Follow Datanami:
March 12, 2012

Google Guru Highlights “Data Imperative”

Datanami Staff

Avinash Kaushik, author and Google statistical analytics guru spoke last week at the Strata Conference, highlighting his thoughts on the “Big Data Imperative”—an imperative that he thinks has slipped through the cracks in the midst of the revived hype around data management, value and analytics.

From his experiences “playing with all the stuff” inside the hallowed halls of Google and other web giants, “the key thing that you try to figure out is how to make all the petabytes and terabytes of data actionable to the person who is sitting on the ground—and then a lot more useful. So I actually don’t really care about the promise of data, unless we can deliver on the promise that comes with the data.”

According to the Market Motive founder, the models we have used to dig for data have been “broken for a very long time.” He claims that for companies in this big data era, there is a sense that there is urgency and value in all the data—and everyone knows many technological methods to tackling it. However, he argues that instead of making actionable use of the data

“…we hire these gods and princelings to go take this data, convert into reports and hopefully pray to god that they will take action at the other end. The problem with this model is that it doesn’t actually scale because as the users multiply, we run around like bunnies trying to find people that don’t exist, whose job it is to produce ever more data and start hitting people with it every single day to try and produce action. Ironically what happens, is the company becomes more inefficient. It’s a very sad, pathetic story.

Then there are still the arrogant people who refuse to touch data because their faith still sustains them. For them, we will create these princes who will play with it and dump data. The difference in this scenario is because we empower every person to take power of the data. Very large decisions will be made in the name of Jesus and God and the all other actual useful decisions will be made by people closer to the data, and this scales. The great thing is that this scales, because it’s part of every person’s job rather than having these people in the blue run around like bunnies trying to produce something.”

As Kaushik told the crowd at Strata, the key to the future of big data is creating a “data democracy that is sans-data-puking and is driven by the quest to find things in data that other people would never know exists.”

While not specific in terms of the technologies or movements required to snap the collective enterprise world out of his big data haze, he claims that his interest rather lies in possibilities. As he stated during the end of his talk, “I am not interested in having lots of big data. I’m not interested in having massive heads explode because of the size of the data we have. I’m really interested in what I hypothesize you should be interested in… It’s trying to figure out how to use that and the intelligence we possess to make life better for the people who are in our companies that deal with our data, people who work at our clients and people on the other end who use our products and services.”