Hacker News Reader

Comment by lifeisstillgood | Hacker News Reader

Let me rephrase that to see why the downvotes

I like patio11's approach and agree we should write our own business systems to meet our particular need - I am finishing off a CMS that I have also opensourced (it's text and templates and cookies for tracking, about 100 lines)

Now I am not promoting that codebase, only releasing it cos I don't want to waste a GitHub private repo. But I do wonder if we should all write our own CMS / mailchimp / an tester or if we should community build them or just roll up packages

Partly I am asking what are the common functions we should all have a custom builds and partly where do we draw the line on build vs "buy"

View on HN

What I need and what is almost impossible to find is a good Autoresponder email system with a template editor, a few stats and basic list management/segmentation.

The catch: I must be allowed to self-host. This is almost impossible to find since everybody seems to rely on 3rd party email services.

Oy. Do NOT self-host an email autoresponder. I wrote a great comment about this a few years ago on Brian Armstrong's blog--and he's now doing Coinbase after deciding the email stuff wasn't worth it.

Look in my profile to see who I am and why I'd say this, but as a quick summary, I have years of experience and have also helped companies set up autoresponders that do over 1M emails a day, and that's exactly why I say this. It is a clusterf*ck and a rabbit hole a mile deep and a mile wide. Pay the money (or in Mailchimp's case, don't pay any money until you get bigger.) It is worth EVERY penny.

By the way, any good autoresponder (I use Mailchimp and Aweber and like them both for different purposes) will let you easily export/back up contacts, and none of the big companies are going anywhere.

Agree! (& listen to Erica, she knows what she's talking about)

Deliverability is a nightmare (client of mine lost 5 figures, because his admins [at Germany's biggest hosting company] messed up the DNS setup).

There's a reason why there are so many companies doing email delivery as a service (SendGrid, MailGun, Postmark - to name a few)

Thanks, I'll check it out. Do you happen to know the blog post title? Since comments are loaded via Disqus, I can't search for them with Google :(

Ah, finally found it! http://brianarmstrong.org/blog/my-next-project/ Ctrl-F 'Erica' and read the whole thread of him/me! :) (He did launch the biz, but shut it down not long after that. Coinbase is doing much better! ;)

Thanks for taking the time to find the blog post. I definitely need to reconsider whether or not self-hosting is such a good idea or not.

Do you want to become an expert in email deliverability, DNS, bounces, white listing, DKIM, SPF?

Or do you want to sell your product?

It boils down to that. Spend your time doing what you are GREAT at and what you WANT to do. Are you great at email deliverability? Do you want to spend your time becoming an expert on that? Probably not.

I know how to change the oil in my car. I know how to mow my lawn. I know how to clean toilets. I don't do any of those things because they are a huge waste of my time that is better spent elsewhere. Same thing goes here.

Thats what I am saying - why must you self-host, and is that need so great and shared by enough that you will write your own (a few hours with the python email library, or weeks of work, whichever it is)

Reply on news.ycombinator.com