A BETTER WAY TO SHARE THE WORK

Explain what comes first—and what must wait

A priority list is useful when people understand the tradeoff behind it.

New to creating with an agent? The request will guide it through the task. Bring your notes; you don’t need to write HTML.

How can an AI agent help explain product priorities?

Give your agent the candidate work, supporting evidence, dependencies and real capacity constraints. Ask for a proposed order with a reason for each choice, explicit deferred work and questions that could change the recommendation. Let the decision owner approve priorities in your existing planning process.

01 / SEE THE DIFFERENCE

The prerequisite comes before the attractive feature

Fictional reporting-team planning notes; no priority has been approved.

THE RAW MATERIAL

Capacity allows one small improvement this cycle. Twelve supplied support notes mention confusing report labels; that is not a customer prevalence estimate. A glossary revision is estimated at two editor days. Custom labels need a settled glossary and are estimated at five developer days; developer availability is unknown. A color refresh takes one designer day but has no supplied problem evidence. Recommend the glossary first, defer custom labels until definitions and capacity are confirmed, and defer the color refresh. Priya must decide; nothing is scheduled.

THE SHAREABLE BRIEF
Proposed first
Glossary revision: two editor days estimated; twelve support notes indicate label confusion, not customer prevalence.
Dependency
Custom labels need a settled glossary; five developer days estimated and developer availability unknown.
Deferred
Color refresh: one designer day, but no supplied problem evidence.
Decision
Priya must decide within capacity for one small improvement. Nothing is scheduled.
Illustrative example · fictional details, not a customer result. The order follows evidence and readiness, not a made-up composite score.

Open a complete fictional HTML example, with its source notes →

Create my prioritization brief

02 / MAKE IT YOURSELF

Start with the evidence. Shape the page around the reader.

  1. Agree what the order should optimize

    Name the outcome and constraints before discussing attractive features. A fixed deadline, a prerequisite and a preference are different kinds of input. Ask for missing constraints rather than creating numerical weights that make the choice look objective.

  2. Put evidence beside each candidate

    Keep the source and its limits visible. Several support notes may justify investigating a problem without proving how many customers have it. Separate evidence of a problem from confidence that a proposed solution will fix it.

  3. Check what each choice displaces

    Show effort as an estimate, identify prerequisite work, and name what will wait if this item comes first. Do not confuse a high-value idea with work that is ready to begin. Never count uncertain availability as committed capacity.

  4. Share a recommendation, not a silent roadmap change

    Lead the HTML brief with proposed order and rationale, then evidence, dependencies and deferred items. Include the decision still needed and a supplied review trigger. Record approval and scheduling in the planning tool; a readable page does not allocate people.

03 / TRY THIS PROMPT

Copy this into your agent. Add your source material.

Use the conversation where you already did the work, or start one with your notes. Replace the placeholders and supply any missing context; only attach information you are allowed to share with your agent.

Get the ready-to-copy creation request for guided questions, this task, and a draft review before anything is published.

Create a prioritization brief from only my supplied or authorized candidate work, evidence, dependencies and capacity constraints. Ask for missing essentials. Separate problem evidence from solution confidence; preserve estimate units and uncertainty. Show a proposed order, what each choice displaces, prerequisite work, deferred items and conditions that could change the order. Do not invent scores, weights, customer prevalence, staffing, dates or approved commitments. Use static HTML with no scripts or forms and my saved style if available. Show the draft and unresolved questions. Keep priority approval and scheduling in our existing planning workflow. Do not publish or change access without separate explicit approval.

04 / BEFORE YOU SEND

A polished page still needs your judgment.

  • Does each priority connect to a supplied goal and evidence?
  • Are dependencies and capacity constraints visible?
  • Are estimates distinct from booked resources?
  • Does the reader see deferred work and the decision owner?

When HTML fits

A dated explanation stakeholders can read before a planning discussion, with detail below a short proposed order.

When another format fits better

Use the backlog or planning system to assign work and maintain live priorities. A brief does not optimize capacity automatically or replace the accountable decision-maker.

FROM DRAFT TO A LINK

Make it clear. Make it yours. Then share it.

Keep creating and revising in the agent you already use. With a compatible connection, share/artifacts publishes the reviewed HTML so you do not have to rebuild it in another editor. Ask your agent to use your saved artifact style for headings, color, and tone; review the result before sending it.

Choose who should be able to open the page and verify the returned link. For corrections to the same deliverable, update the existing artifact without changing its link. Publish a separate artifact when the old edition needs to remain available.

These are static pages, not live apps: no scripts, forms, or automatic data refresh. Source collection and scheduling depend on your agent and its available tools, not on share/artifacts.

Make this with my agent

Questions before you start

Do I need a scoring framework?

No. A clear rationale can be more honest than an unsupported score. If your team supplies a validated scoring method, preserve its inputs and uncertainty instead of inventing weights.

What if the evidence is weak?

Say what is missing and consider a bounded investigation if that is an actual option. Do not turn a request into proof of widespread demand.

Is the first item now committed?

Only if the authorized decision process says so. Label a recommendation as proposed and leave dates unconfirmed until agreed.

Further reading

Explain what comes first—and what must wait | share/artifacts