Methodology

Last updated: 2026-04-23

This is the framework we use to evaluate and rank packages in our index and comparison pages. We publish it so you can judge how applicable our recommendations are to your situation.

Evaluation criteria

For every entry we evaluate, we score against eight dimensions:

1. Pricing transparency + value — published pricing, predictable model, fair free tier where applicable
2. Documentation quality — completeness, examples, getting-started time
3. User experience — onboarding clarity, error states, request/response design where relevant
4. Reliability — published SLA, public status page, historical uptime, incident communication
5. Coverage + features — feature completeness vs category leaders, supported regions, standards
6. Security + compliance — SOC 2, ISO 27001, data residency, audit logs where relevant
7. Ecosystem + integrations — third-party tooling, integration partners, community size
8. Vendor health — funding status, customer concentration risk, deprecation history

Weighting by use case

Different use cases warrant different weightings. Our "best for [use case]" pages apply weights specific to that use case. The framework is published; the weights are documented on each page.

Refresh cadence

Each entry's score is recomputed at least quarterly, or sooner if a vendor ships a major change (new pricing, new region, new feature set, new compliance certification).

What we don't evaluate

We don't grade vendors on subjective brand factors (logo design, marketing site quality). We don't penalize vendors for being newer or smaller — we evaluate on what they ship, not on age.

Conflict policy

See our disclosure for our conflict-of-interest policy and how commercial relationships (where they exist) are surfaced on relevant pages.