Tuesday, June 2, 2009

In Short...A Behavioral Masterpiece

Consumed with the topic of behavioral targeting over the past few days and then experiencing what seemed like a catastrophic outage on Twitter recently, I started to reflect on the incredible amounts of data that Twitter has at its disposal and the many uses it has to target consumers behaviorally.

There are so many solutions that are available to target behaviorally. To avoid some of the red tape behind the creepy privacy issues, some solutions providers are labeling the systems “re-targeting or re-marketing tools”. It doesn’t matter how you slice it, keeping tabs on an individual’s user activity and then formulating a profile with rules attached to it and an ad server is behavioral targeting.

Now, let’s look at what Twitter is sitting on in terms of data. Here’s an example of a timeline:
  • Beautiful Toronto day…sun is shining and the lake is gorgeous.
  • @exampleA I can’t wait to go to Cuba next week.
  • Driving the kid(s) off to school – 12 month old is jealous of 4yr old brother going to kindergarten - must be a girl thing...
  • Looking at cars this weekend…can’t wait!
  • Loving the Car1 but the Car2 is good too. Any suggestions?
  • Ugh Air X sucks – lost our luggage – time for mojitos
  • Planning hubbies 40th – glad I’ve got 2 years to savour my youth.
This 7 entry string provides us with:

Age, gender, location, car purchaser, leisure traveler, # of children and their ages.

Now imagine millions of individuals with hundreds and thousands of entries.

Mining this data for advertisers is a huge step forward in marketing. With an integration like what Google is proposing with Wave - communication could be served in real time to garnish any behavioral cue. Sunscreen promotions leading up to the trip to Cuba, financial offers from car companies etc.. But what about analyzing this incredible amount of data from a more general perspective?

Having the ability to understand trending of global car purchasing intentions, travel profiles, financial profiles and the list goes on. Heck it’s almost as good as having access to someone’s personal email content only better because:
  • "Contained context" - 140 characters are much easier and faster to crawl than email communication
  • Micro-entries are not as filtered and contain raw, impulsive communication
  • The ability to index one individual within a network of followers and trend consequential behavior across the individual’s network (influence factor)
  • The real time factor is subtle but enormous in value – the speed of crowd sourced content is unprecedented and can be indexed with news and current events
Those are just a few examples. There are many more.

In closing, here is some food for thought.

On May 29 I tweeted this:

On June 1st, this appeared in my gmail box.

If you’re wondering why Twitter is taking so long to monetize, picture yourself realizing how much data just landed on your entrepreneurial lap and ask yourself what you would do with it all.