agentleFS
lukesw55lukesw55/ai-pm-toolkit/skills/pm-prioritization-regua-comum/SKILL.md

pm-prioritization-regua-comum

Skill1 starsChanged 42 days agoMITView on GitHub

The "Régua Comum" product-prioritization model — Impact × Effort with a single shared impact ruler. Use whenever the user needs to prioritize initiatives, score a backlog, run or prepare a prioritization session, decide what goes into the next PI, or tell product evolution apart from one-account customization. Triggers on "priorizar", "régua comum", "impacto x esforço", "o que fazer primeiro", "vale a pena fazer isso?", "isso é customização?", "score this backlog", "prioritization workshop", "Business / Abrangência / Strategic risk", "confiança", "HIPO". Scores three impact dimensions (Business impact, Abrangência, Strategic & risk), applies confidence and HIPO weighters, enforces the Abrangência lock against customization, rates effort, and plots the Impact × Effort matrix into a decision. Runs on Amplitude's North Star Framework as a working premise: impact is leverage on the North Star Metric through its inputs. The ruler is the same for everyone; who asks changes the *evidence*, not the *score*.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Write the configuration into the block at the top of references/templates.md and revisit it only when the OKRs change.

    line 41

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Régua Comum — product prioritization

One shared impact ruler for every team. Whoever asks (sales, customer, CS, engineering, leadership) changes the evidence, not the ruler. Impact is always computed the same way, regardless of who brought the request or how urgent it looks.

Progressive loading

Load this SKILL.md first. For large or specialized tasks, use references/progressive-loading.md to choose the narrowest supporting reference before reading more.

Core principle

Optimize for product leverage: prioritize what becomes a reusable capability, serves the OKRs the ruler is configured against, and is strategically relevant. The model exists to avoid two errors: (1) turning the roadmap into a queue of single-account commercial customizations, and (2) faking quantitative precision we don't have. Commercial urgency, a big customer's name, and leadership opinion do not raise impact on their own — they enter as evidence (or as an explicit, limited weighter) and are counterbalanced by confidence.

Configure the ruler

The three dimensions are structural; what two of them serve is configured once per team, before any scoring.

  • D1 Business impact names the commercial OKR it serves (recurring revenue in most instantiations).
  • D2 Abrangência is fixed: does the problem generalize and does the solution become reusable?
  • D3 Strategic & risk names the strategic, regulatory or security OKR it serves.
  • Materiality for D1 is a limit set once from your own revenue base. Never invent a threshold mid-session, and never import one from another company's ruler.

Write the configuration into the block at the top of references/templates.md and revisit it only when the OKRs change. references/rubric.md closes with one worked instantiation as an example of what a filled configuration looks like.

North Star premise (working assumption)

The Régua Comum operationalizes Amplitude's North Star Framework. The OKRs this ruler serves sit downstream of one North Star Metric (NSM), the leading indicator of revenue that captures the value customers get from the product. Lagging business results such as recurring revenue and net-new customers are what the NSM leads; they are not what you score directly. Your product's NSM and its inputs come from your North Star definition; until that converges, treat the NSM as a hypothesis and lean on the confidence factor.

Teams never move the NSM directly. They move a small set of inputs that produce it. Read any initiative through the levels of bets: the NSM is Level 0, an input is Level 1, an opportunity to move that input is Level 2, an intervention or feature is Level 3.

Binding rule: every scored initiative names the North Star input it intends to move (Level 1) and the opportunity it exploits (Level 2). If you cannot trace it to an input, that is the framework's roadmap-check signal: either the work is not valuable, or an input is missing from the model. Flag it before scoring; do not invent the link.

Prioritize by how much the input influences the NSM and how likely this initiative is to move that input. The three impact dimensions below are how the Régua Comum estimates that; they are not three unrelated scores.

Official impact definition (read this aloud at the top of every session)

Impact is how much an initiative moves our OKRs through a reusable, strategically relevant capability, not how much a single account wants it, nor how urgent it seems, nor who champions it.

Procedure

Work each initiative through these steps. Use the template in references/templates.md (copy one block per backlog item). Detailed 1–5 rubrics live in references/rubric.md; worked examples in references/examples.md; the workshop and governance in references/workshop.md.

  1. Frame the demand. Capture origin (Sales / Customer / CS / Security-Compliance / Discovery / Engineering / Pipeline), the problem, the value hypothesis, the evidence per source, and the North Star input (Level 1) the initiative intends to move plus the opportunity (Level 2). Origin never sets the score; it sets confidence. If no input link exists, raise it as a roadmap-check signal before scoring.
  2. Score the 3 impact dimensions, 1–5 each (full rubric in references/rubric.md):

    • D1 Business impact — does it generate, unblock, expand, or protect the configured commercial outcome?
    • D2 Abrangência (includes reuse) — strong evidence it serves more than one customer/segment and the solution becomes a reusable capability, not a one-account customization?
    • D3 Strategic & risk — does it advance the configured strategic, regulatory or security position, or mitigate strategic risk?

    North Star reading: each dimension estimates how the initiative moves an input toward the NSM. D1 Business impact is the lagging commercial result the NSM leads, so score the input movement, not the NSM itself. D2 Abrangência is the breadth and reuse of that movement across the base, which is why a one-account pull scores low (the NSM is customer value, never a single account). D3 Strategic & risk covers strategic, regulatory and system-health inputs (the framework recommends a system-health indicator input).

  3. Compute raw impact (simple average):

    Raw impact = (Business + Abrangência + Strategic) ÷ 3
  4. Apply the two weighters — they are not impact dimensions; they adjust an already-computed impact:

    Final impact = Raw impact × Confidence factor × HIPO factor   (capped at 5.00)
    Confidence (evidence quality — only discounts)Factor
    Low (1 source / hypothesis / no discovery)0.70
    Medium (2+ sources or partial discovery)0.85
    High (discovery + convergence, or documented regulatory/contractual obligation)1.00
    HIPO (leadership conviction — explicit & limited)Factor
    Deprioritize0.85
    Neutral (default)1.00
    Prioritize1.15

    HIPO rules (non-negotiable): default is Neutral; any non-Neutral HIPO must be logged (who decided + one-sentence why) or it reverts to Neutral; HIPO is capped at ±15%, so it moves at most one classification band — it never rescues a weak item nor kills a strong one. Moving more than one band is a decision outside the model and must be annotated as such.

  5. Classify the final impact:

    BandImpact
    1.00 – 2.49Low
    2.50 – 3.74Medium
    3.75 – 5.00High
  6. Apply the Abrangência lock (the main protection against customization):

    No initiative may be classified high impact if it scores low on Abrangência (1–2), except when it is mandatory for a documented regulatory or strategic obligation, critical security, or retention of material recurring revenue (a large deal/renewal provably at risk).

    Material is the limit written into the ruler configuration, never a number invented in the session. Allowed exceptions cover only those three cases and must be logged (owner, one-sentence justification, the OKR/risk it protects). HIPO does not disable the lock or waive the exception log. If Abrangência = 1–2 and no valid exception, it does not enter the product roadmap — route it: parametrize (turn into config that serves everyone, raising Abrangência), deliver as a paid service/project (outside the product roadmap), or decline with a clear reason.

  7. Rate effort — Low / Medium / High (six signals in references/rubric.md). The worst signal pulls the score: any single "High" signal (risk, architecture, compliance) makes effort High. Rule of thumb: all Low = Low; some Medium, no Highs = Medium; any High = High.
  8. Plot Impact × Effort and read the quadrant:

    Low effortMedium effortHigh effort
    High impact🟢 Top priority🟢 Plan (PI)🟡 Evaluate trade-off
    Medium impact🟢 Quick win🟡 Prioritized backlog🔴 Caution
    Low impact⚪ Opportunistic🔴 Avoid⛔ Do not prioritize
  9. Record the decision and owner. Map the quadrant to a recommended decision (Do now / Plan (PI) / Evaluate trade-off — consider phasing an MVP / Backlog / Customization / Decline) and assign a next step + owner. For high/high, prefer phasing (MVP first) to cut risk.

Protection rules (summary)

  • Sales requests enter the ruler like any other; Sales provides evidence, not the final score. No sales request is automatically high impact.
  • An important customer raises confidence in the problem, not the impact of the solution. Score Abrangência separately.
  • The Abrangência lock and exception log are the backbone — keep them honest.

What to produce

A filled matrix/sheet plus a decision table (item → North Star input → quadrant → decision → owner → next step) and the logged list of exceptions and non-Neutral HIPOs — ready as input for PI Planning. When the input is qualitative or thin, be explicit about uncertainty via the confidence factor rather than inventing precision. For deeper quantitative validation of the evidence (datasets, SQL, A/B), chain with data-science-analyst; for turning decisions into roadmap narrative, chain with pm-phase-define.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.