Most sales coaching tips fail for one reason. They describe what good coaching looks like without saying what a manager should do on a Tuesday afternoon with four calls to review and eleven minutes to spare.
These twelve tips are written the other way around. Each one has a drill attached, so a manager can run it this week without redesigning their enablement program first. They are grouped into three parts: coaching the conversation, making practice routine, and measuring what actually improves.
The thread running through all twelve is a loop. Reps practice before the call, get guidance during it, and review what happened after. Break any stage of that loop and the tips stop compounding, which is the real reason coaching advice so often fades within a month.
Coach the conversation, not the CRM
The fastest way to waste a coaching session is to spend it on pipeline hygiene. Deal reviews inspect outcomes that already happened. Coaching changes the conversation that produces the outcome, which means the subject has to be what the rep said and how the buyer responded, not what stage the opportunity sits in.
1. Coach one moment, not the whole call
Pick a single sixty second stretch and work only on that.
A rep cannot fix nine things at once. When a manager reviews an entire call, the feedback arrives as a list, and lists get filed rather than practiced. Isolating one moment gives the rep something they can carry into the next conversation.
2. Ask before you tell
Open with what the rep noticed, not what you noticed.
Reps who diagnose their own calls improve faster than reps who receive verdicts, because self-diagnosis is the skill they will need on every unobserved call for the rest of their career. Telling produces compliance, asking produces judgement.
3. Separate skill problems from will problems
If a rep can do it in practice but not on a call, it is a confidence problem, not a knowledge problem.
These need opposite responses. A knowledge gap is closed with information. A confidence gap is closed with repetition until the response is automatic. Managers who treat the second as the first end up re-explaining something the rep already knows.
4. Coach the question, not the pitch
Weak discovery causes more lost deals than weak presenting.
According to Gartner research, B2B buyers spend only 17 percent of their total purchase consideration time meeting with potential suppliers, and when several vendors are in play any single rep may get five or six percent of that time. There is no room to recover a shallow discovery call later in the cycle.
Make practice routine
The reason most coaching advice does not stick is that it depends on manager availability. A manager with eight reps cannot attend enough calls to build skill through observation alone, so practice has to happen without them in the room. That is a system design problem rather than a discipline problem.
5. Put the reps in before the stakes are real
The most expensive place to attempt something for the first time is a live call with a real buyer.
Reps are routinely asked to try new messaging, a new objection response, or a new product story for the first time on pipeline that counts. Rehearsing first costs nothing. Getting it wrong on a live deal costs the deal. This is where AI roleplay simulations earn their place: unlimited attempts, no prospect at risk.
6. Space the repetitions instead of massing them
Four ten minute sessions across a week beat one forty minute session.
The reason is well established. Hermann Ebbinghaus documented the forgetting curve in the 1880s, showing that people lose the majority of newly learned information within days unless it is revisited at intervals. Sales enablement calendars are still built as if that research does not exist.
7. Practice the specific buyer, not a generic one
A drill against "a difficult prospect" builds nothing. A drill against your actual buyer builds a reflex.
Reps need the objections their market really raises, in the language their buyers really use. The closer the practice conversation is to the real one, the more of the skill transfers.
8. Give reps a way to practice without you
If coaching only happens when a manager is present, coaching capacity is capped at manager hours.
This is the arithmetic that breaks most coaching programs. A team of ten reps making twenty calls a week produces two hundred conversations. No manager reviews two hundred calls. Reps need somewhere to get their reps in on their own schedule, with feedback that arrives without a person attached.
9. Extend practice onto live calls
Practice prepares the rep. Guidance in the moment is what changes the call in front of them.
Rehearsal handles the conversations you predicted. Real calls produce the ones you did not. Guidance that surfaces while the buyer is still talking, which is what live call coaching does, covers the gap between what a rep practiced and what actually happened.
Measure what improves
Coaching that is not measured drifts back to whatever the manager finds most comfortable to talk about. The measurement does not need to be elaborate, but it does need to be behavioral, because quota attainment is affected by too many things outside the rep's control to be a useful coaching signal on its own.
10. Score behavior, not outcomes
Win rate tells you what happened. Discovery depth tells you why.
Behavioral metrics move weeks before revenue does, which makes them the only ones a manager can actually coach against in-quarter. Pick three, hold them steady for a quarter, and resist adding more.
11. Make progress visible to the rep
Reps who can see their own trend line practice more without being chased.
Gallup's research on management has consistently found that managers account for around 70 percent of the variance in team engagement. A visible scoreboard does part of that work on the manager's behalf, because improvement becomes self-evident rather than something the rep has to be told about.
12. Coach the pattern, not the person
When six reps miss the same thing, it is a messaging problem, not six coaching problems.
Team-level data separates individual gaps from systemic ones. If most of the team concedes the moment price comes up, no amount of one-on-one coaching will fix it, because the problem is the value story rather than the sellers. Team analytics exist to tell those two situations apart.
TL;DR
- Coach one sixty second moment per session, not the whole call
- Ask the rep to diagnose before you offer a verdict
- Separate skill gaps from confidence gaps, they need opposite fixes
- Coach discovery questions before presentation skills
- Never let a rep attempt something for the first time on live pipeline
- Space repetitions across the week instead of massing them
- Practice against your real buyers and their real language
- Give reps a way to practice that does not require a manager present
- Extend practice with guidance during live calls
- Score behavior, because it moves before revenue does
- Show reps their own trend line
- Check whether a gap is individual or systemic before coaching it
Where to start next week
Pick tips one, five, and ten. One session, one moment, one drill run before the stakes are real, and one behavioral number written down. That is a coaching loop, and it is small enough to survive a busy week.
The broader argument behind why single training events fail is covered in why sales training does not stick, and the mechanics of building the loop are in sales training reinforcement. For the version that runs during live calls rather than after them, see how real-time coaching fixes the feedback delay.
Frequently asked questions
Short and frequent beats long and occasional. A weekly fifteen minute session on one specific skill outperforms a monthly hour covering everything, because skill is built through spaced repetition rather than single exposure. The cadence matters more than the calendar time.
Training transfers knowledge to a group. Coaching develops skill in an individual, based on what that person actually did on a call. Training is an event, coaching is a habit, and teams that run one without the other usually see behavior drift back within a quarter.
Start from evidence rather than opinion. When the conversation opens with a specific moment in a call, and the rep has already seen their own score, the discussion is about the moment instead of about them. Practice environments help here because the first correction happens without a real buyer present.
Track skill scores over time, not just quota attainment. Quota is a lagging indicator affected by territory and market. Discovery question count, objection recovery rate, and talk ratio move first and predict the revenue change that follows.