making-product-decisions_skill

This skill helps you structure and document product decisions using a framework that improves alignment, traceability, and learning across teams.

136

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill flpbalada/my-opencode-config --skill making-product-decisions

SKILL.md6.8 KB

Overview

This skill is a structured framework for making and documenting product decisions that balances decision science with practical product management. It helps teams handle complex tradeoffs, align stakeholders, and capture rationale so decisions are auditable and learnable. Use it to reduce bias, speed reversible choices, and ensure senior attention for irreversible ones.

How this skill works

The framework guides you through five steps: frame the decision, generate at least three options (including do-nothing), establish weighted criteria, evaluate options against those criteria, and document the final decision with success metrics and review dates. It emphasizes distinguishing Type 1 (one-way door) versus Type 2 (two-way door) decisions, judging decision quality by process rather than outcome, and choosing a data-informed approach when data is incomplete. Outputs include a concise decision record capturing rationale, dissent, and review cadence.

When to use it

Choosing between competing priorities or major product approaches
Making irreversible or high-stakes architecture/business choices
Aligning cross-functional stakeholders with different perspectives
Documenting decisions for future audits or learning
Delegating decisions while keeping accountability

Best practices

Always include a ‘do nothing’ / status quo option to avoid false binaries
Define decision criteria and weights before scoring options
Limit analysis on reversible (Type 2) decisions to avoid paralysis
Capture dissenting views and set a review date for follow-up
Treat outcomes separately from process quality — evaluate how the decision was made, not just results

Example use cases

Build vs buy analysis for analytics tooling with weighted criteria and cost/time tradeoffs
Quarterly roadmapping choice between mobile features and API platform work
Deciding whether to sunset an underused product area vs invest in revitalization
Assigning decision authority: senior-only for irreversible choices, broad delegation for reversible experiments

FAQ

At least three: status quo, Option A, and Option B. Avoid binary framing to surface creative alternatives.

When do we use data-driven vs data-informed?

Use data-driven when metrics are clear and the system is well-understood. Use data-informed when data is incomplete or the context is novel; apply judgment and expert input.