Command of the Message Practice Drills 2026
Chapters
What does Command of the Message practice actually certify?
It certifies message discipline under pressure: whether a rep can name the buyer's business problem, state the required capabilities to solve it, attach the buyer's own metrics, and hold a differentiation claim when challenged. It does not certify that the rep earned that information honestly. Those are two separate things, and teams conflate them constantly.
The framework's value is consistency. Ten reps facing the same buyer should describe the same problem in roughly the same language and reach for the same required capabilities before naming a product. That consistency is gradeable. A manager reading a transcript can point to the sentence where the rep named a capability, and the sentence where the rep skipped it and led with a feature.
What is not gradeable is a message the rep never said out loud in practice. Slide reviews and knowledge checks certify recall. A scored session certifies behavior, because the transcript shows the order the rep chose under time pressure with a buyer interrupting.
A message is certified only when the buyer can restate it, not when the rep can recite it.
Set the certification target before you run anything: the rep passes when a second manager, reading only the transcript, would mark the same four rows the same way. If your two managers disagree on what a required capability sounds like, fix the rubric language first.
TL;DR
Command of the Message practice becomes coachable when you convert the framework into four rubric rows a manager can grade from a transcript alone: the business problem stated in the buyer's words, required capabilities named before any product noun, metrics tied to numbers the buyer said, and differentiation the buyer can repeat back. Run it as two drills - a solo capability statement first, then a buyer who pushes back on differentiation. Score the buyer's restatement, not the rep's confidence.
- Grade four rows: problem in buyer's words, required capabilities before product, buyer's own metrics, repeatable differentiation.
- Drill one is solo and time-boxed; drill two adds an AI or peer buyer who challenges the claim.
- The pass bar for differentiation is the buyer restating it in their own language, unprompted by a script.
- The model disciplines message consistency; it does not certify discovery questioning or qualification hygiene.
- Failed attempts get one behavior in the debrief and a scheduled re-run, not a longer lecture.
Which rubric rows make the message scoreable?
Four rows are enough. Each one names an observable behavior with a location in the call, not an impression of the rep. Keep the wording concrete enough that a manager grades it without asking the rep what they meant.
Load these rows into whatever you score with. On XL Roleplay, your call stages and rubric rows are loaded before the session, so the coaching note cites your own capability language and the flags jump to the exact transcript moment where the rep named a product before a capability. Our position is that roleplay scoring changes behavior only when it is graded against your organization's own methodology and call stages, and that generic rubrics produce feedback reps rationally discount.
Rubric rows must name behavior a second manager could grade from the same transcript without asking the rep what they meant.
Score the rows separately and read trends per rep per scenario. One blended message score tells you a rep landed somewhere in the middle and nothing about which row broke. Reps who fail on metrics need a different drill than reps who fail on differentiation.
| Rubric row | Pass bar | Failing attempt sounds like |
|---|---|---|
| Business problem in the buyer's words | Rep restates the problem using at least one phrase the buyer used earlier in the session | "Most companies your size struggle with efficiency." |
| Required capabilities before product | Rep names two capabilities as outcomes the buyer needs, with no product or feature noun before them | "Our platform has automated routing and dashboards." |
| Metrics tied to the buyer's numbers | Rep uses a figure the buyer stated and names who owns it | "You'll see significant time savings across the team." |
| Differentiation the buyer repeats back | Buyer restates the differentiator in their own words when asked how they would explain it internally | "Honestly, we're similar to them - it comes down to price." |
Drill 1: the solo capability statement
Run this one first and run it solo. No buyer, no interruptions, no rescue. The rep gets a real deal from the pipeline and one prompt: name the business problem, then two required capabilities, then how you deliver them, then the metric and its owner. We recommend a 60-second time box and three consecutive attempts in a 15-minute block.
The order is the whole drill. Required capabilities come before product language, because a rep who leads with the product has skipped the step where the buyer agrees what is needed. Score the sequence from the transcript: mark the first product noun, then check whether two capabilities appeared before it.
Pass bar: two required capabilities stated as buyer outcomes before any product noun, plus one metric the buyer actually said, inside the time box. Anything else is a fail and gets re-run in the same session.
A capability statement that names a product feature before a required capability has failed the row, regardless of delivery.
A failing attempt sounds smooth. "We help revenue teams get visibility into their pipeline with a single source of truth, and our customers see faster cycles." No named problem, no capability, no buyer number, no owner. Reps often score themselves well on that sentence, which is exactly why the row is graded from the transcript rather than from the debrief conversation.
Drill 2: the buyer who pushes back on differentiation
Now add resistance. The buyer persona has already seen a competitor and says so: "They told us the same thing." "What is actually different here?" "Our team thinks these tools are interchangeable." The rep has to hold one differentiation claim, tie it to a consequence the buyer named, and survive a second challenge without adding three more claims.
The pass bar sits with the buyer, not the rep. At the end of the session, the rep asks: "If you were explaining the difference to your CFO, how would you say it?" The buyer's answer is the score. If the buyer restates the differentiator in their own words, the row passes. If the buyer produces a feature list or shrugs, the row fails no matter how articulate the rep was.
Some managers argue that agreed first responses make reps sound canned, and that strong reps improvise better. We disagree on sequence, not on skill. Drill the agreed first response until it is automatic, then let the rep improvise the branches from understanding - improvisation is safe once the opening is not being invented mid-call.
Differentiation survives only in the buyer's language, and pushback is the only test that reveals which version the buyer kept.
A failing attempt sounds like escalation: three differentiators in one breath, a comparison the rep cannot substantiate, or a quiet concession into price. Flag the moment, debrief one behavior, schedule the re-run.
The before-and-after scenario drill
The before-and-after scenario is where the message stops being a claim and becomes arithmetic. The rep states the buyer's current state with the buyer's own figures, names the required capabilities that close the distance, describes the after state in the same units, and names the person who owns that number. Run it as a five-minute segment inside your weekly block, using a live deal.
Score three things from the transcript. First, are the before-state numbers ones the buyer stated, or ones the rep assumed. Second, do the capabilities in the middle match the two the rep used in drill one, or has the message drifted. Third, is the after state expressed in a unit the buyer's finance team would recognize - hours, headcount, days of cycle time, error rate - rather than an adjective.
A before-and-after scenario built on the rep's assumptions instead of the buyer's numbers is a brochure with two columns.
Where the numbers are missing, check whether the gap sits upstream in discovery rather than in the message. That is a useful diagnosis, and it is the reason we keep the two drills separate: this drill tells you the rep cannot quantify the gap, and a discovery drill tells you why.
What is Command of the Message weak at?
It is weak at everything before the rep speaks. The framework disciplines what gets said and in what order. It does not teach the questioning that surfaces the problem, and it does not qualify the deal. A rep can deliver a flawless capability statement inside a deal with no economic buyer, no decision process, and a problem nobody has agreed is worth solving.
Name the gaps and cover them with drills you already run. For questioning sequence - building the cost of inaction before any solution is named - use SPIN practice drills, where implication questions are the scored row. For qualification hygiene, use MEDDPICC exit criteria drills: metrics, economic buyer, decision criteria, decision process, identify pain, champion, paper process, competition, graded as buyer-verified evidence rather than rep opinion. Keep those rubrics distinct from your message rows so a rep's score tells you which skill is missing.
Command of the Message disciplines what a rep says; it does not certify the questions that earned the right to say it.
The model is also silent on next-step commitment. Add a close row from your own call stages, or the rep will finish a strong value message with a vague follow-up and still pass every message row you graded.
How to run and debrief this week
Use a 30-minute weekly block: 5 minutes setup, 15 minutes drilling, 10 minutes debrief. Add roughly 10 minutes of your own prep beforehand to pick the scenario from a live deal and choose the one rubric row you are grading. Do not attempt all four rows in one session.
Debrief one behavior. Pick a single flagged moment - the sentence where a product noun arrived before a capability - play it, and ask the rep to diagnose it before you render a verdict. Then schedule the re-run. Feedback that covers three or more issues at once is entertainment; our full sequence is in the roleplay debrief guide.
Keep this block separate from the pipeline review. Deals are urgent and skills are merely important, so message drills lose every calendar collision unless the block is protected and recurring. A 15-minute 1:1 built on one flagged moment belongs in addition to the pipeline 1:1, never instead of it.
Read scores per rep per scenario, because a blended message score hides which row is actually breaking.
Failed rows get re-run, not re-explained. A rep who fails the differentiation row twice needs the same buyer persona a third time, not a new deck.
Frequently asked questions
Can Command of the Message be scored without a full rollout of the methodology?
Yes. The four rows - problem, required capabilities, metrics, differentiation - are gradeable from a transcript even if your team uses different labels internally. Use your own capability language in the rubric wording so reps recognize it.
How many required capabilities should a rep name in a drill?
We recommend two. One reads as a slogan and four turns into a feature list the buyer cannot repeat back. Two forces the rep to choose the capabilities that matter for the named problem.
Who plays the buyer in the differentiation drill?
A manager, a peer, or an AI persona all work, provided the buyer challenges the claim twice and then answers the restatement question honestly. The restatement is the score, so a buyer who helps the rep out invalidates the drill.
What if the rep has no buyer numbers to use in the before-and-after scenario?
That is a discovery failure, not a message failure. Score the row as a fail, then send the rep to a discovery drill where quantifying the current state is the graded behavior, and re-run the message drill after.
How do we keep managers grading the differentiation row consistently?
Calibrate on transcripts. Have two managers score the same recorded session independently, compare row by row, and rewrite any row where they disagreed before you certify anyone against it.