innonate

continuations:

It has been great for startups to be able to fire up instances on Amazon EC2 or Slicehost for next to nothing, but one thing that has gone down the drain in doing so is email deliverability.  I try out a lot of new services all the time and at this point I would say that well over half of those new services don’t get their email messages into my inbox.  Instead they get swallowed up by Postini.  And because Postini is generally excellent (very few false positives), I rarely bother to check it.  That’s a real problem for the growth and adoption of these services, because if I can’t register or reset my password or see when I have a friend request, I will be less likely to use the service.

The reason for the deliverability problems are that email reputation today is IP address based and it takes time to “warm up” an IP address, i.e. build reputation for it.  Even worse, every once in a while someone will really spam from a cloud provider (whether on purpose or by accident) and potentially damage the reputation of a whole block of IP addresses.  It would be great to see the the major ISPs agree to a different model in which a startup can make an initial payment of some small but meaningful amount (say $500) to get an IP address “whitelisted” right from the start and then have it monitored right away by Returnpath (disclosure: USV portfolio company).  This could even be a great upsell for cloud providers — e.g., when activate a new machine they will say “Planning to send mail? Sign up here.”

Until that is the case, I can only recommend to everyone to start paying attention to email deliverability from day 1.  I know there is a lot of stuff to worry about in getting a service off the ground and this may seem small, but if you can’t reach folks it’s a big deal.

Or you can not send a single email from your servers to start. We’re launching AnyClip using MailChimp for newsletters and SailThru for notifications. I can’t report on their performance yet — we haven’t launched — but I can tell you that the value proposition is spot on. Buy vs. build analysts (i.e. head of product, like me) will undoubtedly say, “I don’t want to worry about being whitelisted and email delivery rates. I’ll pay someone to do that for me, and keep track of it, and let me A/B test my copy, and do a billion other valuable things not clearly alligned with my company’s mission and unique abilities.”

So, good idea Albert. But I’d rather pay $500 to MailChimp and Sailthru, not the ISPs.