Build roadmaps that communicate strategic intent without pretending to predict the future. This skill covers outcome-based sequencing, strategic-bet management, and portfolio-level choices — not feature-level prioritization or project scheduling.
Core Principles
-
Roadmaps are not project schedules. A roadmap communicates strategic direction and sequencing intent. It does not commit to dates by default. The default view is Now/Next/Later; dates appear only when explicitly labeled as commitments.
-
Distinguish four categories in every artifact. Every roadmap output separates evidence (observed data), assumptions (beliefs not yet validated), commitments (decisions with consequences for reversal), and options (alternatives under consideration).
-
Bets, not plans. Each roadmap item is a strategic bet — an investment with an expected outcome, a confidence level, explicit dependencies, and pre-defined continue/pause/kill/revisit criteria. A roadmap is a portfolio of bets, not a to-do list.
-
Portfolio thinking scales up and down. The same bet-record and roadmap structure works for a single product and for a portfolio of products.
Loading Guide
| File | Load when |
|---|---|
| references/outcome-roadmapping.md | Building or reviewing a Now/Next/Later outcome roadmap |
| references/strategic-bets.md | Defining, evaluating, or comparing strategic bets |
| references/dependency-confidence-view.md | Mapping cross-bet dependencies or calibrating confidence |
| references/scenario-comparison.md | Comparing alternative futures or building scenario plans |
| references/roadmap-review-cadence.md | Designing or running a roadmap review process |
| references/discovery-brief.md | Understanding ownership boundaries and routing rules |
| templates/outcome-roadmap.md | Fillable Now/Next/Later roadmap template |
| templates/bet-record.md | Fillable strategic-bet record |
| templates/roadmap-review-record.md | Recording a roadmap review session |
Working Method
1. Start with outcomes, not features
Define outcomes as measurable "from → to" statements. Route evidence to product-analytics-and-measurement.
2. Map the Now/Next/Later landscape
Now = committed (WIP-limited), Next = validated/provisional, Later = options only. Dates only when labeled as commitments.
3. Build bet records
Every bet: outcome hypothesis, evidence, confidence (H/M/L with rationale), risk, dependencies, capacity, continue/pause/kill/revisit criteria.
4. Map dependencies and confidence
Blocking, enabling, shared-capacity, external dependencies. Effective confidence bounded by dependency health.
5. Compare scenarios
Base case + alternative scenarios for consequential decisions. Identify robust, conditional, fragile bets.
6. Define review cadence
Monthly for Now, quarterly for Next/Later. Record decisions with rationale.
7. Communicate to stakeholders
Executive narrative, team view, external/customer view — each tailored.
Routing Table
| When the task involves... | Route to... |
|---|---|
| RICE scoring, MoSCoW prioritization, OST workflow | product-methodology — canonical tactical owner |
| Kano model, OST strategic framing, competitive positioning | product-strategy — strategic frameworks |
| Capital allocation, M&A portfolio choices | strategy-frameworks — corporate strategy |
| Delivery flow, WIP limits, throughput, cycle time | kanban-guru — execution-level flow |
| Evidence, metrics, analytics instrumentation | product-analytics-and-measurement — analytics layer (prose reference) |
| Implementation planning for approved bets | implementation-planning — handoff target (prose reference) |
| Experiment design for low-confidence bets | product-experimentation — downstream consumer (prose reference) |
| Adoption measurement for shipped bets | product-adoption — downstream consumer (prose reference) |
| Lifecycle learning from bet outcomes | product-lifecycle-learning — downstream consumer (prose reference) |
When Not to Use
- For tactical prioritization mechanics (RICE scores, MoSCoW buckets, Kano categorization, OST trees): route to
product-methodologyorproduct-strategy. This skill references these frameworks only to describe where they fit, never to re-explain formulas or mechanics. - For cross-project program coordination (integrated dependencies, shared capacity, program benefits, or transformation delivery): route to technical-program-management. This skill sequences product bets; it does not coordinate their technical program execution.
- For project scheduling, Gantt charts, or milestone tracking: route to technical-project-management for ongoing project control; use implementation-planning for detailed planning of approved work. This skill owns outcome roadmaps.
- For delivery-flow management: route to
kanban-guru. - For corporate capital allocation and M&A portfolio decisions: route to
strategy-frameworks. - For shaping a single bet (setting its appetite, narrowing it into a bounded
problem, writing its pitch): route to
product-shaping. This skill sequences bets across cycles at the portfolio level; shaping packages one bet before it enters a roadmap — "strategic bet" here means a roadmap entry with sequencing criteria, not the shaped single-project commitment thatproduct-shapingproduces.
Evidence, Assumptions, Commitments, and Options
Every roadmap artifact distinguishes these four categories:
| Category | Definition | Example |
|---|---|---|
| Evidence | Observed data, validated learning, experiment results | "A/B test showed 12% lift in retention (p < 0.05, n=4,200)" |
| Assumptions | Beliefs not yet validated | "We assume users grant notification permissions at ~40%" |
| Commitments | Decisions with defined consequences for reversal | "Q3: 8 engineer-weeks for payments migration" |
| Options | Alternatives under consideration | "Build in-house vs. integrate third-party" |
An artifact that mixes assumptions into evidence or presents options as commitments is not a valid output of this skill.