Ask better discovery questions
Sales discovery questions is an Agent Skill: a written method your AI applies on demand, in Zealos or in any client that reads SKILL.md.
Sales discovery questions is an Agent Skill: a written method your AI applies on demand, in Zealos or in any client that reads SKILL.md.
Good discovery asks about what happened rather than what the buyer wants, because people are reliable historians of their own problems and unreliable architects of their own solutions.
| name | discovery-questions |
|---|---|
| description | The discovery question bank and which questions belong at which stage. Use when preparing for a first call, writing a call plan, or reviewing whether a discovery call actually discovered anything. |
| allowed-tools | none — this skill adds judgement, not reach |
| compatibility | Zealos, Claude Code, Cursor, Codex and any client that reads SKILL.md |
| license | Free to use and edit |
--- name: discovery-questions description: The discovery question bank and which questions belong at which stage. Use when preparing for a first call, writing a call plan, or reviewing whether a discovery call actually discovered anything. --- # Discovery questions Ask about what happened, not about what they want. People are reliable historians and unreliable architects. ## Opening (situation) - "Walk me through how this works today, start to finish." - "Who touches it besides you?" - "How long has it worked this way?" ## Pain (the part most calls skip) - "When did this last go wrong? What happened?" - "What did that cost — time, money, or a customer?" - "What have you already tried?" *(Reveals budget, urgency and internal politics in one answer.)* - "What happens if this is still true in six months?" ## Consequence - "Who else feels it when this breaks?" - "Is anyone measured on this?" - "What is this competing with for budget?" ## Process - "If you decided this was worth doing, what happens next?" - "Who has told you no before on something like this?" ## Rules - One question at a time. Stacked questions get the easy one answered. - After an answer, count to two before speaking. The second half of the answer is the useful half. - Never ask a question whose answer is in the CRM. - If you leave without a number and a date, it was not a discovery call.
The qualification call stops being a checklist recited at the prospect and becomes a conversation that finds out whether anything is actually broken.
You stop asking questions whose answers are already in the CRM, which is what buyers mean when they say a call felt like a waste of time.
New reps get the question bank on day one rather than assembling one over their first two quarters.
It asks about something that already happened. When did this last go wrong outperforms what are you looking for, because the first has an answer and the second has an opinion.
Fewer than you think, with longer silences. The second half of an answer is usually the useful half and it only arrives if you wait.
No, and it should not. It can prepare the questions, notice which you skipped, and capture what was said — the conversation is still yours.
Post-call automation turns a recorded conversation into the three things that outlive it: the commitments each side made, the fields that changed, and the follow-up that has to go out today.
/call-recap-formatA pre-call brief is worth writing only if it changes what you ask: who is in the room, what they are measured on, what happened last time, and the one question you cannot afford to leave without answering.
/discovery-questionsAn AI SDR agent is only as good as the definition of a good customer it is working from, which is why the fit criteria belong in a written skill rather than buried in a prompt nobody can find.
/our-icpAI email triage is worth having only when it sorts by consequence rather than by recency: what blocks a deal, what someone is waiting on you for, and what can be read at the end of the week.
/email-rulesCRM data hygiene survives AI only if one rule is written down first: a field left blank is a known unknown, while a field filled in by inference is an unknown unknown that will be reported on as fact.
/crm-field-conventionsAccount research earns its time only when it changes the first sentence you write, which means looking for what recently changed rather than assembling a profile anyone could have found.
/our-icp