A higher score is not the whole decision.
Use a fictional CRM shortlist to see how evidence and must-haves change the outcome.
Open the workspace and choose Explore a fictional example. It creates a separate project without replacing your work.
What the example actually calculates
Northstar’s fictional configuration has a stronger weighted fit but an unresolved integration. Harbor’s lower illustrative cost cannot offset its failed export requirement. Both need follow-up before a buying decision.
| Fictional option | Fit range | Supported ratings | First-year cost | Decision status |
|---|---|---|---|---|
| Northstar (fictional) | 67–83% | 5 of 6 | $3,200 | Evidence incomplete |
| Harbor (fictional) | 53% | 6 of 6 | $2,300 | Blocked |
The method
The six requirement weights are 5, 5, 4, 3, 4, and 3, totaling 24. A scored requirement contributes (score ÷ 5) × weight. The lower fit bound divides the sum of those contributions by 24; the upper bound adds the unscored weights before dividing. Both bounds are rounded to whole percentages. A range represents missing ratings, not statistical confidence.
Northstar has five scores of 4 out of 5, with the weight-4 integration unscored: 16 ÷ 24 = 67% to 20 ÷ 24 = 83%, rounded. Harbor scores 3 on five requirements and 1 on the weight-4 export requirement: 12.8 ÷ 24 = 53%, rounded. A must-have score below 3 blocks an option independently of its total.
First-year cost is monthly seat price × seats × 12 + setup. Northstar uses $45 × 5 × 12 + $500 = $3,200. Harbor uses $35 × 5 × 12 + $200 = $2,300. These example totals exclude tax, discounts, usage charges, staff time, and renewal changes.
“Supported ratings” counts scored requirements with a documentation or test label and a nonempty evidence note or URL. It does not independently verify that a note is true. Here, every note is explicitly illustrative.
Reproduce it
- Open the Software Decision Workspace and choose Explore a fictional example.
- Compare the Score & evidence and Add options views with this table.
- Alternatively, download the editable JSON below and use Import backup in the workspace. Keep the six weights unchanged to reproduce these figures.
Download the fictional example ↓
Written by Stephen Stilwell. Calculation checked October 26, 2026 against this site’s decision model. This is a worked example of a tool built for Useful in Practice, not a real customer study or product performance claim. Attribution & corrections.
1. Inspect the two options
Northstar and Harbor are fictional tools for a five-person sales team. The prices, trial notes, and scores are illustrative, not real product research.
2. Find the unresolved integration
Northstar has stronger scores on the assessed requirements, but its email and calendar workflow is unscored. Its weighted fit appears as a range and its evidence is incomplete. In Score & evidence, review the suggested trial task before assigning a score.
3. Find the failed must-have
Harbor costs less, but its migration and export score is 1 out of 5. Because export is a must-have, the option is blocked. A low price or a high score elsewhere does not erase that failure.
4. Turn the comparison into an action
The Decision brief lists the remaining checks, cost assumptions, and failed requirements. Record your choice and rationale, then download the brief to discuss with your team. You can also save an editable backup.