Hey HN,Chris here. I have been playfully hacking on a few ideas that leverage some of the latest AI tools (both open-source and proprietary). One in particular seems somewhat useful and is ready for a few more eyeballs.The idea is to make building very capable AI phone agents as easy as creating an online form. We think the ease of production will lead to some interesting enterprise use...
"To demonstrate the economics of our business model, we have compared the revenue and gross profit generated from the North American subscribers we acquired in the second quarter of 2010, which we refer to as our Q2 2010 cohort, to the online marketing expenses incurred to acquire such subscribers. The Q2 2010 cohort is illustrative of trends we have seen among our North American subscriber base. The Q2 2010 cohort included 3.7 million subscribers that we initially spent $18.0 million in online marketing to acquire in the second quarter of 2010. In that quarter, we generated $29.8 million in revenue and $12.8 million in gross profit from the sale of approximately 1.2 million Groupons to these subscribers. Through March 31, 2011, we generated an aggregate of $145.3 million in revenue and $61.7 million in gross profit from the sale of approximately 6.3 million Groupons to the Q2 2010 cohort. In summary, we spent $18.0 million in online marketing expense to acquire subscribers in the Q2 2010 cohort and generated $61.7 million in gross profit from this group of subscribers over four quarters."
They are seeing >3x profit on typical cohorts, so they decided to buy as many users as possible (which is smart).
"To demonstrate the economics of our business model, we have compared the revenue and gross profit generated from the North American subscribers we acquired in the second quarter of 2010, which we refer to as our Q2 2010 cohort, to the online marketing expenses incurred to acquire such subscribers. The Q2 2010 cohort is illustrative of trends we have seen among our North American subscriber base. The Q2 2010 cohort included 3.7 million subscribers that we initially spent $18.0 million in online marketing to acquire in the second quarter of 2010. In that quarter, we generated $29.8 million in revenue and $12.8 million in gross profit from the sale of approximately 1.2 million Groupons to these subscribers. Through March 31, 2011, we generated an aggregate of $145.3 million in revenue and $61.7 million in gross profit from the sale of approximately 6.3 million Groupons to the Q2 2010 cohort. In summary, we spent $18.0 million in online marketing expense to acquire subscribers in the Q2 2010 cohort and generated $61.7 million in gross profit from this group of subscribers over four quarters."
A typical cohort that returned >3x what it cost? Sounds like a good business to me.
We're trying to free your purchase data to allow any developer (including us) to add value. Social commerce (as a space) is ripe to explode, and we are at forefront.
Some of our tools include ruby, rails, haml, memcached, mongodb, sphinx.
Backed by August Capital, CRV, Sequoia, Ron Conway, Evan Williams, and many other awesome angels.
Team: http://blippy.com/about
Funny pictures: http://blippy.com/jobs
However, at the end of the day, watching individual user sessions, one after another, doesn't really scale; you need a way to aggregate the data to identify trends that may suggest pain points for the end user, and we haven't really cracked that nut (and neither have our competitors).
Its tough for a startup to attract big players to use their messaging platform or service because it is difficult to guarantee uptime, reliable service, etc. Might be perfect for mashups and hackers, but they won't pay.
Cool stuff though; I had a lot of fun playing with it.
Simple events = mouse movements, clicks, focus events, scrolling etc.
Advanced events = DOM mutation events triggered by ajax and other types of sophisticated javascript: http://en.wikipedia.org/wiki/DOM_Events
We will likely have to do customer specific fixes to get advanced event captures working well during our beta phase, which is why we want you to email us if you are at all interested.
As for your second question, we can certainly setup the service such that it captures a certain percentage of users, and it's certainly a route we would consider depending on the size of the client.
Thank you!
Check it out at http://userfly.com/In short, Userfly captures actual browsing sessions from real users as if you were looking over their shoulder. What we've found is that watching real users actually use your site can be remarkably insightful, so we built Userfly to accomplish that goal.There's a screencast that will walk you through the basic functionality. Click on the demo button above it...
In fact, I ended up building this because I didn't want to pay for the other mobile services platforms (there are a few of them) when attempting to add mobile features to another project I'm experimenting with.
But to your original point, I absolutely agree, pitching this as a "mobile" "platform" for "widgets" is probably hopeless. Quite simply, it must solve somebody's problem. And I believe that we can get it to do just that. One area that we think is promising is using email to invoke applications that do interesting things because (1) everyone knows how to send email, and (2) you can attach lots of useful things to email.
So, what if we added the ability to recognize a ton of different file formats and allowed you to query the data within an attached file to invoke other web services?
For example, lets say you have a site that does group payments, and you want to let your users send you an Excel spreadsheet with the names of your friends and how much they owe you for that trip you just took together and automatically create an invoice on your site, and then notify everyone that they owe money. That seems interesting, and our product could support that pretty easily.
Anyways, it is definitely our immediate burden to focus this technology on solving _real_ problems.
I don't think we have the killer app yet, but we implemented a mobile interface to google calendar with ~5 lines of code which I've been using for a little while, and its surprisingly useful. When all is said and done, I think we might want to simply pitch this as a platform for adding mobile features to your existing app, so it becomes more of a developer tool than a consumer facing product.
As for monitoring, yes, it becomes a tricky task to achieve high availability (and scalability), but its a really fun problem to tackle :)
http://vimeo.com/2231638Would love to know what you guys think.Beware of the discontinuity @ 6:44 :) Sorry, no public demo available (yet).