It's the most common move in sales management. Top rep crushes the quarter. Leadership decides to scale what they're doing. Interviews are scheduled. Playbooks are drafted. Templates get circulated. Six months later, the playbook is in CRM, adoption is at thirty percent, the slope hasn't moved, and the top rep is still the top rep.
Nothing replicated. This pattern is not because top reps are unicorns. It is because the methodology — interview the top rep, document the answers, publish the playbook — fails for three predictable reasons.
Why interviews fail.
Self-report bias. Top reps don't know what they're doing. Not literally — they're doing it expertly. But the cognitive layer that articulates what they're doing is downstream of the doing, and it is a notoriously unreliable narrator. Ask a top rep how they handle discovery and they'll describe a process. Sit in on three of their discovery calls and you'll watch them break the process they just described — in ways that work.
Attribution error. The top rep tells you what they think made the deal close. What actually made the deal close was often three or four moves earlier, in a moment they don't even remember as significant. The recency of the win biases the attribution toward the final move.
Context-dependence. The top rep's moves work because of who is making them. The same words, said by a different rep with a different relationship history, don't produce the same result. The move isn't replicable as written.
What to do instead.
Replicate behaviors, not people. The distinction sounds subtle. It is not.
Observe, don't interview. Sit in on real calls — across multiple reps, not just the top one. Note the actual moves, not what the rep says they did. Tag those moves. Look for the ones that show up across won deals, across multiple reps, across deal types. Those are the behaviors that compound.
The behaviors that only work for one rep are art — admirable, but not transferable. The behaviors that work across three or four reps are signal. Build the playbook from the signal.
The gap is almost always the same shape. The top rep, when interviewed, will describe a structured process — discovery, qualification, demo, close. Watch them, and you'll see something different. They open with a question that has nothing to do with discovery and everything to do with relevance: a specific operational pain that signals whether the buyer's organization is even in a position to act this quarter. The rep doesn't think of it as a move. They think of it as a normal opener. It is the move. And it is invisible to the playbook the interview would produce.
The four behaviors that almost always show up.
Patterns vary by org, but four behaviors recur across nearly every won-deal analysis I've run:
- Multi-thread before the demo, not after. Won deals consistently show a second or third stakeholder engaged before the product is shown. Lost deals show single-threaded discovery extending into procurement.
- Discovery question that uncovers a budget cycle, not a budget number. Won deals have a moment where the rep learns when the buyer's planning cycle starts. Lost deals stop at "we have budget."
- A pre-emptive objection handle. Won deals show the rep raising the most predictable objection before the buyer does. Lost deals let the objection emerge in the eleventh hour, when there's no time to address it.
- An executive sponsor with a stake. Won deals have an internal champion whose role-success is tied to the buying decision. Lost deals have a champion whose role-success is unaffected by the outcome.
These four are observable. They are taggable. They are coachable. They are what should be in the playbook — not the top rep's reconstructed narrative of their own process.
The top rep can't be replicated because the top rep is doing something the documentation can't capture. The good news is that you don't need to replicate them. You need to find the three to five moves that show up in won deals across your whole team — and make those moves the operating standard.