Discovery Call Structure: A Four-Phase Arc from Rackham, Keenan, Dixon, and Moesta

When to use

• First discovery call in a mid-market or enterprise deal where more than one stakeholder will touch the decision.
• Any complex sale where time-to-close exceeds 30 days.
• A re-qualification call after a deal has gone quiet or stalled.
• Competitive displacements where the buyer currently uses something else and switching is a risk in their mind.
• Wherever the real loss driver is indecision, not competitor preference. Dixon's research puts "no-decision" at 40-60% of lost pipeline. 40-60% of B2B deals are lost to "no decision", and 87% of deals show medium-to-high indecision

Not for:

• Transactional inside sales calls where the need is already established and the conversation is confirmation and pricing.
• Renewal check-ins with low-ACV accounts where a lighter discovery rhythm is sufficient.

The four phases

Phase 1. Frame (5-8 minutes)

Open by establishing what the counterpart came to accomplish, not by pitching context about your product. Most discovery calls fail here before any questions are asked: the rep opens with a company overview and the buyer's attention goes elsewhere.

Two moves make this phase work.

Set the agenda as a question. "I want to make sure we focus on what's actually useful for you today. What would make this an hour well spent?" This hands agenda-setting to the counterpart. What they say tells you where the live problem is, before you've asked a single diagnostic question.

Name your POV upfront. Orlob's data shows reps who lead with a specific, contrarian point of view enter discovery with higher buyer engagement than those who lead with questions about goals. A POV is not a pitch. It is a hypothesis about the counterpart's world that they can confirm, refine, or reject. "Most [role]s I talk to are dealing with [pattern]. Is that live for you?" This is not asking "tell me about your challenges." It names a specific problem and invites the counterpart into the diagnosis as a peer. Lead with a per-account point of view, not a templated cold ask

Why it matters: framing the call around the counterpart's agenda rather than yours surfaces the live problem faster. Framing it with a named hypothesis shows you already know something about their world, which raises the perceived quality of the conversation before a single question is asked.

Phase 2. Surface (15-20 minutes)

This is Rackham's investigation stage. Star performers spend more time here than on any other phase. Average performers rush past it toward the demo or the close.

Surface has two sub-layers.

Situation questions: establish the current state. What does the current process look like? How long has this been in place? Who owns this? These questions gather facts and should be used sparingly. Rackham found that too many situation questions annoy buyers because they could have been answered with pre-call research. Do the homework. Only ask what you genuinely cannot know from a website, a LinkedIn, or a prior call.

Problem questions: name the friction. What is not working about the current approach? Where does the process break down? Where does time get lost? Problem questions locate the pain. Without a named, felt problem, there is no gap. Without a gap, there is no sale. No problem, no sale. No gap, no large sale. Change = (Dissatisfaction × Desirability) > Cost of Change.

Moesta's switch-interview framing adds a temporal layer to problem questions. The buyer's current dissatisfaction was not born yesterday. There was a moment when the old way became untenable. "When did you first notice this was a problem?" surfaces the event-stream behind the pain, not just the state. The language the counterpart uses to describe that moment is the language your messaging should use. JTBD interviews surface the customer's actual language and the switch trigger

What good Surface sounds like:

• "Walk me through the last time this broke. What happened?"
• "Who gets pulled in when it breaks? What does that cost them?"
• "Is this a new problem or one you've been working around for a while?"
• "When did you first start looking for a different approach?"

What bad Surface sounds like:

• "Tell me about your goals." (Too open. No hypothesis. The counterpart reads it as the rep running out of preparation.)
• "What's your budget?" (Too early. Before the gap is established, budget is adversarial not diagnostic.)

Kapoor's insight on listening mode is the discipline that holds Phase 2 together: when you are in discovery, you are listening, not selling. Reps who start mentally preparing their pitch while the counterpart is still talking miss the detail that would have changed the call. Switch from selling mode to listening mode when the market breaks

Phase 3. Widen (10-15 minutes)

A felt problem has a stated size and an actual size. The stated size is what the counterpart says when you ask. The actual size is what emerges when you ask where the problem propagates.

Rackham calls these implication questions. They are the single question type that most separates star performers from average performers. Star reps ask four times more implication questions than average reps. Not because they are more curious. Because they understand that the buyer's felt urgency for change is not fixed at the stated problem. It grows when the buyer maps the downstream effects themselves. Top reps ask 4× more Implication questions, the highest-leverage question type in large sales

Implication questions follow a pattern: "You mentioned [problem]. What happens downstream from there?" or "When that breaks, who else feels it?" or "What does that cost in [time / headcount / deals / churn]?"

The goal is not to manufacture urgency by piling on negative emotion. Dixon's research shows that hammering the status quo after the buyer already understands it triggers defensiveness and activates omission bias, the tendency to prefer inaction because the risks of the current state feel familiar and the risks of change feel unknown. When buyers are indecisive, 73% of reps double down on hammering the status quo, and it backfires 84% of the time

The distinction: implication questions help the counterpart discover the full scale of the problem in their own words. That is different from the rep lecturing about how bad the status quo is. The buyer moving themselves to urgency is durable. Urgency induced by the rep's enthusiasm is not.

Keenan's change formula frames the threshold: change happens when (dissatisfaction × desirability of the future state) exceeds the cost of change. Phase 3 develops the left side of the equation. Phase 4 develops the right. Both have to exceed the switching cost or the deal stalls. No problem, no sale. No gap, no large sale. Change = (Dissatisfaction × Desirability) > Cost of Change.

Miller's three-level problem is useful scaffolding here. Every felt problem has three layers: an external problem (what the counterpart can describe), an internal problem (what it costs them emotionally or politically), and a philosophical problem (what it says about how things should work). Reps who stay at the external layer leave the internal and philosophical problems for the counterpart to resolve alone, often in favour of inaction. Implication questions that touch the internal and philosophical layers move the buyer further. Articulate the buyer's problem at three layers, external, internal, and philosophical, or your message rings shallow

What good Phase 3 sounds like:

• "When that breaks, who else in the org gets pulled into cleanup?"
• "What does that mean for your quarter if it happens again in Q3?"
• "Is this on a roadmap to fix internally, or has it been in backlog for a while?"
• "What would it cost you to keep the current setup for another year?"

Phase 4. Confirm (5-10 minutes)

The close does not live here. The confirmation of fit does. Most reps confuse the two and lose the deal. Closing techniques work in small sales and backfire in large sales, the more closing pressure, the lower the success rate

Phase 4 has one job: check whether the gap you've surfaced and widened is one your product can close. If yes, show how. If the fit is partial, say so. If the fit is absent, say so. Deals closed on shallow fit generate churn, implementation failures, and reference accounts that actively harm future pipeline.

Rackham calls these need-payoff questions. "If we could [specific capability], would that address [specific problem you surfaced]?" The structure matters: specific capability, specific problem. Generic capability against generic problem is not confirmation. It is mutual vagueness.

Dixon's indecision research adds the key constraint: most deals lost after a strong discovery are not lost to a competitor. They stall on buyer indecision. The driver is omission bias: buyers overweight the risks of the new and underweight the cost of staying put. Confirmation done well gives the buyer a reason to move and a specific picture of the first step. The biggest source of lost B2B deals is customer indecision, not competitor wins. JOLT them out.

Three things to confirm before ending Phase 4:

1. The problem is real and owned. "Does this match what you've been experiencing?" The counterpart should say yes, not something vaguer.
2. The fit is specific. "The part we'd address directly is [X]. The part that would stay on your side is [Y]." Half-fits named honestly close faster than full-fits claimed vaguely.
3. The next step is concrete. A time, a name, a format. Not "let's stay in touch." Dunford's framing: the real competitor in most deals is no-decision. A vague next step is a gift to inaction. 40–60% of B2B buyers say "no decision", your real competitor is the status quo

Sequencing failure modes

Four common mistakes in how reps run this arc:

Pitching before surfacing. Moving to capabilities before the counterpart has named their problem. The counterpart has no frame to attach the capabilities to, so the pitch slides off.

Skipping widening. Jumping from a surface-level problem statement straight to a demo. The problem stays small in the buyer's mind. The switching cost stays large. The deal stalls after a positive demo.

Closing without confirmation. Asking for a next step before the counterpart has confirmed the fit. They agree to the next meeting out of politeness. By the next meeting, inaction has won.

Over-implicating. Continuing to develop urgency after the buyer is already there. This triggers defensiveness, not momentum. Once the buyer has moved themselves to urgency, shift to desirability of the future state.

Pre-call preparation

The arc above fails without one upstream input: a hypothesis. Not a goal ("I want to sell them"). A hypothesis about what the counterpart's world looks like and where the live problem is. Orlob's POV-led framework makes this concrete. Before the call, write one sentence: "I think [role] at [company type] is dealing with [specific problem] because [observable signal]." That sentence is the POV you open with in Phase 1.

Two things the hypothesis surfaces during prep:

The research ceiling. List the facts you can confirm before the call (tech stack from the website, team size from LinkedIn, recent funding from Crunchbase, any mutual connections). Everything above that ceiling is a question. Everything below it is wasted call time. Reps who ask for information available on the company's homepage lose credibility before the first implication question.

The switching event. Moesta's switch-interview work maps a predictable event structure for every buying decision: passive looking (dissatisfaction exists but has not yet prompted action), active looking (the buyer is researching), the event that triggered the search, and the anxieties that slow or stop the final decision. If you know the counterpart is in active-looking, your implication questions can move faster. If they are still passive, your Phase 1 frame needs to surface whether they are ready to commit to the discovery process at all. Use the JTBD switch interview on recently-converted high-value customers, segments emerge from jobs, not demographics

Preparation is also where Dunford's positioning discipline enters. Before you can confirm fit in Phase 4, you need to know what your actual differentiated value is against the most likely alternative. Most buyers are not evaluating you against a named competitor. They are evaluating you against their current workaround. Discovery that surfaces what the workaround is, and where it fails, is more useful than a feature comparison matrix. Test positioning in a live sales pitch, marketing stories are unfalsified theory until then

Running the arc across a multi-stakeholder deal

Enterprise discovery rarely resolves in a single call with a single counterpart. The same four-phase arc applies per stakeholder, but the phase emphasis shifts by role.

Economic buyer. Weight toward Phase 3 (widening). Economic buyers care about scale and cost. They are often insulated from the day-to-day friction the operational buyer lives with. Implication questions that translate operational pain into financial and organisational exposure are the moves that move them. Skip or thin out Phase 3 here and the economic buyer never develops enough urgency to sponsor the deal internally.

Champion. Weight toward Phase 2 (surface). Champions live inside the problem and know it well. Your job in Phase 2 is to earn their trust by demonstrating you understand their world accurately. Champions who feel understood become internal advocates. Champions who feel interviewed by someone who doesn't know their domain go quiet after the call.

Technical buyer or security/compliance reviewer. Weight toward Phase 4 (confirm). Their question is not "is the problem real" or "is the gap wide." It is "can your product do what you say, and at what risk." Confirmation with technical buyers is about precision, not warmth.

Gartner's research on adoption shows that change statements, framing why the change matters before capability framing, drive higher stakeholder buy-in at every layer. This maps directly to the arc: Phase 2 and 3 build the change statement. Phase 4 is where capability is introduced. Flipping the order (leading with capability) produces higher initial interest but lower downstream conviction, because the buyer never moved their own need state. 60% of users get new apps without a "why it helps", fix that before any capability rollout

What to do when the call goes off-arc

Three common off-arc scenarios and the recovery move for each.

The buyer pitches you. Some counterparts arrive with a solution in mind and spend Phase 1 describing what they want. Recovery: let them finish, then ask "what problem were you solving when you started looking for this?" This resets to Phase 2 and surfaces the gap behind the solution-request. A buyer who has already decided what they want is not in discovery mode. You need to earn your way back to the problem before the fit can mean anything.

The call becomes a demo. The buyer asks to see the product before Phase 3 is complete. Recovery: "Absolutely. Before I show you the part most relevant to you, can I confirm I understand the problem correctly? I want to make sure I'm showing you the right thing." Then surface and widen before the screen share. Demos without a developed need state produce "looks interesting" responses and second meetings that go nowhere.

The buyer disqualifies on price. Budget comes up before Phase 3 is complete. Recovery: do not defend the price. Instead, ask "before we get into pricing, can I check whether I understand what we'd be solving? The cost question looks different once we've mapped the problem." This is not a deflection. It is the honest position: without a mapped gap, neither party knows whether the price is high or low relative to the value.

Metrics

Roberge's engineering-discipline model of sales gives this arc its measurement layer. Phase outputs are observable and coachable. Sales is engineering, not art, system design beats individual talent at scale

Manager call review should track question-type ratios, not just outcome. Rackham's finding is consistent: the behaviours that differentiate stars from average reps are visible in the questions, not the close. If coaching only touches outcomes (did they get the next meeting?), it misses the mechanisms.

Call documentation

The output of a well-run discovery call is a problem statement in the buyer's own language. Not "they're struggling with efficiency." That is a category, not a problem. A real problem statement is: "Their current tool breaks when more than three people are editing the same template, so the ops team manually merges versions every sprint, which takes four hours and introduces errors that downstream teams catch in QA."

That statement is what you write in the CRM immediately after the call, before the memory softens. Reps who wait until end of day paraphrase. Paraphrase loses the buyer's language. The buyer's language is the raw material for messaging, positioning tests, and the next call's POV. JTBD interviews surface the customer's actual language and the switch trigger

Three fields, not a paragraph:

1. The current state (how the process works now, in their words).
2. The friction event (the specific moment it breaks, and how often).
3. The downstream cost (what happens to people, time, or money when it breaks).

If you cannot fill all three fields, the call was not complete. Either you did not surface enough (Phase 2 was thin) or you did not widen enough (Phase 3 was skipped). Go back in the next touchpoint. Orlob's five-touch sequence model is relevant here: discovery is rarely complete in one call, and knowing which phase was thin tells you what the next touchpoint's job is. Each outbound touch must build on the prior, not restate it; the 2nd and 3rd touches double reply rates when this holds

Relationship to the Tactical Empathy playbook

This playbook and the-tactical-empathy-discovery-playbook.md run in parallel on the same call. The arc above (frame, surface, widen, confirm) is the structural skeleton. Voss's tools (mirroring, labelling, accusation audit) are the relational tissue. The skeleton without the tissue produces a call that feels like an interrogation. The tissue without the skeleton produces a warm conversation that surfaces nothing actionable.

When to weight toward this playbook: when the counterpart is cooperative and the primary risk is shallow discovery, missed implications, or premature closing.

When to weight toward the Tactical Empathy playbook: when there is visible resistance, a stalled deal, a renewal in trouble, or an escalation where rapport has broken down.

In most new enterprise discovery calls, run both.

Quick-reference: phase checklist

The checklist is not a script. It is a minimum. Calls that complete all four phases with specific outputs close faster and close more cleanly than calls that feel good but leave any phase thin.