The first version of Coachvyne's behavior model came from a disagreement. Half the team believed talk-to-listen ratio was the strongest predictor of discovery call outcomes. The other half believed it was question rate. Both camps were wrong — or rather, both were measuring proxies rather than the underlying behaviors that actually correlated with closed-won.
We spent 14 months annotating discovery calls across a range of B2B SaaS teams, tagging 23 distinct behaviors per call and then running correlation analysis against CRM outcome data. Most behaviors washed out. Seven didn't. What follows is a precise account of what those seven are, why they outperform simpler metrics, and what it actually looks like to coach them.
Why single-metric coaching fails
A manager who coaches "talk less" based on a 45% rep talk ratio is optimizing a proxy. The underlying question is: what does more listening actually buy you? If a rep is silent while a prospect rambles about an irrelevant topic, the silence is worthless. The behavior that matters is whether the rep's listening generates useful signal — confirmed pain, quantified impact, stakeholder names, decision timelines.
This distinction between proxy metrics and behavior-outcome correlation is why most conversation intelligence implementations produce dashboards that get checked once and then ignored. A metric that correlates at 0.3 with win rate across all reps tells you almost nothing about which rep to coach on what, this week, with what drill.
The seven behaviors below each correlate significantly with closed-won at the individual call level, not just at the aggregate level. That's the standard we held ourselves to during the annotation process.
Behavior 1: Problem Articulation Depth
This is distinct from "did the rep ask about pain." It measures whether the rep got the prospect to articulate the problem in their own words, with specificity, at least twice in the same conversation. A prospect saying "yeah, our forecasting is a mess" once is noise. A prospect saying "we close the quarter with a 40% variance between commit and actuals, and our CRO has been in two board meetings this year explaining overages" is signal. The second version has been drawn out through follow-up probes.
On closed-won calls, reps triggered this behavior in 74% of cases. On closed-lost calls at the same stage, the rate dropped to 31%. The gap is not because top reps are cleverer — it's because they have a repeatable follow-up pattern: initial open question, acknowledgment, clarifying probe, then a quantification bridge ("how much is that costing you in aggregate?").
Behavior 2: Quantified Impact
This is the most commonly missing behavior, and also the one with the highest correlation to win rate among the seven. The behavior is defined as: did the rep establish a dollar figure, a time cost, or a headcount cost associated with the stated problem, and did the prospect confirm it?
It's not enough for the rep to estimate the impact. The prospect must confirm or generate the number themselves — the psychological ownership of "we're losing about $200K a quarter to this" is categorically different from the rep saying "so that's roughly $200K impact?" and hearing "yeah sure."
We're not saying quantification is only about ROI calculation. It's about whether the prospect has connected the problem to a budget-level pain. A VP of Sales who says "our ramp time is seven months" has a problem. A VP who says "seven months means we're paying $30K in OTE before a rep generates a dollar" has a problem with a budget owner's fingerprints on it.
Behavior 3: Decision Process Mapping
In MEDDIC terms, this is the "D" — decision process and decision criteria. Most reps ask one of two questions: "Who else is involved in this decision?" or "What's your timeline?" Neither is wrong, but neither is sufficient.
Decision process mapping, as we define it, requires the rep to have established: (a) the formal steps required before a contract is signed, (b) who reviews each step, (c) what criteria are applied at each gate. A rep who knows there's a security review, that security requires SOC 2 Type II, and that the review takes 3 weeks has mapped the process. A rep who knows "legal will look at the contract" has a single data point.
Calls where reps mapped at least two distinct decision steps converted at 2.1x the rate of calls where they mapped zero or one. The correlation held even after controlling for deal size.
Behavior 4: Champion Identification
A champion is not the person you're talking to. A champion is someone who will actively sell on your behalf when you're not in the room — and on a discovery call, the behavioral test is whether the rep identified that person explicitly and verified their internal influence.
The tell-tale pattern on winning calls: the rep asks something like "who internally would be most affected if this problem doesn't get solved in the next six months?" — and gets a name that's different from the current contact. Then they probe whether that person has budget access or authority to prioritize. Reps who do this are three times more likely to have an active champion by the time they reach proposal stage, versus reps who default to assuming their contact is the champion.
Behavior 5: Objection Surfacing
This behavior confuses managers more than any other, because it's counterintuitive: reps on winning calls surface more objections, not fewer. The reason is that surface-level discovery calls generate surface-level objections late — in the pricing call, in the security review, in the legal red-line. Winning reps pull objections forward into the discovery call by asking directly: "What would make this a no-go for you or your team?" or "What's the most common reason a project like this stalls at your company?"
Objections surfaced in discovery are objections that can be addressed, reframed, or factored into the deal structure. Objections that surface at proposal stage are deal-killers. The behavioral score is binary: did the rep proactively invite at least one objection, and did they handle it using the acknowledge–clarify–respond–confirm pattern rather than deflecting?
Behavior 6: Next Steps Specificity
The close of a discovery call predicts the deal's trajectory more reliably than almost anything in the middle of it. The behavioral standard is not "did the rep propose a next meeting" but whether the next step had: a specific date confirmed, a named attendee from the prospect side, and a stated purpose (not just "follow-up call").
Consider the difference between "I'll send you a recap and we'll find some time next week" versus "I'll send you a recap by Thursday. Can you confirm whether Sarah from your IT security team can join us on Tuesday the 14th? We want to cover the SOC 2 question directly with her." The second version commits a second stakeholder, advances the decision process, and sets a concrete expectation the prospect agreed to — all in 30 seconds.
Calls with specific next steps closed at 38% versus 19% for calls with vague or no next steps. That's the single largest win rate gap between any high-scoring and low-scoring behavior pair we measured.
Behavior 7: Competitive Positioning
This does not mean asking "are you looking at anyone else?" — that's a surface probe. Competitive positioning as a discovery behavior means the rep established whether the prospect is evaluating alternatives, which alternatives, and what the selection criteria are, and then connected at least one key differentiator to a stated prospect priority without trashing the competitor.
Reps who do this in discovery control the evaluation frame. Reps who don't do this find out in the technical evaluation that the prospect was simultaneously running a POC with a cheaper alternative and has already half-committed to it.
There's a specific pattern on winning calls: the rep confirms the competitive landscape, then asks which criteria matter most to the prospect, then makes one explicit connection between a differentiator and a high-priority criterion. This is the Challenger Sale's "commercial teaching" applied at the discovery stage.
The composite score and what it means for coaching
No single behavior is determinative. A call that scores 7/7 doesn't guarantee a close, and a call that scores 3/7 doesn't guarantee a loss. What the composite score gives you is a coaching prioritization signal: reps who consistently score below 4/7 on behaviors 1, 2, and 6 need different coaching than reps who score above 5/7 across the board but have a specific gap in behavior 3 or 7.
The practical application is call-level scoring fed into a weekly coaching plan. A manager with 9 direct reports who each run 5 discovery calls a week is looking at 45 data points per week. Without behavioral scoring, the manager will review the 2 calls the loudest rep mentioned in the 1:1, form a subjective impression, and give generic advice. With scoring, the manager sees that three reps have a consistent behavior 2 gap, one has an isolated behavior 7 gap, and the rest are running at an acceptable composite. The first group gets the quantified impact drill this week. The second gets one targeted competitive positioning exercise. The rest get pipeline review time instead of coaching time.
That's the difference between a coaching program that moves metrics and one that just generates activity. The seven behaviors aren't a checklist — they're a diagnostic vocabulary for having the right conversation about the right gap at the right time.