Feb 09

Google Trends Versus Twitter Twist Trends

Both Franz Ferdinand Tonight:Franz Ferdinand and Andrew Bird Noble Beast have released new albums in the past 30 days. I started thinking about how good they are, relative to one another.

In my initial preview I found “Tonight” to be the same old Franz — in a boring way, while “Noble Bird” was the same old Andrew in a good way. I turned to Twitter and Google to see what everyone else is thinking.

First, Google shows parity in the US across the two acts in search, with Franz Ferdinand getting substantially more news hits. Oddly, the albums were released in separate weeks and search results don’t reflect this.

Andrew versus Franz

Andrew versus Franz

The most interesting distinction through Google Trends came through the location information. I started with no regional filters. Franz Ferdinand popped very high in Croatia and other seemingly odd places. I then filtered for only in the United States.

Andrew versus Franz by location

Andrew versus Franz by location

Bird crushed Franz Ferdinand in Austin, Chicago, Boston, Philadelphia, and Minneapolis. FF dominated Irvine, CA and LA. Not sure what that means? Bird rocks and FF doesn’t (yes I live in Chicago).

Franz versus Andrew

Franz versus Andrew

Twitterers are making more noise about Franz Ferdinand (sorry about the color swap, blue = bird in these graphs) — though in general, the tweet rate looks low. They start to look similar over the last few days.

The real metrics here are actual listens, downloads, or sales. Based on current Amazon Sales Rank, in CD format, “Tonight” ranks #36 and “Noble Beast” ranks #94. However, as an MP3 download,”Tonight” ranks #54 and “Noble Beast” dominates at #12.

I have a few tools I can use moving forward to track this stuff in more detail. Not sure what to make out of this just yet — but interesting nonetheless.

Dec 08

Smarter Recommendations: Mood and Context

This post on TechCrunch, Clerk Dogs Takes a Curated Approach To Movie Recommendations, got me thinking about recommendation engines again.  Namely, the Netflix contest to optimize its algorithm.  The problem I’ve had with the Netflix approach is that it assumes it’s just a matter of being smarter at using inputs to predict outputs.  But what if your inputs are wrong?

An illustration will help.  I remember the first time, some time in the 90′s, I used Amazon’s book recommendation engine.  I had fun rating all the books I’d read.  Book lovers enjoy remembering the vast history of books we’ve had affairs with.  This type of engine works relatively well for books — looking at co-occurence of books read (and ratings) across users.  Music and movies present a more difficult challenge.

My premise is that many times movie watching (more so than music) is a social experience.  The movies you choose to see, order from Netflix, or buy, are all products of who you plan to watch them with.  Further, there is an element of mood.  How many times has someone said, like choosing where to eat out, “what are you in the mood for?”

Imagine a Netflix rating system that asked a few questions in addition to the stars:

  1. Who did you watch the movie with?  What would they rate the movie?
  2. What day of the week did you watch?  What time?
  3. Was this movie recommended to you?

These might not be the right three… but the way I’d approach the Netflix Prize challenge (though of course this strategy is not within the rules) is to cycle through a few qualifying questions asked in an easy yet entertaining way.  Imagine the fun byproducts of some of these questions?  The best date movie for a Saturday night at home, the most popular hump day movie, etc.

Just a beginning of a thought here.  I just think all 5 star movies (even my own) don’t always deserve 5 stars (though some might — though I’ll use another post to talk about the likelihood that 1 and 5 starts are more mood and context dependent than 2-4 stars).  And if you could tease this out, you’d get vast improvements over the “optimization” approach.  A lesson that could be applied outside the world of recommendations.

Dec 08

What’s Wrong With Email?

Twitter, Facebook, wikis, blogs, and a smattering of other web 2.0 companies are being used for business.  Yammer, specifically, caught my eye recently.  It’s described as Twitter for business.  But why?  What gap is it filling?  For me, the question is, what’s wrong with email?

Managing (if that’s possible) email is a full time job, at least at my company.  With a Blackberry in hand there is an expectation that you will read, and sometimes reply to, emails as you receive them.  If we could make money, as a company, reading and responding to emails, this might be okay.  There’s probably a startup idea there.  Regardless, that can’t be the only thing wrong with email.

The magnitude of emails can be troublesome, especially with the implicit expecation of response.  The problem, however, is predicated on the push nature of the medium.  If you take a random 10 emails from your inbox (or wherever your last 10 are), you’ll probably have a few that you wanted to receive — and in fact, are later arriving than you hoped (you have that same expectation of response for people you send emails to).  But, there are probably a few that are just important enough that you have to read, and maybe doing something with, just not as you read them.  And finally, you’ve got ones that are just cluttering your inbox.  These are probably the easiest to deal with.  The thing is, you don’t control what gets to your inbox in the first place.

Then there’s a matter of the content.  Forget the continuing degredation (or increasing efficiency?) of the resolution of language.  I’m concentrating here on the facts, requests, attachments, etc, that come along with emails.  When you combine the magnitude of emails with their walled content obsured behind time, sender, and title, you get the inability to extract value from within your email program.  All these other bits need to be removed from email and used somewhere else — in another program, in a contract, in a request, etc.

And a chain reaction from the previous problem?  The group feature of email is lost after the first email.  Meaning, if I send an email to 5 people, they all get the information inside.  But what happens when one person replies?  If we all wait to reply, in some predetermined order, there isn’t a problem.  In reality, however, people respond (go back to the first problem) and sometimes a barrage of multiple responses flood the cc’s of the email in a disjointed mess.  Further, if the intention is to progress on some file within the email, each of the recipients must open the attachment, make changes, and then hope to reassemble all of the simultaneous edits people are making.

This is just a start to some of what’s wrong with email as a corporate communication tool.  I want to lay out this initial list and see how far it gets when I start talking about the solutions to email, including Yammer.