When Postmark is the right call
Choose Postmark when the cost that actually matters is your time. Setup is measured in minutes rather than days, the support team knows deliverability properly, and the event data is clear enough that debugging a specific message is straightforward rather than archaeological.
For a solo operator or a small studio, the price difference against SES is frequently less than the value of one afternoon not spent reading AWS documentation. For a portfolio pushing millions of messages a month, it is not.
The list-quality caveat, stated plainly
Postmark is stricter than most providers about bounce and complaint rates, and they will get in touch about a bad import faster than SES would.
Treat this as a feature rather than a risk — it is the same posture SendNectar takes, and the two reinforce each other. If you import suppressions before contacts, use double opt-in, and run frequency caps, you will never hear from them. If you do not, you will hear from them before the damage is permanent, which is better than the alternative.
Streams map cleanly onto message classes
Postmark’s split between broadcast and transactional streams lines up almost exactly with SendNectar’s marketing and transactional message classes.
The clean setup is two provider connections — one per stream — with sending routes directing each message class to the matching one. Marketing volume then cannot delay a password reset, and the reputation of each stream stays legible on both sides.
Notes for multi-brand senders
Create a server per brand. It costs nothing structurally and it keeps event data, activity logs, and reputation signals separated in Postmark the same way SendNectar separates them — so when you investigate a problem, both systems agree on which brand you are looking at.