One good argument I found [^1] about not exposing primary keys is that primary keys may change (during system/db change) and you want to ensure users have a consistent way of accessing data.
It's not the same as what Mailgun does. Mailgun lets you send emails programmatically. Sendy/Sendwithses lets you send campaigns from an interface (using your SES).
I think a better route is getting someone to handle the marketing. As engineers, even when we know what to do in terms of marketing, we still find it hard to do because it's not something we enjoy like engineering.
I'd rather recommend using a cheaper service like Amazon SES than running your own mail server. Email deliverability is complex and figuring it out may not be worth it.
Yes it's intentionally as Mandrill is still used solely for transactional emails and Mailchimp for marketing. Same reason I didn't merge Mailgun and Mailjet.
Thanks for the feedback. Actually makes sense. The target is for newsletters so no attachment support, spam filtering (for now) and access to full email headers (no RSS element this will fit in).
I actually started it as opensource but maintenance and monetisation was difficult. The source is still a good start point to build anything off Amazon SES/Mailgun analytics https://github.com/suetco/suet
[^1]: https://softwareengineering.stackexchange.com/questions/2183...