Sort of true, in the sense that the product isn't separately trained for sales emails vs personal emails vs internal business emails, etc.
But the calculations we chose don't provide a lot of constraints, and the variances were not as high as you'd likely expect. So I'd be comfortable saying that the recommendations generalize well to a vast majority of situations.
Declaring that somebody is falsely claiming to use ML without looking at their product is a bit irresponsible ;)
The positivity, subjectivity, and politeness calculations are all outputs from neural networks, and the overall likelihood is calculated using a decision forest so that we can explain the results to people. There are plenty of emails with a high likelihood of getting a response, even though each calculation may score poorly.
Thanks, Borski! We're really excited about the new tools - it's been a ton of fun figuring out how to combine the machine learning algorithms to get them to spit out actionable advice, figuring out how to make them run fast in Gmail and Outlook, and figuring out how to scale this to 45 million people :)
It's seed funding. The seed funding ecosystem is a vortex of death in Boston.
We (Boomerang) moved to San Francisco after going 0/34 with Boston investors. It took less than 30 days to have our full round complete after we got here, and that was in 2010, before the seed market even started to feel frothy.
All the Boston talk about how "a good company can get funded here just as well as it can in San Francisco; they just fund a lot of crappy companies out there" is nonsense. Over half of those 34 Boston investors have paid us for Boomerang.
The ideal company for a Boston tech investor is one gunning to be the #6 company in an already-established market, run by a 45-year-old dude who graduated from HBS then worked for a 55-year-old VC's former company for a few years. If you don't fit that profile, you need to be rich enough to not need funding, you need to have a way to get yourself to $5m in revenue without outside funding, or you need to move.
There were a lot of young, energetic founders working on exciting stuff alongside us back in 2010-2011. I can think of one or two who still are. Some of the rest moved. Most shut down their companies and went to work for a 45-year-old HBS graduate building the 6th most successful company in some already-well-established vertical.
I've been loving Apps Script for things that are too lightweight to need something like Boomerang, with a server and a full browser extension, but heavy enough that you can't do it with a search string.
Awesome, thanks! I just found out about google Script today - was thinking my only choice would be something like Greasemonkey - so very grateful for all this.
Way back when we were just starting to think about Baydin, Xobni was one of the models for how we could build a company based on an email plugin.
Their tech was so cool -- we'd never seen anything like it. Then Bill Gates demoed Xobni on stage, they had so many downloads, and rumors about acquisitions swirled in the press. It sounded so glamorous and so exciting. And, honestly, so fast and easy.
We didn't know that Xobni had been at it for two years already when they were featured in that keynote. And we certainly weren't thinking at all about a story of 7 years of grind, improvements, and putting one foot in front of the other every day. Helicopter rides, keynotes, WSJ profiles, and closing fundraising deals get the headlines, but behind the scenes, it's releases, updates, support, and hard work.
Now that we're a few years into our own story of grind, improvements, and putting one foot in front of the other every day, we have a ton of respect for what Xobni put together over the past 7 years. Obviously, this isn't the outcome that everyone was hoping for, but it's easy to look at the numbers and forget 2008-2009, when Xobni was at the top of the world.
Thanks, Adam, Matt, and the rest of you who were early in Xobni for building something magical and inspiring us to build something of our own. And congratulations on the end of a long journey.
I love it. For decades, my flame war offensives against the evil crusaders of Emacs have stalled out against their Maginot Line of "But Emacs can be your mail client too!" This is like a Blitzkrieg of flame war glory. PARIS PREPARE TO BE MINE!!!
A little late to the party here, but Boomerang doesn't do anything at all except for when you Boomerang something and when it's scheduled to be sent/returned. The rest of the time, it doesn't make any API calls.
But the calculations we chose don't provide a lot of constraints, and the variances were not as high as you'd likely expect. So I'd be comfortable saying that the recommendations generalize well to a vast majority of situations.