roadmap-communication

What this skill does

# Roadmap Communication Expert

## Overview

A single roadmap cannot serve every audience. Executives need confidence and strategic framing. Customers need plausible promises with credible timelines. Internal teams need detail, dependency awareness, and honest risk. When PMs publish one roadmap to all three, one audience is always wrong -- either the executives are bored, customers feel misled, or engineers feel managed.

This skill produces three variants of the same underlying roadmap, each tailored to its audience. The underlying source of truth -- the roadmap data -- stays single. The framing, the level of detail, the risk language, and the visualization all change. The variants share names and identifiers so a customer reference can be traced back through the internal variant to the actual ticket.

The structural backbone is **Now / Next / Later** (popularized by Janna Bastow at ProdPad and now used widely), which Marty Cagan calls "right-sizing the roadmap" -- forecast confidence decreases with time horizon, so commit explicitly to "Now" items, name "Next" items by theme, and keep "Later" deliberately fuzzy. Each variant exposes a different slice of that backbone.

### When to Use

- **Quarterly roadmap publication** -- the regular cadence where you communicate to multiple audiences.
- **Pre-board meeting** -- the executive variant feeds the board packet.
- **Annual customer conference** -- the customer variant becomes the conference roadmap session.
- **Engineering kickoff** -- the internal variant grounds sprint planning across multiple teams.
- **Sales enablement** -- the customer variant feeds sales reps' "what's coming" pitch.

### When NOT to Use

- For a single-team backlog with one stakeholder (one roadmap suffices).
- For tactical sprint planning (use `../scrum-master/` outputs).
- For incident communication or in-flight project status (use `status-update-generator/`).

## The three variants

| Variant | Audience | Time horizon detail | Risk framing | Format |
|---------|----------|---------------------|--------------|--------|
| **Executive** | Board, exec team, investors | Outcome-led; quarterly | Confidence levels (high/med/low) | 1-page narrative + simple visual |
| **Customer** | Customers, partners, sales prospects | Theme-led; "this year / later this year / next year" | No risk language; "exploring", "in development", "shipping soon" | Themes + benefit statements, no dates beyond current quarter |
| **Internal** | Engineering, design, support, account teams | Feature-led with dependencies | Explicit risks with mitigations | Detailed table with owners, dependencies, links to PRDs |

### What changes across variants

| Element | Executive | Customer | Internal |
|---------|-----------|----------|----------|
| **Granularity** | Outcomes (3-7 items) | Themes (5-10 items) | Features (15-50 items) |
| **Time horizon** | This quarter / next quarter / H2 | Now / Next / Later (themes only) | Sprint-by-sprint for "Now"; quarter for "Next" |
| **Dates** | Quarter only | Quarter only for "Now"; "later this year" for "Next" | Specific sprint and release dates for "Now" |
| **Risk** | Confidence per outcome | None (use language like "exploring" / "shipping soon") | Explicit: tigers, mitigations, dependencies |
| **Owner** | Exec sponsor per outcome | None | PM, EM, design lead per feature |
| **Metric** | Outcome metric + target | Benefit statement | KR + leading indicator |
| **What's NOT here** | Implementation detail, risk discussion | Risk, dependencies, internal team names | Marketing language, exec framing |
| **Visual** | Now/Next/Later 3-column or Gantt-summary | Themed cards or simple Now/Next/Later | Detailed table or Gantt |

## The "right-size the roadmap" principle

Marty Cagan's observation: **forecast confidence drops sharply with time horizon**. A 2-week commitment can be 90% confident. A 1-quarter commitment is maybe 70%. A 1-year commitment is 30% at best. Pretending otherwise (by promising features 9 months out) trains customers to disbelieve every roadmap you publish.

The Now / Next / Later structure encodes this honesty:

- **Now (this quarter):** Specific features with committed dates. Build is in flight or starting imminently. High confidence.
- **Next (next 1-2 quarters):** Themes with named outcomes, but specific features are not committed. Medium confidence.
- **Later (beyond):** Directions, hypotheses, big bets. Deliberately fuzzy. Low confidence by design.

Each audience needs this honesty calibrated:

- **Executives** want confidence ratings on each item (so they can plan).
- **Customers** want forward visibility without firm dates (so they can buy with confidence in the direction).
- **Internal teams** want explicit dependencies and risks (so they can execute).

## Outcome-based vs feature-based roadmaps

A second axis: **outcomes vs features**. Cagan and Torres both argue for outcome-based roadmaps -- the roadmap promises a customer outcome ("reduce onboarding from 14 days to 3 days"), not a feature ("ship guided onboarding wizard"). Outcomes survive scope changes; features become commitments that bind you to wrong solutions.

| Variant | Default orientation |
|---------|---------------------|
| Executive | Outcome-based always |
| Customer | Mix: themes (outcomes) for Next/Later; specific features for Now (when they are committed) |
| Internal | Feature-based for execution, but each feature must trace to an outcome |

The translation from feature to outcome is the highest-leverage move a PM can make in roadmap communication. "We are shipping X" is a feature. "Customers can now achieve Y" is an outcome.

## Variant 1: Executive Roadmap

**Purpose:** Build confidence that the team is solving the right problems and is on track.

**Length:** 1 page (slide or memo).

**Structure:**

1. **Header:** Quarter, sponsor, the one-sentence strategy this roadmap serves.
2. **Now (this quarter) -- 3-5 outcomes:** Each outcome named, with metric target, confidence rating (High/Med/Low), and lead PM.
3. **Next (next quarter) -- 3-5 outcomes:** Theme-level. No specific features. Confidence is implicit (Medium).
4. **Later (H2 directions) -- 2-3 big bets:** One line each. Deliberately fuzzy.
5. **What we are explicitly NOT doing this period:** 2-3 items. Builds credibility by showing tradeoffs.

**Voice:** Direct, plain, confident-but-honest. No marketing adjectives. No "leveraging" or "synergizing." Bezos memo style.

**Visualization:** Now/Next/Later 3-column or a slim Gantt of the 3-5 Now-quarter outcomes. Avoid swim-lane explosions.

## Variant 2: Customer Roadmap

**Purpose:** Help customers and prospects make confident purchase or expansion decisions; help sales sell direction without overcommitting.

**Length:** 2-5 themed cards (web page or 1-2 slides).

**Structure:**

1. **Header:** "What's coming in [year]" -- positions roadmap as a window, not a contract.
2. **Now (this quarter) -- 3-5 themes with named features:** "Workflow automation: Connect tasks to your CRM with built-in triggers. Available this quarter."
3. **Next -- 3-5 themes, features unnamed:** "Reporting: deeper drill-down across teams and saved-view collaboration. Coming later this year."
4. **Exploring (Later) -- 2-3 directions:** "Exploring: AI-assisted task creation from meeting notes. Early experiments underway."

**Voice:** Benefits-led, customer-perspective. No internal team names. No release versions. No risk language. Use directional language ("exploring", "in development", "shipping soon") instead of dates beyond the current quarter.

**Visualization:** Themed cards (Notion / Productboard / Cards format). Avoid Gantt -- it reads as a commitment.

**Important caveats to repeat on the page:**

- "This roadmap is directional and may change as we learn from customers."
- "Items in 'Exploring' are not commitments."
- "Dates are based on current planning and may shift."

## Variant 3: Internal Roadmap

**Purpose:** Coordinate execution across teams;