A bit of shameless self promotion: Haraka is a great mail server for coding up systems for processing incoming email, including parsing out attachments. We use it at Hubdoc for streaming attachments directly from email to S3.<p><a href="http://haraka.github.com/" rel="nofollow">http://haraka.github.com/</a>
Solving problems around email has a wide audience.<p>After our startup launched a private and secure messaging service, we weren't able to drive easy and repeatable traction.<p>We broadened the appeal to include email privacy which not only got press but created a ton of sticky users. Eventually, we turned it into our first point of revenue.<p>Email allowed us to reach a wider audience while still preserving our focus on privacy. While we weren't technically "email first" it was a big second that ultimately served our primary focus.
So would this be considered an email first project? I had an idea for website called "Global Voices" (domain will be glblvcs.com) Users from all over the world will enter their email and subscribe to a micro-newsletter that comes out every Friday night. By entering their email they're also automatically entered into a "lottery". Every week 3 lucky people are chosen and each get to ask all the users 1 question which will appear in that week's newsletter. The 3 questions in each newsletter will link to a message board where people from all over the world with diverse backgrounds and cultures can answer the questions if they want.<p>I think email in this case would encourage participation. If you're not subscribed and getting the emails you don't get a chance to ask a question. By being subscribed you have a greater chance at returning to the message board and writing an answer. Thus fueling the community.<p>Sound good?
I'll add another one:<p>Email users stick around.<p>I've seen a lot of web companies fail because their retention sucked. Even users that absolutely love your products forget about you because they are being bombarded with noise from everyone else.
I'm on my 3rd startup. The first 2 involved building product first and then finding users. This time around I started blogging and accidentally built up thousands of email subscribers.<p>Validation happens so freaking fast over email. As long as you can solve the user's problem over email who cares if there is a product. Wrote about my experience here - <a href="http://blog.goodsense.io/2013/03/27/making-guesses-for-your-startup-is-better-than-careful-logical-deductions" rel="nofollow">http://blog.goodsense.io/2013/03/27/making-guesses-for-your-...</a>
I'm the founder of a stealth startup that focused on email first as a key part of our product.<p>Swombat was one of my early adopters (<a href="https://news.ycombinator.com/item?id=5280289" rel="nofollow">https://news.ycombinator.com/item?id=5280289</a>). Hopefully I'll be launching as a Show HN sometime next month.<p>Email first allowed me to quickly validate the business and generate revenue without having to worry about anything other than the core value proposition.
This is an interesting approach. It's also a great way to build audience way before your real product launch.<p>I wish I know this before launching my latest startup, <a href="http://www.bucketlistly.com" rel="nofollow">http://www.bucketlistly.com</a>.<p>I always thought that you need a product first in order to build audience. Boy, how I was wrong and wasted countless days just to learn that you can build audience even without a product. :/<p>Thanks for sharing it here.
Beeminder started this way too (it was Kibotzer -- the kibitzing bot [<a href="http://blog.beeminder.com/beenamer" rel="nofollow">http://blog.beeminder.com/beenamer</a>] -- back then). Bot emails you asking for your number (eg, your weight) and sends you back a graph of your progress.
I am trying this with the reboot of my project <a href="http://cluedb.com/" rel="nofollow">http://cluedb.com/</a> - it's a daily tip project. It efficiently reuses old content so it isn't strictly a mailing list.
<i>Email has identity built in. Email is identity.</i><p>I'm no security expert, but my understanding is that email is pretty flawed when it comes to establishing the true identity of the sender. I guess you could use something like DKIM or SPF, but plenty of people don't have that set up.<p>If you use obfuscated inbound email addresses then it's not really a problem. But, if you're identifying people by their FROM address on a very public inbound address, be aware that it's trivially easy to spoof that.
Indoctrinator: <a href="http://indoctrinator.com" rel="nofollow">http://indoctrinator.com</a> is my first email startup. It's a weekly program that reveals the secrets to becoming obsessed over anything you want.<p>Anyone whose been obsessed before knows how it can accelerate learning and mastery of skills.<p>I thought email would be the ideal platform for Indoctrinator because I can roll it out in phases and give more personalized service.<p>EDIT: I invite anyone who reads this and is interested in the program to sign up for more info and early notification of launch.
Is this resuming <a href="http://ryanhoover.me/post/43986871442/email-first-startups" rel="nofollow">http://ryanhoover.me/post/43986871442/email-first-startups</a> ?<p>It's getting noisy in here.<p>Edit: Here's the discussion: <a href="https://news.ycombinator.com/item?id=5279590" rel="nofollow">https://news.ycombinator.com/item?id=5279590</a>