Feature Kickoff Workflow¶
Quick-start workflow for feature development
Feature Kickoff is a streamlined workflow for developing well-understood features where requirements are relatively clear and deep discovery isn't needed. It gets you from problem to launch efficiently.
Workflow Metadata¶
| Field | Value |
|---|---|
| Workflow | Feature Kickoff |
| Command | /workflow-feature-kickoff |
| Skills | define-problem-statement → define-hypothesis → deliver-prd → deliver-user-stories → deliver-launch-checklist |
| Phases Covered | Define, Deliver |
| Estimated Duration | 4-6 hours |
| Prerequisite Inputs | A feature idea or customer request |
| Final Output | Sprint-ready user stories with launch plan |
Overview¶
┌──────────────────────────────────────────────────────────────┐
│ │
│ problem-statement → hypothesis → prd → user-stories → launch│
│ │
└──────────────────────────────────────────────────────────────┘
graph LR
A["problem-statement"] --> B["hypothesis"]
B --> C["prd"]
C --> D["user-stories"]
D --> E["launch-checklist"]
This workflow assumes: - The problem space is understood - Users/customers are known - Solution direction is clear - Focus is on execution, not exploration
When to Use¶
Use Feature Kickoff when:
- Building incremental features on existing products
- Requirements are clear from customer requests or internal needs
- Team has context on the problem space
- Similar features have been built before
- Time-to-market is a priority
Don't use Feature Kickoff when:
- Entering new markets or user segments → Use Triple Diamond
- High uncertainty about problem or solution → Use Lean Startup
- Strategic pivots or major direction changes → Use Triple Diamond
- Complex stakeholder alignment needed → Add discovery phase
Core Sequence¶
Step 1: Problem Statement¶
Skill: define-problem-statement
Purpose: Ensure everyone agrees on what problem we're solving before diving into solutions.
Time: 30 minutes - 1 hour
Key Outputs: - Clear problem framing - User impact articulated - Success criteria defined - Scope boundaries set
Example output:
Mobile users abandon checkout 73% of the time because the 4-step process is too long. Success means reducing abandonment to under 60% within 3 months.
Step 2: Hypothesis¶
Skill: define-hypothesis
Purpose: State what we believe will solve the problem and how we'll know if we're right.
Time: 15-30 minutes
Key Outputs: - Testable hypothesis statement - Primary success metric - Secondary/guardrail metrics
Example output:
We believe that simplifying checkout to a single page for returning users will reduce abandonment from 73% to 55% as measured by checkout completion rate, without negatively impacting average order value.
Step 3: PRD¶
Skill: deliver-prd
Purpose: Specify what we're building in enough detail for engineering to execute.
Time: 1-2 hours
Key Outputs: - Requirements specification - Scope (in/out) - Technical considerations - Success metrics - Dependencies and risks
Tip: For Feature Kickoff, keep the PRD concise. Focus on what's needed to build and ship, not exhaustive documentation.
Step 4: User Stories¶
Skill: deliver-user-stories
Purpose: Break the PRD into implementable, estimable stories for sprint planning.
Time: 30 minutes - 1 hour
Key Outputs: - User stories with acceptance criteria - INVEST-compliant stories - Story points/estimates - Dependencies identified
Story format:
As a [persona], I want [action] so that [benefit]
Acceptance Criteria:
- Given [context], when [action], then [result]
Step 5: Launch Checklist¶
Skill: deliver-launch-checklist
Purpose: Ensure nothing is forgotten before shipping.
Time: 30 minutes to create, ongoing to complete
Key Outputs: - Cross-functional checklist - Owner assignments - Go/no-go criteria - Rollback plan
Complete Feature Kickoff¶
Day 1: Problem Statement (team alignment)
↓
Day 1: Hypothesis (testable assumption)
↓
Day 1-2: PRD (specification)
↓
Day 2: User Stories (sprint-ready tickets)
↓
Sprint: Development
↓
Launch: Launch Checklist (pre-launch validation)
Total time to kickoff: 4-6 hours (can be done in one day)
Optional Extensions¶
Add these skills when the situation calls for more depth:
Before PRD¶
| Add When | Skill |
|---|---|
| Complex UX decisions | develop-design-rationale |
| Technical uncertainty | develop-spike-summary |
| Architecture impact | develop-adr |
During Development¶
| Add When | Skill |
|---|---|
| Complex feature | deliver-edge-cases |
| Needs measurement | measure-instrumentation-spec |
After Launch¶
| Add When | Skill |
|---|---|
| A/B testing | measure-experiment-design |
| Team reflection | iterate-retrospective |
| Communication | deliver-release-notes |
Example: Feature Kickoff in Action¶
Context¶
E-commerce platform adding "Save for Later" feature to shopping cart.
Step 1: Problem Statement¶
Problem: Users often leave items in cart that they're not ready to buy today, then forget about them. Cart clutter leads to decision fatigue and abandoned carts.
User Impact: 35% of cart items sit for 7+ days untouched. Users report "cart anxiety" in interviews.
Success Criteria: - Reduce untouched cart items from 35% to 20% - Increase cart-to-purchase conversion by 5%
Step 2: Hypothesis¶
We believe that adding a "Save for Later" section below the cart will help users organize their shopping intentions, reducing cart anxiety, because it separates "ready to buy" from "considering" items.
Primary metric: Cart-to-purchase conversion rate Guardrail: Average order value (should not decrease)
Step 3: PRD Summary¶
In Scope: - "Save for Later" button on each cart item - Separate "Saved Items" section below cart - Move back to cart functionality - Persisted for logged-in users
Out of Scope: - Guest user persistence - Wishlists (separate feature) - Social sharing of saved items
Step 4: User Stories (excerpt)¶
US-001: Save item for later
As a shopper, I want to move an item from my cart to "Saved for Later" so that I can keep it visible without cluttering my active cart.
Acceptance Criteria: - Given I have items in my cart - When I click "Save for Later" on an item - Then the item moves to the Saved Items section - And the cart total updates to exclude that item
US-002: Move saved item to cart
As a shopper, I want to move a saved item back to my cart so that I can purchase it when I'm ready.
Step 5: Launch Checklist (excerpt)¶
- Feature complete in staging
- QA sign-off
- Mobile responsive verified
- Analytics events implemented
- Support team briefed
- A/B experiment configured (10% rollout)
- Rollback plan documented
Tips for Effective Kickoffs¶
Do¶
- Time-box each step to maintain momentum
- Include engineering in PRD creation
- Define "done" before starting
- Keep documents lightweight for simple features
Don't¶
- Skip the problem statement ("we know what to build")
- Over-document for small features
- Forget instrumentation until launch
- Treat kickoff as a handoff—it's a collaboration
Quick Reference¶
| Step | Skill | Time | Output |
|---|---|---|---|
| 1 | problem-statement | 30-60 min | Problem framing doc |
| 2 | hypothesis | 15-30 min | Testable hypothesis |
| 3 | prd | 1-2 hrs | Requirements spec |
| 4 | user-stories | 30-60 min | Sprint-ready stories |
| 5 | launch-checklist | 30 min | Pre-launch checklist |
Total: ~4-6 hours to kick off a feature
Quality Checklist¶
Before considering this workflow complete, verify:
- Problem statement has measurable success criteria
- Hypothesis is specific and falsifiable
- PRD scope is clear (in/out)
- User stories have Given/When/Then acceptance criteria
- Launch checklist has owners for every item
See Also¶
- Triple Diamond Workflow — For comprehensive product development
- Lean Startup Workflow — For rapid hypothesis validation
Part of PM-Skills — Open source Product Management skills for AI agents