There's a number that gets cited constantly in sales coaching circles: top performers talk 43% of the time. It comes from a frequently shared analysis of call data across a broad pool of B2B sales calls, and it's become the kind of benchmark that gets dropped into coaching sessions and sales methodology decks as if it's a law of physics.
We looked at 8,400 calls scored in Tunlai over a 14-month period, across teams with different average contract values, buyer personas, and deal cycles. The 43% figure is not wrong — it's just not useful. The optimal talk-to-listen ratio varies enough by context that applying a single benchmark across all calls is more likely to produce bad coaching than good.
The Aggregate Problem
The 43% number comes from averaging across call types — prospecting calls, discovery calls, demo calls, negotiation calls — and across deal sizes and industries. When you average across all of that, you get a number that's technically accurate for the aggregate and misleading for any specific situation.
Here's what the breakdown looks like when you disaggregate. In our data, first discovery calls with technical evaluators — where the prospect is assessing whether your product is even worth a second conversation — the winning calls have the rep talking between 35% and 40% of the time. The rep's job on that call is primarily to ask good questions and confirm they understand what the evaluator cares about. Talking more than that tends to correlate with lower stage conversion rates.
Demo calls look different. When a rep is presenting to a buying group that has already completed discovery and expressed intent to evaluate, the winning calls trend toward 50-55% rep talk time. The rep is presenting, handling questions, connecting features to stated pain points. The ratio is higher because the call purpose is different.
Negotiation calls and mutual action plan reviews are different again. On those calls, the rep who talks less than 40% of the time often isn't driving the conversation forward — they're being reactive rather than guiding the deal to a close.
Buyer Persona Matters More Than Stage
Even within a call type, buyer persona shifts the optimal ratio more than most coaching frameworks acknowledge. When the economic buyer is on the call — the person who actually controls the budget — winning calls consistently show higher rep talk time than calls with technical evaluators or end users. Economic buyers are typically time-constrained, and they expect the rep to come in prepared with a clear business case. Lengthy open-ended discovery with an economic buyer often reads as unpreparedness rather than active listening.
Technical evaluators and practitioners are the inverse. They want to be heard. They have specific concerns — integration complexity, data security posture, customization limits — and they're evaluating whether the rep actually understands their environment. A rep who listens 65% of the time on a technical deep-dive call is not performing poorly. They're doing exactly what the situation requires.
This distinction has direct implications for how you coach. Telling a rep "you need to get your talk time up to 43%" before they hop on a technical evaluation call is counterproductive. Telling them "on the economic buyer call this Thursday, plan to lead with the business case for 10 minutes before you open it up for questions" is the kind of specific, situational coaching that actually changes outcomes.
What the Number Misses Entirely
Talk-to-listen ratio captures quantity — how much of the call the rep is speaking. It says nothing about the quality of either the talking or the listening. We've seen calls where a rep talks 60% of the time and every minute of it is useful: tight questions, crisp feature explanations, clear handling of objections with genuine diagnosis before response. And we've seen calls where a rep talks 40% of the time and most of it is filler — restating what the prospect just said, overly long transitions, unnecessary preamble before questions.
We're not saying talk ratio is a useless metric. It's a useful first signal that something might be off. A rep consistently running 70% talk time across their discovery calls has a problem worth investigating. But the investigation needs to go deeper than the number. What are they talking about? Are they filling silence because they're uncomfortable, or are they structuring the conversation? When they're listening, are they actually processing what the prospect is saying — or waiting for their next talking point?
The behaviors that actually predict closed-won outcomes are downstream of the ratio: question-to-statement ratio within the rep's talk time, response latency after prospect statements (a proxy for active listening vs. scripted follow-up), and whether the rep's next question connects directly to what the prospect just said or shifts topic.
Deal Size Creates Its Own Effect
We also see a deal-size effect that isn't discussed much. In early-stage deals under $20,000 ACV, the difference in talk ratio between closed-won and closed-lost calls is relatively narrow — a few percentage points in either direction. The rep's talk time matters less at lower deal sizes because the purchase decision involves fewer stakeholders, shorter cycles, and less risk for the buyer.
In deals above $80,000 ACV — where you're typically running multi-stakeholder evaluations, security reviews, and legal negotiations — the talk ratio difference between winning and losing calls is more pronounced, but the direction is not uniform. On early-cycle calls in large deals, winning reps talk less. On late-cycle calls with procurement and legal, winning reps often talk more — because they're guiding a complex process rather than discovering a problem.
A useful thought experiment: take your top three closed-won deals from last quarter. Pull up the call recordings. Listen to the demo call and the final negotiation call. The rep's talk ratio is almost certainly different between those two calls, and that difference reflects appropriate adjustment to conversation purpose — not inconsistency to be corrected.
How We Approach This in Tunlai Scoring
When we score calls on talk-to-listen ratio, we don't score against a universal benchmark. We score against the pattern from your team's own closed-won calls, segmented by call type and stage. If your best discovery calls average 38% rep talk time and your current call is at 55%, that's a coaching signal worth surfacing. If your best demo calls average 52% and a rep's demo runs at 53%, there's nothing to coach there.
The scorecard also weighs talk ratio alongside the quality indicators — question density within talk time, topic follow-through, objection diagnosis before response. A rep who talks 60% of the time but has high question density and strong follow-through on prospect statements is being coached differently than a rep who talks 60% of the time with long monologue sequences and shallow objection handling.
The goal is to give managers a coaching signal that's specific enough to act on. "Your talk ratio is high" isn't actionable. "In your last three discovery calls, you averaged 58% talk time and your question density dropped significantly after minute 15 — the point where prospects typically start raising process concerns" is something a rep can work with in the next call.
A Practical Alternative to the 43% Rule
For VP Sales and enablement leaders who want something practical to replace the single-number benchmark: define the optimal talk ratio range for each call type in your deal cycle, using your own closed-won data as the reference. Discovery calls at your ACV and deal complexity might land at 35-42%. Demos might land at 48-55%. Use those ranges as the coaching reference, not an industry average that was derived from a different mix of deals than yours.
Then look at variance, not just average. A rep who's at 60% talk time consistently across all call types is different from a rep who adjusts their ratio depending on the call. The adjustment is the skill. Reps who can't modulate their talk ratio based on the conversation they're actually in are going to hit a ceiling — because the ability to read when to talk and when to hold silence is one of the things that separates the top quarter of performers from the middle.
The 43% number isn't useless. But if it's the main thing your coaching framework is tracking, you're measuring a proxy when you could be measuring what actually moves deals forward.