Feature scoring

# Feature scoring

Use when scoring a set of features on customer evidence and strategic fit to produce a ranked prioritization.

## Definitions

Use these meanings consistently in every step.

- **Feature set:** The features in a release, a shortlist from a report, or a mixed set of candidate roadmap items the user names
- **Customer evidence:** Explicit demand signals: linked ideas, number of promoted ideas, vote counts on promoted ideas, and repeated customer comments
- **Strategic fit:** Explicit strategy signals: linked initiatives, goals, key results, record naming that shows strategic intent, and visible roadmap themes
- **Total score:** Customer evidence plus strategic fit, out of 10

## Workflow

1. Start from the feature set the user names or is currently analyzing
2. If the feature set comes from a report, read the report data first
3. Gather only explicit evidence from the features and their directly linked records
4. Score each feature on customer evidence from one to five
5. Score each feature on strategic fit from one to five
6. Add the two scores to produce a total out of 10, then rank highest to lowest
7. Break ties by stronger customer evidence, then by more explicit strategy linkage
8. Summarize the top priorities, the weakly justified items, and the important trade-offs

## Rules

**Customer evidence scale:** Five means strong evidence from multiple customers or very high repeated demand with a clear problem definition. Four means good evidence with more than one meaningful signal. Three means limited, narrow, or partly indirect evidence. Two means anecdotal demand or a minor signal. One means no meaningful customer evidence found.

**Strategic fit scale:** Five means the feature directly advances a current strategic priority with clear outcome linkage. Four means it strongly supports strategy but is not central to this period. Three means the connection is moderate or indirect. Two means it is loosely connected. One means no clear strategic connection found.

**Interpretation:** Flag a feature as a strategic bet when strategy fit is strong and customer evidence is weak. Flag it as demand that may not fit current priorities when the reverse is true. Note that idea votes underrepresent bugs, platform work, and internal enablers. Say when many items tie in the middle, because that means the real decision boundary sits near the top of the list.

**General:** Do not invent evidence. Call out missing evidence rather than averaging or guessing. Reference every feature by reference number and name. Score every feature in the set. Headers use sentence case. No em dashes, en dashes, or curly quotes.

## Output

```
<1 to 2 sentence summary leading with the strongest features and the weakest pattern in the set>  
  
| Rank | Feature | Customer evidence | Strategic fit | Total | Notes |  
  
### Highest priority  
<the features to act on now, with one line of reasoning each>  
  
### Needs more evidence  
<features whose scores rest on thin evidence, and what evidence would firm them up>  
  
### Strategic bets  
<strong strategy fit, weak customer evidence>  
  
### Trade-offs  
<the decisions this ranking forces, stated plainly>
```

## Quality checks

- Every feature received both scores
- Totals match the two dimension scores
- Ties were broken consistently
- Missing evidence is explicitly identified
- The narrative separates what to prioritize now from what needs more justification