evalops

diligence-sprint

0
0
# Install this skill:
npx skills add evalops/open-associate-skills --skill "diligence-sprint"

Install specific skill from multi-skill repository

# Description

A time-boxed diligence sprint (3-7 days) with daily deliverables and decision gates. Burns down top 4 risks and produces IC-ready evidence. Use when a deal is live and you need speed without losing rigor.

# SKILL.md


name: diligence-sprint
description: "A time-boxed diligence sprint (3-7 days) with daily deliverables and decision gates. Burns down top 4 risks and produces IC-ready evidence. Use when a deal is live and you need speed without losing rigor."
license: Proprietary
compatibility: Works offline; improved with calls and data access; Salesforce logging recommended.
metadata:
author: evalops
version: "0.2"


Diligence sprint

When to use

Use this compound skill when:
- A deal is live and you have a hard decision deadline
- You need a clear risk burn-down plan and daily deliverables
- You want to coordinate multiple diligence workstreams with gates

Inputs you should request (only if missing)

  • Deadline (term sheet / IC date)
  • Intended role (lead/follow) and check size
  • Top 3 "must be true" claims from first meeting
  • Customer access constraints
  • Sprint length: 3-day, 5-day, or 7-day

Outputs you must produce

  • Day-by-day plan with owners and deliverables
  • Ranked risk register (MAX 4 risks)
  • Evidence pack with anti-confirmation
  • Decision at each gate: kill, continue, or escalate
  • Memo-ready findings or explicit kill decision

Every sprint ends with one of two outcomes:
1. Partner-ready memo + IC scheduled
2. Explicit kill with logged reason + "what would change our mind" + recheck date

5-day default sprint plan

Day 0: Setup (2-4 hours)

Deliverables:
- Aligned on top 4 risks + fastest tests
- Owners assigned and calls scheduled
- Salesforce Tasks created for each workstream

Day 0 Gate: Can we test the top 4 risks in the time available? If not, scope down or extend timeline.

Day 1: Customer diligence

Deliverables:
- 3 buyer calls completed
- 2 user calls completed
- 1 churned/lost deal call scheduled or completed (REQUIRED)
- Customer summary written

Day 1 Gate: Did customers validate or invalidate the core thesis? Any kill signals?

Risk Before Day 1 After Day 1 Change
1
2

Decision: Kill / Continue / Escalate

Day 2: Product diligence

Deliverables:
- Hands-on demo completed
- Deployment reality assessed
- Integration blockers identified
- Time-to-value documented

Day 2 Gate: Does the product deliver on the promise? Any kill signals?

Risk Before Day 2 After Day 2 Change
1
2

Decision: Kill / Continue / Escalate

Day 3: GTM + metrics diligence

Deliverables:
- Pipeline review completed
- Unit economics validated by segment
- Cycle time and pricing confirmed
- Competitor/alternative user call completed (REQUIRED)

Day 3 Gate: Is this repeatable? Are the economics real? Any kill signals?

Risk Before Day 3 After Day 3 Change
1
2

Decision: Kill / Continue / Escalate

Day 4: Team + competition

Deliverables:
- Team learning rate assessed
- Decision rights clarity confirmed
- Competitive map updated
- "Incumbent build" threat evaluated

Day 4 Gate: Is this team credible? Can they win vs competition? Any kill signals?

Risk Before Day 4 After Day 4 Change
1
2

Decision: Kill / Continue / Escalate

Day 5: Synthesis + decision

Deliverables:
- All evidence synthesized
- Memo sections drafted (market, product, traction, GTM, team, risks)
- Recommendation written (first line, no hedging)
- Steelman "best argument against" written
- IC readout prepared

Final Gate: One of two outcomes:
1. ADVANCE: Partner-ready memo + IC scheduled
2. KILL: Pass logged with reason + "what would change our mind" + recheck date

3-day sprint (urgent deal)

Day 0: Setup + risk alignment (2 hours)

  • Align on top 2-3 risks (fewer risks, faster tests)
  • Schedule all calls for Days 1-2

Gate: Are the tests achievable in 3 days?

Day 1: Customer + product

  • 3-5 customer calls (buyers + users)
  • Hands-on product eval
  • Churned/lost deal call (REQUIRED)

Gate at end of Day 1: Kill or continue? No "need more time."

Day 2: GTM + team + competition

  • Pipeline review
  • Team assessment
  • Competitor validation

Gate at end of Day 2: Kill or continue? All top risks tested?

Day 3: Synthesis + decision

  • Evidence synthesis
  • Memo draft
  • IC decision

Final gate: Memo or kill. No limbo.

7-day sprint (deeper diligence)

Same structure as 5-day, with:
- Days 1-2: Customer diligence (more calls, deeper coverage)
- Day 3: Product diligence
- Days 4-5: GTM + metrics + competition
- Day 6: Team + synthesis
- Day 7: Memo + decision

Decision gates (mandatory checkpoints)

At each gate, you must answer:
1. Kill? Is there evidence that this deal should not proceed?
2. Continue? Is there enough positive signal to invest more time?
3. Escalate? Does a partner need to weigh in before Day X?

Kill criteria (stop immediately if any are true):
- Customer calls reveal fundamental misunderstanding of ICP
- Churned customer call reveals systemic product/service issue
- Metrics don't support the stated traction
- Competitive position is weaker than pitched
- Team shows inability to change mind with evidence

Continue criteria:
- Evidence supports core thesis
- Risks are testable and manageable
- No kill signals triggered

Escalate criteria:
- Risk requires partner relationship/expertise
- Deal dynamics require partner involvement
- Unclear whether to kill or continue

Anti-confirmation requirements (MANDATORY)

Every sprint must include:
1. Churned customer or lost deal call - What caused them to leave/not buy?
2. Competitor/alternative user - Why do they use something else?

If these calls are impossible, document why and flag the gap prominently in the memo.

Guardrails

  • If you can't get a failed customer reference, discount retention claims.
  • If the deadline compresses: do fewer risks, not more shallow work.
  • Never extend a sprint without a partner decision. Time-box or kill.
  • Every day must produce a decision-relevant deliverable.

Salesforce logging

  • Create Tasks for each diligence workstream with owner and due date
  • Log each call as Activity with notes
  • Update Opportunity stage at each gate
  • Attach final evidence pack and memo to Opportunity

# Supported AI Coding Agents

This skill is compatible with the SKILL.md standard and works with all major AI coding agents:

Learn more about the SKILL.md standard and how to use these skills with your preferred AI coding agent.