Methodology
How a product gets a score
Every product or solution evaluated on this site is scored against the same six categories, computed the same way every time. The rubric below is published in full — including the exact thresholds — so a low score is explainable, not arbitrary.
The six categories — 0 to 10 each, 60 total
| # | Category | What it scores |
|---|---|---|
| 1 | Workflow improvement | Does it measurably reduce procedure time, steps, staff burden or complexity versus current standard of care? |
| 2 | Scientific proof | Strength and quantity of published evidence — preclinical through peer-reviewed clinical data, not just claims. |
| 3 | Clinical efficacy | Demonstrated clinical outcome improvement where evidence exists. |
| 4 | Cost-benefit evaluation | Cost of acquisition/use weighed against workflow, outcome, or downstream-cost benefit. |
| 5 | Uniqueness | Genuinely novel mechanism or approach versus an incremental variation on an existing category. |
| 6 | Overall benefit | Composite of Patient benefit (0–5) and Clinician benefit (0–5) — a holistic "so what," not a restatement of categories 1–5. |
Recommendation tiers
Evaluated top-down; the first tier whose two conditions are both met wins. A high total that's concentrated in only a few categories can still fall through to a lower tier — the rubric rewards well-rounded products, not products that ace one or two categories and ignore the rest.
| Tier | Category count | Total score |
|---|---|---|
| EXCELLENT | ≥5 of 6 categories score ≥7 | total > 45 |
| GOOD | ≥4 of 6 categories score ≥6 | total > 30 |
| MAY PROVIDE A VALUE | ≥3 of 6 categories score ≥6 | total > 25 |
| NO VALUE | fallback — none of the above met | |
Where the score comes from
An AI-assisted first pass proposes the six category scores from the submitted weblink or document, with justification tied to specific evidence in the source. The final score and tier are then computed deterministically from those numbers — never asserted directly by the AI — and every published evaluation is checked by an editorial reviewer before it goes live.
A submitter who disagrees with a result can request a correction, or resubmit once new evidence exists. See Submit a Product.