above it.

Most popular coaching models were written for coaching in general, not for sales. Jason Jordan of Vantage Point Performance made this point in a Sales Management Association article: general models are not sales-specific, and they were designed for periodic planning sessions, not the steady rhythm of a sales week. So the first decision is the type of conversation, not the acronym.

Separate skill coaching from deal coaching

Skill coaching, deal coaching, and pipeline reviews answer different questions. When they share one meeting, the deal with a close date this month usually wins, and the skill work moves to "next week."

Deal coaching can surface a skill gap. A rep who keeps losing deals at the pricing conversation probably has one. When that happens, take it to the skill session. Deal coaching answers "what do we do in this deal?" Skill coaching answers "what will the rep do differently in every deal like it?"

Our recommendation: give skill coaching its own time on the calendar, even if it is only 20 minutes a week, and keep deal and pipeline work out of it.

Four coaching models and where each fits in sales

You do not need seven acronyms. Four models cover the conversations above. Pick one per conversation type and use it the same way across managers.

GROW: for skill conversations where the rep owns the answer

GROW stands for Goal, Reality, Options, and Will (or Way Forward). Performance Consultants credits its founder, Sir John Whitmore, and his peers with developing the model, which became widely known through Whitmore's book Coaching for Performance. The manager asks questions instead of giving the answer: what the rep wants from the call, what happened, what they could try, and what they will do.

In sales, GROW works well when the rep can already see the problem. "I keep presenting before I understand the problem" is a good GROW conversation. It works less well when the rep does not yet know what good looks like. In that case, show an example first, then use GROW to plan the practice.

Did-Doing-Do: for weekly deal and activity check-ins

RAIN Group's Did-Doing-Do framework runs a check-in in three parts: review what the rep committed to last time (Did), talk through current priorities such as opportunities and prospecting (Doing), and agree written commitments for the next period (Do).

It fits deal coaching and short weekly check-ins because every session starts from last week's commitments. Its limit is skill work: a commitment to "ask more about impact" can be written down and still not happen on the call. Pair it with a skill model when the commitment is a behavior.

AOR and a management rhythm: for pipeline and activity reviews

Jordan's model in the same Sales Management Association article links sales activities to sales objectives and business results (AOR), then puts structured conversations on a regular management rhythm. Start from the result you need, work back to the activities that drive it, and review those activities on a fixed schedule.

This fits pipeline reviews and activity coaching. It tells a manager which activities to watch and when to talk about them. It says less about how a rep improves a single conversation, so it needs a skill loop beside it.

Evidence, practice, check: for changing one behavior

This is the loop we recommend for skill coaching, and the one the rest of this guide uses. The manager starts from a real call moment, names one behavior with a pass condition, has the rep rehearse it before the next buyer conversation, and checks a named later call for the same behavior. GROW can run the conversation in the middle. What this loop adds is the evidence at both ends.

Choose a framework with five questions

Answer these questions once for the team, then write the answers on one page that every manager uses.

The one-page answer should name the conversation types, the model for each, the weekly time each gets, and the evidence that counts as progress. Keep it short enough that a new manager can read it before their first 1:1.

The weekly coaching loop

The loop has four steps. It is small enough to run with each rep every week, as long as each loop stays on one behavior.

1. Pick the moment from evidence

Start from something both people can inspect: a short stretch of a recorded call, or a practice attempt. Choose the moment tied to the stage result the rep needs, such as getting a real answer on business impact or agreeing on a next step. "The rep rushed the pricing part" is an opinion. "The rep gave the price before asking how many teams would use it" is evidence.

2. Name one behavior and its pass condition

Write the behavior so a second manager could judge it from the transcript. "Confirm who else signs off before proposing a next step" is a behavior. "Be more consultative" is not. Then write the pass condition: what the rep has to say or do for the attempt to count. For help writing criteria that produce evidence, use the sales call scorecard guide.

3. Rehearse before the next buyer conversation

Feedback without practice leaves the first real attempt to a live buyer. Have the rep retry the same moment, then a harder version of it: keep the target behavior fixed and change how the buyer responds. You can do this in the 1:1 with the manager playing the buyer. An AI sales simulation gives the rep a private retry against a buyer built from your playbook and real calls, scored against the scenario's own scorecard. If the moment came from a recording, turning that call into a roleplay scenario keeps the practice close to what happened.

4. Check a named later call

Before the session ends, agree on which upcoming call the manager will review for the same behavior. That later call is the test. A better practice score is useful, but the coaching worked only if the behavior shows up with a real buyer. With real-call grading, each recorded call is graded against the scorecard your team built, and every judgment points to the transcript line behind it. The manager opens that moment and skips the rewatch.

If the behavior shows up, pick the next one. If it does not, run the loop again with a harder practice scene, or check whether the pass condition was clear.

Roll the framework out across managers

A framework only helps if every manager runs it the same way. Roll it out in this order:

1. Agree on definitions first. Write down the behaviors, pass conditions, and scorecard criteria before managers start coaching to them.
2. Calibrate on shared calls. Have managers score the same two or three calls on their own, then talk through the evidence behind each score before comparing totals. The calibration steps in the scorecard guide walk through it.
3. Start with one conversation type. Most teams should start with skill coaching, because it has the clearest evidence. Add deal and pipeline models after the skill loop runs every week.
4. Decide who owns what. Enablement can own the definitions, scenarios, and scorecards. Managers own the weekly conversation and the later-call check.
5. Review after four weeks. Look at which behaviors moved on live calls, which loops stalled, and which pass conditions managers read differently. Update the one-page framework itself, then tell managers what changed.

For a team-wide practice habit to sit underneath the framework, the sales practice culture plan covers leader participation, private first attempts, and a short weekly rhythm.

What sales and enablement leaders told us

How we know this: this section draws on nine of our 2026 conversations with seven people: two enablement leaders, a former second-line sales leader, a former first-line sales manager, a fractional sales consultant, a former head of sales training who now advises and coaches, and a former sales engineering leader. Names and company details are removed.

• Deals and forecasts take the coaching time. A senior enablement leader said managers who were promoted from top rep often go back to working deals themselves, and that forecast meetings take much of the time that is left. Her version of deal coaching was a set of questions about the rep's own plan, such as "how do you know that deal is going to sign by this date?", followed by a request to see the plan.
• A month has fewer coaching weeks than it looks. A fractional sales consultant said the last week of the month goes to closing and the first week to resetting, so managers "really have two weeks in four where they can effectively coach their team."
• Call review is smaller than the plan assumes. The same enablement leader said managers can't listen to dozens of recorded calls a week, and that getting them to review one call per rep each week is already a lot. She also argued that managers should run the same roleplay as their reps, plus a separate roleplay that practices the coaching conversation.
• Who picks the focus skill is contested. A former second-line sales leader said the skill should come from what he heard in call and deal reviews (in his case, reps giving ground too quickly in negotiations) and go to enablement as a request. Corporate initiatives had taken that list away from him. An enablement leader at a large cloud company described a middle ground: enablement builds a shared baseline learning path that senior leadership signs off on, and each manager adjusts it, for example more prospecting for SDR teams.
• Weekly coaching is not universal. A former sales engineering leader said managers on his teams rarely reviewed recordings unless someone raised a problem, so coaching was mostly reactive.
• Test whether the fix held. A former head of sales training described coaching a rep through a price objection, then bringing the same objection back a few minutes later without warning to check that the answer stuck.
• Effort and skill are different problems. A former first-line sales manager said a rep who has the tools but does not practice has a motivation problem, while a rep who practices and still does not improve is a "very different problem."

How we apply this to the framework

• Schedule the skill slot in the weeks managers can coach, and keep it short in the closing week.
• Plan for one reviewed call per rep per week, and make that call the named later check.
• Let managers propose the focus behavior from their own call reviews, inside the shared definitions.
• In practice, bring the same buyer pushback back later in the session without warning.
• Before coaching a skill, check whether the rep practiced. Missing practice and practice without progress need different conversations.

Frequently asked questions

What is a sales coaching framework?

It is the team's shared agreement on how managers coach: which coaching conversations happen (skill, deal, pipeline), which model runs each one, how often, and what evidence counts as progress. A template is the page a manager fills in for one session; the framework decides how those sessions fit together.

What is the best sales coaching model?

There is no single best model, because sales managers run different kinds of coaching conversations. GROW suits skill conversations where the rep can see the gap. Did-Doing-Do suits weekly deal check-ins. AOR with a fixed review rhythm suits pipeline and activity reviews. For changing one call behavior, wrap the conversation in a loop that starts and ends with call evidence.

How is a coaching framework different from a sales methodology?

A sales methodology, such as MEDDIC or Challenger, describes how reps should run deals and conversations. A coaching framework describes how managers help reps do that. The methodology supplies many of the behaviors and scorecard criteria; the framework decides how managers coach and check them.

How do you know the framework is working across the team?

Track how often the agreed behavior shows up on the named later call, how many loops stall without a check, and whether managers score the same shared calls the same way. If behaviors improve in practice but not on live calls, the practice scenes are too easy or too far from real buyers.

Can AI replace the manager in sales coaching?

AI practice can take on the repetitions: a rep can retry a moment against an AI buyer and get a score each time. The manager still chooses the behavior, judges the deal context, and checks the live call. Live in-call prompts are a separate tool with their own tradeoffs, covered in our guide to real-time AI sales coaching.

Start with one conversation type

If you change one thing this month, give skill coaching its own weekly slot and run the four-step loop on one behavior per rep. Add the deal and pipeline models once that loop runs every week without reminders.

When the agreed behavior needs rehearsal before the next buyer call, book a free practice call and have a rep try it against a realistic AI buyer.