Walk in with the questions that qualify.
Tailored discovery questions sequenced problem to impact to next step, each mapped to your qualification framework.
Inside: The question set, the impact quantifiers, and the field each question maps to.
Install it in one line, or paste it in.
~/.claude/skills/ and runs automatically when it is relevant.Connect your context. Set it to your motion.
it reads the open opportunity and the qualification fields already filled, so it only asks what is still missing.
it can pull prior-call notes so you do not re-ask what you already learned.
pulls the role and context so the questions land in the prospect's world, not a generic one.
This was built for B2B AEs running staged discovery. Set these to your motion:
Qualify with any framework you like. The skill maps its questions to your fields, not anyone else's letters.
| Set this | What it is | Default / Example |
|---|---|---|
| PERSONA | who is on the call | economic buyerchampionend user |
| PROBLEM | the hypothesis you are testing | stated as a business pain |
| FRAMEWORK | how you qualify | MEDDPICCMEDDICBANTyour own |
| DEPTH | how many questions per stage | 2 to 3 (re-tune to call length) |
| NEXT_STEP | the exit you want to earn | a scoped next meeting or action |
| CRM | your CRM connector | a CRM you already run |
| QUALIFICATION fields | your framework's fields in the CRM | the fields you actually track |
Everything the skill does, in full.
Takes a persona and the problem you think they have, and returns a sequenced set of discovery questions that move from surfacing the problem, to sizing its impact, to agreeing a next step. Each question is mapped to the qualification field it fills, so a good call also fills your framework instead of leaving you to backfill it later.
- 1Problem to impact to next step
Questions run in three passes. First surface and confirm the problem in the prospect's words. Then quantify what it costs them, in time, money, or risk. Then agree the next step. Impact questions before you have confirmed a problem land flat; this order keeps them earned.
- 2Map every question to a field
Each question is tagged with the framework field it fills, Metrics, Economic Buyer, Decision Criteria, Pain, and so on for your framework. A discovery call is not done because time ran out; it is done when the fields that gate the next stage are filled.
- 3Layered, not interrogation
Questions layer: an open question, then a follow-up that goes one level deeper on the answer. The plan gives you the follow-up so you are not caught flat when they say "it's a problem." No rapid-fire checklist feel.
- 4Impact quantification
At least one question forces a number or a consequence, "what does that cost you a quarter," "what happens if this is still true next year." A pain with no size does not survive a deal review. Any example figure is illustrative, drawn from the prospect, never asserted by you.
- 5Earn the next step
The plan ends by scripting the transition to the next step, tied to what they just told you, so the ask feels like their logic, not your quota.
- Every question maps to a named framework field. No orphan questions.
- Impact questions come after a confirmed problem, never before.
- At least one question forces a quantified consequence.
- The next-step ask references the prospect's own stated pain, not a generic close.
DISCOVERY PLAN · economic buyer · framework: MEDDPICC
Pass 1 - Problem (fills: Pain, Metrics)
Q: How is [process] handled today, start to finish?
follow-up: Where does it break most often?
Q: What made this worth looking at now?
Pass 2 - Impact (fills: Metrics, Economic Buyer)
Q: When it breaks, what does that cost, roughly, per quarter?
Q: Who else feels that cost? Who signs off on fixing it?
Pass 3 - Next step (fills: Decision Process, Champion)
Q: If we showed this solved, what is your process to move on it?
Ask: "Based on the [cost] you mentioned, worth a working session
with [role] next week to map it?"
Where an operator takes this next.
The read is step one. Here is where an operator takes it once the manual version proves out.
A good call should fill the framework while you talk, not after.
Connect Salesforce so the question set reads the open opportunity's qualification fields first and only asks what is still blank.
Connect a meeting tool so post-call, the answers map straight back to the framework fields on the deal record.
Connect Clay or an enrichment tool so questions land in the prospect's actual role and situation, not a generic persona template.
One skill is the on-ramp.
A single skill does one job. Chained into a playbook, or run as a full build, it becomes a system. Here is where this one plugs in.