A couple weeks ago, at SXSW Interactive I spent some time talking to OtherInbox.com at their trade show booth. I think email is a space that's long overdue for some serious innovation so I was interested to try out their beta and understand what they were doing a bit better.

Basically, they're trying to address the problem of automated emails clogging up your inbox. They give you another webmail space, and automatically move non-human-generated emails out of your original email account's inbox and give you some interesting ways to slice and dice everything over in your otherinbox.com account.

It sounded ok, but after having it set up for a day, I couldn't see the real problem that it was solving. I was totally confused. I went back to their booth and asked them about it. They just told me that I just needed to try it for a while, then everything would become obvious. They said, "Give it two weeks, then write a blog post about it." I figured that sounded fair, they seemed to genuinely believe in the product, so maybe it would make more sense with time. (Although, for the record, saying "you need to try our service for two weeks to understand how much it rules" is a shitty marketing pitch. Just sayin').

Anyways, I feel like I have a better sense of it now, and while there's a couple things that are sorta neat, it's not really solving a big pain for me.

Where it's cool

• The auto-organizing of the emails is pretty cool. Being able to see a view where emails are grouped by the company that sent them is really a pretty nice way to organize them.
• It did, in fact remove some noise from my inbox. There are some emails that I get that aren't spam, but I still almost always never read them and delete them without thinking about it. It's nice to have that step be automated, without having to set up special rules in gmail. (if only it did a better job of it.)
• They hinted at some upcoming features that sounded pretty neat. Automatically detecting coupons, receipts, and calendar items seems like it could have a lot of potential.

Where it's not so cool (for me, at least)

• I actually cared about some of the stuff they were pulling out of my inbox. At least twice at SXSW, I got direct messages from someone on Twitter wanting to meet up for lunch or drinks or something. Because I don't get DMs on my phone, and the email got moved out of my inbox, I didn't see the message until later on that night. Frustrating. I wonder if computer-generated email is the right thing to be filtering on, given that messaging is so fragmented these days?
• They missed a LOT of stuff, my inbox is still cluttered. Just as an experiment, I didn't manually move anything out of my inbox for the last couple days, and now here's what it looks like:
  
  If their value-prop is "separate the real people from everything else", that's not doing a terribly good job of it.
• It's yet another account to keep up with. I already have too many places that want to be the the place I go to see my data. Social networks, Calendar applications, Flickr, Last.fm, etc. It's overwhelming, and anything that adds to the list is unwelcome. Ironically, that's half of what OtherInbox is trying to solve (at least the constant email notifications you get from those places).
• Just about everything they offer can be done with any email client (albeit with a little bit more work in some cases). On-the-fly email addresses? check. Automatically route emails from company X to a folder where they're all grouped together? check. Flag messages to look at later? check.

At the end of the day, there are a couple of features that I truly wish I could have within my normal email client (but not as an external service). But they're nice-to-haves. Other than that, my take on the it was that it's basically a new webmail space that you use to sign up for services that you don't trust not to pester you with messages you didn't want to read. Having the "spam account" isn't a new concept. It's also a sub-par solution to the problem.

But, the concept of weeding out the non-urgent emails is interesting, and I hope they can figure it out. It's a hard problem to solve, though.