How we test

We compare email tools on what can be checked and sourced: vendor pricing pages and docs, published deliverability tests and first-hand user reports. Every ranking publishes its criteria and weights before any tool is scored, and every number on the site has a source and a date. We don't run our own sending tests yet.

Last updated

What we score

  1. Price at volumeThe real monthly price at fixed volumes (for example 50,000 and 100,000 emails), from the vendor's own pricing page.
  2. DeliverabilityThe tooling a tool gives you (authentication setup, suppression, dedicated IPs, warm-up, insights, separate streams), published inbox placement tests and what users report.
  3. Developer experienceOfficial SDKs, the steps from sign-up to the first email in the vendor's quickstart, a test mode or sandbox, and what users say about setup and docs.
  4. AI readinessAn official MCP server, llms.txt on the docs, official AI integrations and AI help in the docs.
  5. Free tierHow many emails a month you can send free, for good.
  6. Support and reliabilityFirst-hand user reports on support and uptime, counted as positive or negative.
  7. Onboarding frictionWhether you must wait for an account or domain approval before you can send.
  8. Data residencyWhether your data can be stored in the EU, and on which plans.

Each ranking picks the criteria that matter for its job and weights them, out of 100. The weights for the best transactional email services are on that page.

How a ranking is scored

  1. We fix the criteria, weights and scoring rules before we score any tool.
  2. We collect each tool's facts from its own pricing pages and docs, and note the date we checked.
  3. We add published deliverability tests, credited to their publisher with the test date, and first-hand user reports from public reviews and developer forums, counted by theme.
  4. We score every tool with the same rules. A criterion with no verified data scores zero and shows "–"; we never estimate.
  5. We publish the scores with a one-line basis for each one.

Where the data comes from

Vendor pricing pages, documentation and quickstarts; public status pages; independent deliverability tests with a named publisher and date; public user reviews and developer communities. Screenshots come from vendor docs and help centers and are credited and linked.

What we don't do yet

We haven't run our own sending or inbox placement tests. Deliverability results on this site come from the tests their publishers ran, and we name them. When we add our own tests, each result will say when and how it was run.

How rankings work

A ranking compares tools for one job, such as sending transactional email. Each ranking explains which criteria matter most for that job and why. Tools we haven't finished checking are marked "In testing" and get no score.

Keeping results current

Email tools change fast. We re-check pricing and features and date every page with its last check. Pages log real changes with their date.