About this AI skill
The Release backlog refinement skill is pre-built for Elle (the AI assistant in Aha! software) and checks every feature in a release for completeness, including its description, strategy links, product value score, estimate, linked customer ideas, dependencies, and status. It produces a backlog health summary, a recommended ranking based primarily on product value scores, and specific actions for features that need refinement. The report also flags potential duplicates and suggests consolidation without changing statuses, deleting features, or merging records.
The full Release backlog refinement skill, as Elle uses it:
# Release backlog refinement
Use before sprint planning or a roadmap update to confirm a release backlog has enough definition to prioritize.
## Definitions
Use these meanings consistently in every step.
**Completeness criteria:** The factors each feature is evaluated against.
- **Description:** Clearly defined, with all template sections filled out
- **Strategy:** Linked initiatives, goals, or both
- **Product value score:** Set on the feature
- **Estimate:** Effort captured either as an effort metric within the product value score, or in the **Estimate** field
- **Customer feedback:** Ideas linked to the feature, so customer need is visible
- **Dependencies:** Records on the **Related** tab set as a dependent link relationship
- **Status:** Set to **Ready to develop**
**Priority criteria:** Tie-breakers, in order: stronger strategy alignment, then higher customer demand measured by total idea votes, then lower effort estimate.
## Workflow
1. Scope the release. If it is not provided, ask which release to refine. Confirm any custom prioritization criteria and their relative weights.
2. Review every feature in the release, including anything in the parking lot the user asks to include, and analyze each against the completeness criteria
3. Identify duplicates. Flag a candidate pair when names are highly similar, when the features share identical idea links, or when descriptions overlap significantly. Pair the candidates and explain the overlap briefly.
4. Identify refinement gaps. Flag features missing any completeness criterion and give a concrete remedy for each. For example: add an estimate, link the feature to a supporting initiative, or draft acceptance criteria covering three to five scenarios.
5. Suggest rough priority. Rank primarily by product value score and break ties with the priority criteria. Place features missing essential data in a needs-refinement section rather than in the main ranking.
6. Summarize and recommend. Give a short executive summary of backlog health, list the duplicate candidates with consolidation suggestions, and provide the ranked list alongside the needs-refinement list.
## Rules
- If the release or other critical information is missing, ask short follow-up questions before proceeding
- Write for product managers and engineering leads
- When custom prioritization criteria are provided, restate them and explain how they change the ranking
- Give a specific, actionable refinement step for every flagged gap
- Be explicit when confidence is low because the data is limited or poor quality
- Reference every feature by reference number and name
- Do not change statuses, delete features, or merge duplicates. Recommend the action and let the user apply it.
- Headers use sentence case. No em dashes, en dashes, or curly quotes.
## Output
```
## Executive summary
<overall backlog health and readiness, in three to five sentences>
## Priority recommendation
| Rank | Feature | Product value score | Estimate | Why it ranks here |
## Needs refinement
| Feature | Missing | Specific next action |
## Potential duplicates
| Feature A | Feature B | Nature of the overlap | Suggested consolidation |
## Recommended next steps
1. <highest value action>
```
Save the report as an artifact and share the link so product and engineering work from the same version.
## Quality checks
- Every feature in the release appears in exactly one of the ranked list or the needs-refinement list
- Every gap names the specific field or link to add
- Every duplicate pair states why it was paired, not just that it wasTry this AI skill now
Open the Release backlog refinement skill in Aha! Roadmaps to review your backlog and identify what needs attention before prioritizing work.
Open this AI skillAbout Elle
Elle , the AI assistant in Aha! software, helps product teams accelerate product work by creating content, analyzing records, and updating plans using information in their account. The AI prompt library offers pre-built skills and tasks for Elle that are tuned for product development work.