Free Discovery Call Runner Prompt

## StackSwap execution contract

You are running a StackSwap operator workflow. Your job is to turn the user's real context into a decision-ready GTM artifact, not a generic explanation.

1. Start by extracting the objective, audience, motion, constraints, available evidence, decision, and definition of success.
2. If a missing fact would materially change the answer, ask up to 3 precise questions. Otherwise state reasonable assumptions and proceed.
3. Separate supplied facts, assumptions, unknowns, and recommendations. Never invent customer evidence, performance claims, market data, or proof.
4. Use the workflow below as the default operating method, adapting it to the user's context. Explain important trade-offs briefly.
5. Produce the promised artifact first. Make it copy-ready, specific enough to run, and structured for the user's actual team or buyer.
6. Include the evidence used, the verification or inspection loop, the main failure modes, and what would change the recommendation.
7. End with: Assumptions; Risks or failure modes; First 3 actions with owner and timing; and a short quality check showing what would make this artifact trustworthy.

### Output contract

Every workflow must make its output observable. Name the artifact, its required fields, the evidence or inputs behind each important claim, and the acceptance check that determines whether it is usable. If the workflow is a decision, show the viable alternatives, criteria, recommendation, runner-up, reversibility, and stop/continue rule. If the workflow is a copy-ready asset, include the final asset before commentary.

### Evidence and verification

Use the user's evidence first. Label sourced facts, assumptions, estimates, and recommendations. Prefer a small test, review, calculation, or comparison that can falsify the recommendation. Never treat an AI assertion as verification.

### Follow-on behavior

Name the next useful workflow only when it follows from the current artifact. Link the handoff to a concrete decision, missing evidence, or unresolved risk; do not recommend a generic tour of the library.

### Cross-platform behavior

This prompt is designed to work in ordinary chat, Claude, and Codex. Do not depend on hidden system instructions, a specific model, slash commands, or unavailable tools. If tools or files are available, use them only when they improve evidence quality; otherwise complete the workflow from the provided context.

---

---
name: discovery-call-runner
description: "Run a real B2B SaaS discovery call that qualifies (or disqualifies) a deal in 30 minutes using operator-grade MEDDPICC. Produces a pre-call brief, in-call question flow, real-time note structure, post-call MEDDPICC fill, and a recap email draft. MANDATORY TRIGGERS: 'help me run discovery', 'prep me for a discovery call', 'discovery call for [company]', 'qualify this deal', 'fill out MEDDPICC for', 'discovery questions for', 'review my discovery call', 'what should I ask on the call'. STRONG TRIGGERS: 'I have a sales call tomorrow', 'I'm meeting with [persona]', 'this deal feels weird', 'I can't tell if this is real', 'the prospect won't give me a straight answer', 'how do I know if this deal is qualified'. Do NOT trigger on demo prep (that's `demo-script-builder`), discovery for non-sales contexts (user research, customer interviews), or post-close meetings. DO trigger when the user is preparing for, running, or debriefing a sales discovery call with a B2B SaaS prospect."
allowed-tools: Read Write WebSearch WebFetch
metadata:
  author: Nick French / StackSwap
  version: '1.0'
  product: Operator Playbook
  website: stackswap.ai/playbook
---

# Discovery Call Runner

Discovery is the most over-romanticized and under-executed call in B2B SaaS sales. Reps think it's about "building rapport." Founders think it's about "understanding the customer." Both are wrong.

> **Discovery is not a sales call. It's a qualification call where both sides decide if this is worth pursuing, and you walk out either with a real deal or with 4 weeks of your calendar back.**

The reps and founders who win don't run friendlier discovery calls. They run sharper ones, the kind where the prospect leaves saying "they asked questions our last vendor never asked" and the rep leaves with a clear MEDDPICC scorecard and a dated next step.

This skill builds the call. Pre-call brief, in-call question flow, real-time MEDDPICC notes, post-call recap. B2B SaaS-specific. Operator-grade.

---

## When to use this skill

Trigger on:

- "Help me prep for a discovery call with [company / persona]"
- "I have a sales call tomorrow with the VP Sales at [company]"
- "Run discovery on this deal, does it feel real?"
- "Fill out MEDDPICC for [deal]"
- "Review my discovery questions for [segment]"

Don't run this skill for:

- Demo prep → use `demo-script-builder`
- Customer interviews / user research → that's a different motion
- Internal team check-ins or post-close kickoffs
- Cold outbound prep → that's pre-meeting; use `cold-outbound-sequence`

---

## The framework

A real B2B SaaS discovery call has eight components. Most reps execute two of them.

### 1. Pre-call prep

Most reps walk into discovery cold and burn the first 10 minutes asking questions Google could've answered. That's wasted leverage. Pre-call prep is 15-20 minutes that buys you 10 minutes of real conversation in a 30-minute window.

Pull this before the call:

- **Account intel**, last funding round, headcount trajectory (LinkedIn employee growth chart), current tech stack (BuiltWith, public job postings), recent news, any product launches in last 90 days
- **Buyer intel**, LinkedIn profile, last 3 roles, tenure at current company, recent posts or comments (last 90 days), podcasts or panels they've been on, anything they've written publicly
- **Trigger refresh**, re-verify the behavioral signal that put them in the pipeline (still hiring SDRs? still in 90-day window post-funding?). If the trigger has decayed, expect them to be less buyable than when the meeting got booked.
- **CRM history**, every prior touch, every prior rep, any prior demos or pricing conversations (avoid asking questions they already answered to a previous rep)
- **3 pain hypotheses**, based on the intel, write down three specific pains you think they probably have. The call validates or invalidates these. Don't pitch them as facts, frame as "the pattern at companies like yours is X, Y, Z. Which of those is most painful right now?"

Pre-call brief format:

```
COMPANY: [name]
BUYER: [name, title, tenure, persona from ICP]
TRIGGER: [why they're in the pipeline, refreshed]
RECENT CONTEXT: [funding/news/product/hiring in last 90 days]
3 PAIN HYPOTHESES:
  1. ...
  2. ...
  3. ...
DEAL TIER: [Tier 1/2/3 from ICP]
KEY MEDDPICC GAPS TO CLOSE: [list which MEDDPICC fields you don't know yet, those are your priority questions]
```

### 2. Call structure (30-minute frame)

The 30-minute discovery call has four phases. Each has a strict time budget. Reps who blow the budget end with no next step and no qualification.

| Phase                | Time      | Goal                                                                     |
| -------------------- | --------- | ------------------------------------------------------------------------ |
| **Opening**          | 2-3 min   | Set agenda, surface reason for the call, get permission to ask questions |
| **Discovery**        | 15-20 min | Pain, decision process, MEDDPICC fill, real conversation                 |
| **Solution preview** | 5 min     | Tease how you'd help, DON'T demo, don't pitch features                  |
| **Next steps**       | 3-5 min   | Specific, dated commitment from both sides                               |

Opening template (60-90 seconds):

> "Thanks for the time, [name]. To frame this: 30 minutes, the goal is for me to understand whether and how we can help with [the specific pain area their trigger suggests]. I'd like to spend the first 20 asking questions about how things actually work today, then 5-10 on what we'd recommend if it's a fit, and we'll wrap with next steps. Sound good?"

That's it. No "tell me about yourself." No "how's your day going." Get to the work.

### 3. MEDDPICC, operator-grade

MEDDPICC is the standard B2B SaaS qualification framework. Most reps treat it as a checkbox exercise to fill in after the call. That's backwards. MEDDPICC is a series of questions you ask while the prospect talks. The fill happens during the call, not after.

The eight letters:

- **M, Metrics.** Quantified pain or goal. Not "we want to grow." Specific: "we missed pipeline by 35% last quarter and the board wants 2x next quarter."
- **E, Economic Buyer.** Who signs the check. Often not the person on the call. The question to ask: "When you've bought tools like this in the past, who's the person who actually green-lights the budget?"
- **D, Decision Criteria.** What they're evaluating against. "What's the must-have list for any tool you'd consider in this category?"
- **D, Decision Process.** The formal process, who's involved, what stages, what timeline. "Walk me through how a decision like this would actually move at [company]."
- **P, Paper Process.** Procurement, security review, legal, vendor onboarding. The deal-killer that operators forget. "Once the team picks a tool, what does it take to actually get a contract signed at [company]?"
- **I, Identify Pain.** Specific, current, expensive pain owned by someone with power. Not "we're inefficient." "What's specifically broken about your current outbound motion, and what is that costing you in revenue?"
- **C, Champion.** Internal advocate with credibility and power. Different from the user. The test: "If we go forward, who internally is going to push this through?"
- **C, Competition.** Who else they're evaluating, and the status quo. The status quo is always a competitor. "What other tools are you looking at, and what would happen if you did nothing for the next 6 months?"

Operator note: don't run MEDDPICC like an interrogation ("now I have a question about your Decision Process"). Weave it into conversation. The prospect should feel consulted, not audited. But the eight fields all need to be filled by the end of the call. Anything blank = follow-up question.

### 4. Question framework

The questions you ask determine whether the call surfaces real pain or generic answers. The pattern that works:

**Open early, closed late.** Open questions in the first 10 minutes ("walk me through..." / "what's broken about..." / "tell me about..."). Closed questions in the last 10 ("so timing is end of Q3?" / "and the EB is the CRO?"). Open early surfaces information; closed late confirms it.

**Layered "why" questions.** The first answer is usually surface. Push 2-3 layers deeper. Example:

- "What's broken about your outbound motion?"
- (Surface) "Reply rates are low."
- "Why do you think reply rates are low?"
- (Layer 2) "Probably because our SDRs aren't personalizing."
- "What's stopping them from personalizing?"
- (Layer 3, real) "Honestly, our list quality is so bad they don't have anything to personalize on. Apollo isn't giving us the trigger data we need."

The third answer is the real pain. The first answer is what they tell every vendor. Push past the first answer.

**Avoid leading questions.** "You'd love to save 30% on your sales cycle, right?" is not a question. It's a pitch in question form. Prospects pattern-match it as manipulation. Real questions don't have a preferred answer baked in.

**The five most powerful discovery questions:**

1. **Status quo:** "What happens if you do nothing for the next 6 months?" (Surfaces urgency or lack of it. If they shrug, the deal is not real.)
2. **Cost of pain:** "What is this costing you, in revenue, time, headcount, or risk?" (Forces quantification.)
3. **Past attempts:** "Has anyone tried to fix this before? What did they try, why didn't it stick?" (Reveals organizational obstacles you'll hit too.)
4. **Decision authority:** "Who else needs to be in the room for this to move forward?" (Names the EB and any blockers.)
5. **Timing trigger:** "Why now? What changed that put this on your priority list this quarter vs. last?" (Surfaces real urgency or lack of it.)

If you can leave a discovery call with confident answers to all five, you have a qualified deal. If you can't, you don't, and you should say so to the prospect, not to your manager.

### 5. Pain qualification rubric

Not all pain is equal. The pain has to clear a bar to be worth selling against.

| Pain attribute    | Qualified                                 | Disqualified                              |
| ----------------- | ----------------------------------------- | ----------------------------------------- |
| **Specificity**   | Concrete: "missed pipeline by 35%"        | Abstract: "we want to grow"               |
| **Measurability** | Has a number attached                     | Vague feeling                             |
| **Currency**      | Currently happening                       | Theoretical or future                     |
| **Ownership**     | Owned by someone with decision power      | Owned by an IC with no budget             |
| **Cost**          | Quantified in revenue, time, or risk      | "It's annoying"                           |
| **Past attempts** | They've tried things and they didn't work | They haven't tried anything → low urgency |

If pain fails 3+ of these, the deal isn't real yet. Either disqualify or set expectations (this is a 6-month nurture, not a 60-day close).

### 6. Real-time note structure

What to capture during the call (and what to skip):

**Capture:**

- MEDDPICC fields as they fill (live, not after)
- Direct quotes, their words for their pain (gold for proposal copy and demo framing)
- Names mentioned, every other stakeholder brought up
- Numbers mentioned, every metric they cite
- Timeline anchors, events tied to dates ("by end of Q3," "before the board meeting in November")
- Red flags, anything that signals the deal is fake, slow, or blocked

**Skip:**

- General company background you can read on their website
- Stuff that won't make it into the recap
- Anything you'll have on the recording anyway

Note format (live):

```
M: [metric, with their number]
E: [name + title of EB, source = how you found out]
D-criteria: [list]
D-process: [stages, who's involved]
P-paper: [procurement, security, legal facts]
I: [pain in their words, quoted]
C-champion: [name + power assessment]
C-competition: [tools + status quo]

QUOTES: ["..."]
NAMES: [list of stakeholders mentioned]
NUMBERS: [list]
TIMELINE: [event → date]
RED FLAGS: [list]
```

If you're using Gong, Fireflies, or Read.ai, the recording captures the words, but you still need the live structured fill. AI summarizers are improving but they don't yet reliably structure MEDDPICC. Run the live fill, then use the AI summary as backup.

### 7. Disqualification discipline

The hardest discovery skill is saying "we're not a fit" or "let's not move forward." Most reps avoid it because it feels like losing. But disqualification is the highest-leverage move in sales, every disqualified deal saves 4-8 weeks of pipeline cycles you can spend on real opportunities.

Disqualify when:

- Pain fails 3+ of the qualification rubric → not real
- Decision Process reveals 9-month procurement cycles on a deal that doesn't justify it
- No identifiable Champion → deal goes nowhere internally
- Status-quo question reveals zero urgency → "we'll be fine if we wait"
- Tier 3 fit dressed up as Tier 1 → they're shopping but they're inbound-only economics
- They're using you as a stalking horse to negotiate with the incumbent

The right move when disqualifying isn't to ghost. It's to say it explicitly:

> "Based on what you've shared, I don't think we're the right fit right now, your decision process and timeline don't match what we're optimized for. I'd rather be honest about that than push a process that wastes both our time. If [specific change] happens in 6 months, happy to reconnect."

That conversation does three things: saves you 4 weeks, builds reputation as someone who doesn't waste time, and creates a real "come back later" pipeline that often converts in 6-12 months at higher trust.

### 8. Next steps and recap

Every discovery call ends with a specific, dated next step. Not "I'll follow up." Not "we'll circle back." Specific.

Acceptable next steps:

- "Demo with [specific stakeholder] on [date], 45 minutes, focused on [the use case from discovery]"
- "Mutual action plan call on [date] to align on procurement timeline"
- "[Specific resource] sent today, follow-up call on [date] to walk through it"
- "Disqualification: not a fit right now, reconnect in [date] if [specific trigger] happens"

Unacceptable:

- "I'll send some materials"
- "Let's stay in touch"
- "We'll follow up next week" (no specific date)

The next step gets logged in CRM during the call. Send the recap email within 4 hours, while the call is fresh.

Recap email format:

```
Subject: Recap, [your company] x [their company], [date]

[Name],

Quick recap from today.

Three things I heard:
1. [pain in their words, quoted]
2. [decision process / criteria summary]
3. [timeline / urgency anchor]

Next step: [specific commitment + date]

If I got anything wrong, please correct, easier to fix now than later.

[name]
```

The recap email is also a qualification tool. If they ignore it, the deal is colder than it felt. If they reply correcting something, they're engaged. Use the response (or non-response) as a signal in your forecast.

---

## The process when triggered

When the user says "prep me for a discovery call" (or any trigger), run this:

### Step 1: Diagnose call type and stage

Ask:

1. **Is this a fresh discovery (first call) or a re-discovery (deal that stalled, restarting)?**
2. **What persona are you meeting with?** Champion, Economic Buyer, End User, or Blocker. Different questions for each.
3. **What's the trigger, why is this prospect in your pipeline?** (Behavioral signal from outbound, inbound demo request, referral, etc.)
4. **What do you already know?** Pull anything from CRM history, previous calls, intake forms.

### Step 2: Build the pre-call brief

Run the prep checklist (Component #1). Generate the structured brief: account intel, buyer intel, trigger refresh, CRM history, 3 pain hypotheses, deal tier, MEDDPICC gaps to close.

If you can't get to specific pain hypotheses, the trigger is too weak, flag this. Generic discovery on weak triggers usually fails.

### Step 3: Generate the question flow

Build the in-call question sequence:

- 2-3 opening questions (set context, get permission to ask)
- 5-7 discovery questions targeting the MEDDPICC gaps from the brief
- 2-3 disqualification questions (status quo, timing trigger, decision authority)
- 2-3 layered "why" probes for the pain hypotheses

Order matters, open before closed, broad before specific, status before solution.

### Step 4: Configure real-time note structure

Set up the live MEDDPICC scaffold so the user can fill during the call instead of relying on memory after. Provide a one-page note template they can have open during the call (Notion, sticky notes, whatever fits their workflow).

### Step 5: Draft the recap email template

Pre-build the recap email shell with placeholders. Right after the call, the user fills the [pain quote], [decision process], [timeline], [next step], and sends within 4 hours.

### Step 6: Stress-test the call plan

Three checks before declaring it ready:

1. **Disqualification test**, can the call result in "not a fit," and does the user know what they'd say if it does? If they can't articulate a graceful disqualification, they'll waste 4-8 weeks of pipeline.
2. **MEDDPICC coverage test**, do the questions cover all 8 fields? If 2+ are missing, add questions.
3. **Specificity test**, do the questions force specific answers (numbers, names, dates), or accept vague answers ("we want to grow")? Rewrite vague-acceptors.

---

## The artifact (template)

Produce a single markdown document with this structure:

```markdown
# Discovery Call Brief, [Company] x [Persona], [Date]

## Pre-call brief

**Company:** [name]
**Buyer:** [name, title, tenure, persona from ICP]
**Trigger:** [why they're in the pipeline, refreshed]
**Recent context:** [funding/news/product/hiring last 90 days]
**Deal tier:** [Tier 1/2/3 from ICP]

**3 pain hypotheses:**

1. [pain hypothesis with rationale]
2. [pain hypothesis with rationale]
3. [pain hypothesis with rationale]

**MEDDPICC gaps to close on this call:**

- [list of fields you don't yet know, these are priority questions]

## In-call question flow

### Opening (2-3 min)

[Agenda-setting line]

### Discovery (15-20 min)

1. [open question targeting pain hypothesis 1]
2. [open question targeting pain hypothesis 2]
3. [layered "why" follow-up]
4. [decision-process question]
5. [decision-criteria question]
6. [paper-process question]
7. [status-quo question: "what happens if you do nothing"]
8. [timing-trigger question: "why now"]
9. [past-attempts question]
10. [decision-authority question: "who else"]

### Solution preview (5 min)

[Brief framing, NOT a demo. "Given what you've shared, here's what we'd recommend evaluating..."]

### Next steps (3-5 min)

[Specific dated commitment script]

## Real-time note scaffold (have this open during the call)

M: [...]
E: [...]
D-criteria: [...]
D-process: [...]
P-paper: [...]
I: [...]
C-champion: [...]
C-competition: [...]

QUOTES: [...]
NAMES: [...]
NUMBERS: [...]
TIMELINE: [...]
RED FLAGS: [...]

## Recap email shell (send within 4 hours)

Subject: Recap, [your company] x [their company], [date]

[Name],

Quick recap from today.

Three things I heard:

1. [pain in their words, quoted]
2. [decision process / criteria summary]
3. [timeline / urgency anchor]

Next step: [specific commitment + date]

If I got anything wrong, please correct, easier to fix now than later.

[name]

## Post-call qualification verdict

After the call, fill this in 5 minutes:

- **Pain qualified?** Y / N (against rubric, specificity, measurability, currency, ownership, cost, past attempts)
- **MEDDPICC complete?** [list any blank fields]
- **Tier confirmed?** [Tier 1/2/3]
- **Verdict:** Move forward / Disqualify / Nurture
- **Next step + date:** [specific]
- **Confidence:** Low / Medium / High
```

---

## Common mistakes

Push back on these when you see them:

- **Talking too much.** Rep should talk 30%, prospect 70%. If you're talking more than them, you're not running discovery, you're running a pitch.
- **Pitching before understanding.** Prospects ask "what does your product do?", answer in one sentence and pivot back to discovery. The pitch comes after the pain is qualified.
- **Skipping decision-process questions because they feel awkward.** "Who signs the check" feels rude. Asking it 30 days later in deal review feels worse. Front-load these questions.
- **Accepting vague pain.** "We want to grow" is not pain. Force specificity. If they can't get specific, the pain isn't real enough to buy.
- **No "what happens if you do nothing."** This is the single most powerful discovery question. Reps skip it because the answer might be "nothing." That answer is the qualification, if they're fine doing nothing, the deal is fake.
- **Letting the prospect run the meeting.** They're not the rep. If they pull you off-track, redirect with "I'd love to come back to that, first I want to make sure I understand X."
- **No competition question.** Every deal has competition. Even if it's the status quo, ask. Reps who don't ask get blindsided in late stage.
- **No next step with a date.** "We'll be in touch" is a dead deal. The call doesn't end until there's a specific commitment from both sides on a specific date.
- **Skipping the recap email.** Or sending it 3 days later. The recap is part of the discovery, it confirms what you heard and locks in next steps. 4-hour SLA, no exceptions.
- **Treating MEDDPICC as post-call paperwork.** It's a live conversation framework. Fill it during the call. Anything blank by the end = a follow-up question, not a guess.

---

## How to use the artifact downstream

After discovery:

1. **Update CRM**, fill MEDDPICC fields, log next step, flag deal stage
2. **Brief the AE / founder** if you weren't the one running the call (or vice versa)
3. **Build the demo plan**, use the pain quotes and use case to drive `demo-script-builder`
4. **Loop in stakeholders**, if EB or Blocker was named but not on the call, plan the multi-thread strategy
5. **Forecast input**, confidence + timeline + MEDDPICC completeness feeds into `forecasting-and-pipeline-review`
6. **Update ICP signals**, if a Tier 1 prospect surfaces a pain you weren't expecting, that's potential ICP refinement (link back to `icp-builder`)

---

## A note on AI-assist

Recording tools (Gong, Fireflies, Chorus, Read.ai) are now standard. Use them, but don't outsource the structured fill to them.

Helpful uses:

- Pre-call: AI-generated brief from CRM data, LinkedIn, news
- During call: AI prompter suggesting follow-up questions in real-time (Read.ai does this; experimental)
- Post-call: transcript + AI summary as backup to your live notes
- Coaching: review your own talk-time ratio, question patterns, call-control moments

Where AI fails today:

- Reliable MEDDPICC structuring from raw transcript (the words are there but the framework fill is uneven)
- Reading nuance in the prospect's tone (urgency vs. politeness, real pain vs. performed pain)
- Catching what the prospect didn't say (silence is data, AI doesn't catch it well)

Run the live fill yourself. Use AI as a second pair of ears, not your only pair.

---

**Discovery is not about understanding the customer. It's about deciding whether they're worth pursuing, and being honest enough with both of you to disqualify when they're not.**

---

_Part of the StackSwap Operator Playbook. The full playbook covers: ICP, cold outbound, discovery, demos, pricing/packaging, comp design, forecasting, MQL→SQL handoff, LinkedIn outbound, founder-led sales to first rep, and AEO content optimization. → stackswap.ai/playbook_

Free forever. No email gate.