ToolPilot Methodology
The ToolPilot Score
Every tool on DoWithAI receives a ToolPilot Score on a 0–10 scale. The score is computed from five criteria with fixed weights — the formula does not change per tool and is not subject to opinion.
Scoring formula
| Criterion | Weight | What it measures |
|---|---|---|
| Output Quality | 30% | Usefulness of the tool output on real tasks for the target profession |
| Value for Money | 25% | Price-from relative to feature coverage, normalized by category |
| Profession Fit | 25% | How well the tool matches the workflows of its target profession |
| Ease of Use | 10% | Onboarding speed and learning curve |
| Integrations | 10% | Number of native integrations relative to category median |
Data sources
- Official vendor sites — pricing, features, trial terms, integration lists
- G2 — ratings, ease-of-use subscores, sentiment signals
- Capterra — cross-validation of ratings and user complaints
- Reddit & YouTube — practitioner feedback, unfiltered complaints
Update cadence
- Pricing — re-scraped weekly
- Ratings — re-scraped monthly
- Features — re-checked quarterly or on vendor announcement
Every review shows a Last Updated timestamp. If data is more than 90 days old, a warning banner appears on the page.
Quality gates
Reviews that fail any of the following conditions cannot be published:
- Fewer than three documented limitations
- No computed ToolPilot Score
- Missing last-scraped timestamp