When to use: Execution planning with engineering. Identifying scheduling conflicts. NOT good for communicating externally (creates false precision expectations).
Score each initiative on four dimensions, then calculate RICE = (Reach x Impact x Confidence) / Effort
Reach: How many users/customers will this affect in a given time period? Use concrete numbers (e.g., “500 users per quarter”).
Impact: How much will this move the needle for each person reached? Score on a scale: 3 = massive, 2 = high, 1 = medium, 0.5 = low, 0.25 = minimal.
Confidence: How confident are we in the reach and impact estimates? 100% = high confidence (backed by data), 80% = medium (some evidence), 50% = low (gut feel).
Effort: How many person-months of work? Include engineering, design, and any other functions.
When to use: When you need a quantitative, defensible prioritization. Good for comparing a large backlog of initiatives. Less good for strategic bets where impact is hard to estimate.
A roadmap is a communication tool, not a project plan. Keep it at the right altitude — themes and outcomes, not tasks.
When reprioritizing, always ask what changed. Priority shifts should be driven by new information, not whim.
Flag capacity issues early. If the roadmap has more work than the team can handle, say so.
Dependencies are the biggest risk to roadmaps. Surface them explicitly.
If the user asks to add something, always ask what comes off or moves. Roadmaps are zero-sum against capacity.
About this skill
Trigger
Update, create, or reprioritize your product roadmap. Use when adding a new initiative and deciding what moves to make room, shifting priorities after new information comes in, moving timelines due to a dependency slip, or building a Now/Next/Later view from scratch.
The verbatim description from this skill’s front matter — the string an agent matches on to decide whether to load it.
Apache-2.0 — the text of every skill is reproduced unmodified, frontmatter included, under the upstream licence.
Discovery
200 skills found by walking the repository tree for SKILL.md, not by matching a directory convention. 27 distinct layouts observed: bio-research/skills/*/SKILL.md, cowork-plugin-management/skills/*/SKILL.md, customer-support/skills/*/SKILL.md, data/skills/*/SKILL.md, design/skills/*/SKILL.md, engineering/skills/*/SKILL.md, enterprise-search/skills/*/SKILL.md, finance/skills/*/SKILL.md, human-resources/skills/*/SKILL.md, legal/skills/*/SKILL.md, marketing/skills/*/SKILL.md, operations/skills/*/SKILL.md, partner-built/apollo/skills/*/SKILL.md, partner-built/brand-voice/skills/*/SKILL.md, partner-built/common-room/skills/*/SKILL.md, partner-built/slack/skills/*/SKILL.md, partner-built/zoom-plugin/skills/*/SKILL.md, partner-built/zoom-plugin/skills/contact-center/*/SKILL.md, partner-built/zoom-plugin/skills/meeting-sdk/*/SKILL.md, partner-built/zoom-plugin/skills/meeting-sdk/web/*/SKILL.md, partner-built/zoom-plugin/skills/video-sdk/*/SKILL.md, partner-built/zoom-plugin/skills/virtual-agent/*/SKILL.md, partner-built/zoom-plugin/skills/zoom-mcp/*/SKILL.md, pdf-viewer/skills/*/SKILL.md, product-management/skills/*/SKILL.md, productivity/skills/*/SKILL.md, sales/skills/*/SKILL.md.
Issue colours
Resolved from the awesome-design-md registry — hue 39°, chroma 0.113. Two accent tones are generated per issue and each is proven against its own ground before it ships: a single accent that passes AA on both light and dark paper is arithmetically impossible.
Typefaces
Display: Copernicus is not available to us; set in Newsreader.
We never load a typeface at request time from a third-party origin, so a brand face we do not already self-host is substituted rather than fetched.
Heading repairs
1 repair applied to this skill so the document has one h1 and no skipped levels:
Removed “Roadmap Update”, a leading h1 that duplicated the skill title.
Spec compliance
53 editorial notes across 41 of 200 skills. They are printed in the margin of each skill rather than as errors here.
Images inside a skill come from the upstream repository. Where the author gave no alternative text we mark the image decorative rather than inventing a description — a plausible caption we made up is worse than none for the reader who depends on it.
Marketplace
A plugin manifest is published at .claude-plugin/marketplace.json by Anthropic, declaring 120 plugins. It is read for editorial metadata only — never as the skill index, which is always the repository tree.
Signal
Install counts come from skills.sh. They measure downloads, not quality, and an unranked repository is not an unread one.
Set by Skills Docs from the source repository. Body text is Literata at the reader’s chosen size and measure; code is Geist Mono. Nothing on this page was written by us except this paragraph.