# Feature Decision Workflow (Go/No-Go) ## Overview This workflow provides a structured framework for making major product decisions. It ensures decisions are made deliberately with clear rationale, not reactively or politically. ## When to Use This Workflow Use this for decisions that are: - **Significant:** More than a few days of work - **Strategic:** Affects product direction or positioning - **Irreversible:** Hard to undo once started - **Resource-intensive:** Requires significant investment ## Decision Options | Decision | Meaning | When to Use | |----------|---------|-------------| | **GO** | Commit resources, proceed | High confidence, strategic fit, acceptable risk | | **NO-GO** | Reject, do not proceed | Poor fit, high risk, better alternatives exist | | **DEFER** | Not now, revisit later | Need more information, timing not right | --- ## Workflow Stages ### Stage 1: Understand the Initiative **Objective:** Ensure complete clarity on what is being proposed. **Questions to answer:** 1. **What exactly is being proposed?** - Feature/product description in 2-3 sentences - What problem does it solve? - Who requested this? (Customer, team, competitive pressure) 2. **What does success look like?** - How would we know if this succeeded? - What metrics would change? 3. **What's the scope?** - MVP scope vs. full vision - Timeline estimate - Resource requirements **Clarity Check:** - Can you explain this to someone unfamiliar in 30 seconds? - Is there ambiguity about what we'd actually build? - Do we have clear acceptance criteria? **Output:** Clear initiative description --- ### Stage 2: Strategic Fit Assessment **Objective:** Evaluate alignment with vision and current priorities. **Strategic Alignment Scorecard:** | Question | Score (1-5) | Notes | |----------|-------------|-------| | Does this align with our vision? | | | | Does this serve our target customer? | | | | Does this strengthen our differentiation? | | | | Does this align with current priorities? | | | | Would this help or hurt focus? | | | **Scoring:** - 5: Strongly supports - 3: Neutral - 1: Works against **Priority Conflict Check:** - Does this compete with current priorities for resources? - Would this delay existing commitments? - Is this more important than what we're already doing? **Strategic Fit Score:** [Sum / 25] = X% | Score | Interpretation | |-------|----------------| | 80%+ | Strong strategic fit | | 60-79% | Moderate fit, consider tradeoffs | | <60% | Weak fit, needs strong justification | **Output:** Strategic fit assessment --- ### Stage 3: Impact Analysis **Objective:** Assess the potential positive and negative outcomes. **Positive Impact Assessment:** | Impact Area | Expected Impact | Confidence | Evidence | |-------------|-----------------|------------|----------| | Revenue | [High/Med/Low] | [High/Med/Low] | [What evidence?] | | User acquisition | | | | | User retention | | | | | User satisfaction | | | | | Competitive position | | | | | Technical capability | | | | **Negative Impact Assessment:** | Risk | Potential Impact | Likelihood | Mitigation | |------|------------------|------------|------------| | Technical debt | | | | | Scope creep | | | | | Customer confusion | | | | | Team burnout | | | | | Opportunity cost | | | | **Opportunity Cost:** > "What are we NOT doing by doing this?" - [Alternative 1 we won't pursue] - [Alternative 2 we won't pursue] - [Feature/fix that gets delayed] **Output:** Impact analysis summary --- ### Stage 4: Risk Assessment **Objective:** Identify and evaluate risks. **Risk Categories:** **1. Execution Risk** - Do we have the skills to build this? - Is the timeline realistic? - Are there technical unknowns? **2. Market Risk** - Will customers actually want this? - Is the timing right? - Will competitors respond? **3. Business Risk** - Can we afford this investment? - What if it fails? - Are there regulatory concerns? **Risk Matrix:** | Risk | Likelihood (1-5) | Impact (1-5) | Score | Mitigation | |------|------------------|--------------|-------|------------| | [Risk 1] | | | | | | [Risk 2] | | | | | | [Risk 3] | | | | | **Risk Tolerance Check:** - What's the worst case scenario? - Can we survive the worst case? - Is this a reversible or irreversible decision? **Reversibility Assessment:** - **Easily reversible:** Low-risk, can experiment - **Partially reversible:** Some sunk cost but can pivot - **Irreversible:** High commitment, proceed carefully **Output:** Risk assessment with mitigations --- ### Stage 5: Make Decision **Objective:** Decide: GO, NO-GO, or DEFER. **Decision Framework:** ``` Strategic Fit High Low ┌─────────┬─────────┐ High │ GO │ DEFER/ │ Impact │ │ NO-GO │ Low │ DEFER │ NO-GO │ └─────────┴─────────┘ ``` **GO Criteria (all must be true):** - [ ] Strategic fit score ≥ 60% - [ ] Expected impact justifies investment - [ ] Risks are acceptable and mitigated - [ ] Resources are available - [ ] Team has capacity and capability - [ ] Not conflicting with higher priorities **NO-GO Criteria (any is sufficient):** - [ ] Poor strategic fit (<50%) - [ ] Unacceptable or unmitigatable risks - [ ] Better alternatives exist - [ ] Insufficient resources or capability - [ ] Would derail critical priorities **DEFER Criteria:** - [ ] Good idea but timing is wrong - [ ] Need more information to decide - [ ] Waiting on external dependencies - [ ] Current priorities must complete first --- ### Stage 6: Document & Communicate **Objective:** Record the decision and next steps. **Decision Record:** ```yaml decision: id: "[DECISION-XXX]" date: "[Date]" initiative: "[Name]" decision: "[GO | NO-GO | DEFER]" rationale: | [2-3 sentences explaining why] # If GO: next_steps: - "[Action 1]" - "[Action 2]" owner: "[Name]" deadline: "[Date]" success_criteria: - "[Criterion 1]" - "[Criterion 2]" # If NO-GO: alternatives_considered: - "[Alternative 1]" revisit_conditions: "[What would change our mind]" # If DEFER: revisit_date: "[Date]" information_needed: - "[What we need to know]" ``` **Communication:** | Audience | What to Share | |----------|---------------| | Team | Decision + rationale + impact on them | | Stakeholders | Decision + high-level rationale | | Requestor | Decision + detailed rationale | --- ## Events Published Based on decision: **If GO:** ```yaml type: decision.go payload: initiative: "[Name]" rationale: "[Why]" owner: "[Name]" deadline: "[Date]" success_criteria: [list] ``` **If NO-GO:** ```yaml type: decision.no-go payload: initiative: "[Name]" rationale: "[Why]" alternatives_considered: [list] revisit_conditions: "[Conditions]" ``` **If DEFER:** ```yaml type: decision.defer payload: initiative: "[Name]" reason: "[Why]" revisit_date: "[Date]" information_needed: [list] ``` --- ## Downstream Effects | Decision | Other Agents | |----------|--------------| | GO | Compliance reviews, Growth plans measurement, UX plans design | | NO-GO | No downstream actions | | DEFER | Calendar reminder set for revisit | --- ## Anti-Patterns to Avoid 1. **HIPPO Decisions** - Don't decide just because the Highest Paid Person's Opinion says so 2. **Analysis Paralysis** - Don't defer indefinitely; set a deadline 3. **Sunk Cost Fallacy** - Past investment doesn't justify future investment 4. **Scope Creep in Disguise** - One feature hiding many 5. **Decision by Committee** - One person must own the decision