Email API Ledger

Transactional email APIs, compared by the constraint you actually have

Transactional email APIs compared by real constraint — every number read off a vendor page on a dated visit, affiliate links labeled, gaps marked UNKNOWN instead of invented. Last verified: 2026-07-12.

Postmark vs Resend for low-volume transactional email (under 10,000/month)

Free-tier math decides it: Resend's 3,000/month free tier vs Postmark's cliff from 100 free straight to $15.

Amazon SES vs Mailgun for high-volume transactional email on a budget (500,000+/month)

About $50 vs about $530 at 500k emails/month (computed from published rates) — and the AWS ops tax that gap pays for.

MailerSend vs Postmark for transactional email under $10/month

The only sub-$10 seat at the table is MailerSend's $7 Hobby plan — and its 1-day retention is the catch.

SMTP2GO vs Mailgun for agencies sending on behalf of multiple client domains

Sub-accounts on every SMTP2GO paid plan from $10/month vs Mailgun's one-domain $15 tier. Plus the log-retention gap.

Resend vs Amazon SES for side projects that need a permanent free tier

Permanent free tier vs 12-months-then-paid, with a sandbox in the way. Unmonetized verdict.

Postmark vs Mailgun for inbound email parsing (receive-and-process workflows)

Receive-and-process lives or dies on the debugging window: 45 days everywhere vs 1 day below $35/month.

How we verify

Every price, cap, retention window, and rate limit on these pages was read from the vendor's own pricing or docs page on the date shown, and each page lists its sources with access dates. Where a vendor doesn't publish a number, we print UNKNOWN instead of guessing. Where we compute (e.g., a 500,000-email monthly bill from a published per-1,000 rate), the figure is marked computed. Some links are affiliate links and are always labeled; three of the six tools covered pay us nothing, and they still win the comparisons the data says they should win.