Big Data’s Dirty Secret

Do you know what you’re doing with your Big Data? Based on my experience, only a small group of companies really do.

There’s just one little problem, as Intel cloud platforms group vice president and general manager Jason Waxman, told investors in a webcast last week.

This is the dirty little secret about big data: No one actually knows what to do with it,” said Waxman, as ComputerWorld reported. Link (emphasis mine)

While I disagree with his statement that no one” actually knows what to do with it, I do agree with his overall premise.

It’s been my experience that many companies adhere to the Volume aspect of the Volume, Variety, and Velocity definition. Typically they’re so scared to miss any potential data point that they just save everything and clutter up their clusters.

What many companies lack is a well thought out Data Policy that builds a cluster with the Right Volume, the Right Variety, and the Right Velocity. Unfortunately, formulating that data policy takes up a lot time and requires broad input from across the organization.

Not only will it require input from the Data Science team (i.e. data analysis), but it will require input from the Line of Business (LOB) groups and corporate strategy/vision from upper management. Sounds great! Let’s get to work, right?

Just ask yourself this: when was the last time all three of these groups aligned with minimal friction?

Answer: Usually never.

Tl;dr: Right Volume > Volume. Big Data’s dirty secret isn’t that dirty if you get it right.

Prescriptive Analytics and My Heart

The world right now is awash in Predictive Analytics, the mystery of Big Data, and the rise of the glorious and magical Data Scientist. Most of the time we hear these buzz words in relation to some marketing campaign, election, or credit score application process, but what about applying these tools and people to a project that can benefit the welfare of humanity?

Well, one group of data scientists and a healthcare provider in Washington State are doing just that.

A partnership between data scientists at the University of Washington Tacoma (UWT) and MultiCare Health System, the South Puget Sound’s largest hospital and healthcare system, is fine-tuning algorithms to better predict which chronic heart failure (CHF) patients would be most susceptible to readmission within 30 days, with the goal of understanding which adjustments to treatment plans have the most impact on preventing readmissions.

Their goal is to create a readmission score as-a-service” that would enable providers to use the bank of predictive models for many chronic conditions, not just heart failure.

Of course this project has a multi-pronged benefit, not only does it help CHF suffering people to get better healthcare, but it also reduces costs in the long run. Win-win in my book.

There was one other interesting nugget that that caught my eye.

The current project status is prototype, he said. We are working on physician validation in a clinical setting. Once we are able to show that it works well and gets the kinks out, then the next step would be a full-on EHR implementation.” (emphasis mine)

Prescriptive Analytics

Operationalizing Predictive Analytics falls into the realm of Prescriptive Analytics, where you automatically extract your insight and act on it. Unfortunately, this is typically where organizations stumble, and stumble hard. It’s one thing to build a new algorithm, model it, and predict the different classes with a high degree of confidence, but it’s another problem altogether to operationalize it.

Tl;dr: A Data Scientist a day keeps the Doctor away .

Churn Models, Cable TV, and My Wife

Churn models are not new in the analytics world, they’re heavily used by mobile telcos and other corporations that want to keep their loyal customers happy, and bring back customers from the brink of about to churn.” In some cases, these models will help classify a group of customers that are such a pain in the butt to keep, it’s better to let them go.

Makes business sense, right? Keep the best customers happy, group the customers that are about to leave into the ones you can save or not. I can’t help but wonder if the ones that you think are better off churning, aren’t the ones that are really telling you something valuable about your business. Like my wife.

My wife is one sharp cookie and she can easily ferret out the best deals for our family. It’s like a game for her, she wants to spend $1 to get $5 worth of value. Granted, I’m biased, but I consider her one of the ultimate shoppers and she routinely evaluates our monthly subscriptions for something better.

Enter our Cable TV fiasco.

We’ve had Cable TV, Internet, and VOIP for many years because we can’t get any other alternative services in our area. FIOS isn’t available and heavy tree cover prevents us from getting DirectTV. When we first got these services 10 years ago, the Cable TV company gave us an introductory rate for about $100/month, and ever year when the deal was about to expire, my wife would call to complain about the cost and then get them extend the offer for another year. If she didn’t, our $100/month bill would jump even higher.

Just think about that for a moment. Basic Cable TV, Internet, and VOIP cost us over $100 per month. I consider paying over $100/month for a landline we barely use, basic Cable (no premium channels) and Internet a bit costly.

Then something changed.

The media consumption habits of Ott household have changed over the past few years. We spend more of our time watching Netflix and Amazon Prime, and less time on CBS, NBC, or even Cable TV. We hardly use our landline phone and spend all our time on our mobiles. We just consume our daily information and entertainment via the Internet more than ever.

This led to several conversations with my wife about cutting the Cable TV cord and porting our VOIP to a new provider, but keeping the Internet, because that is truly the nerve center of the Ott household. So right after Christmas, we decided to make a change. We got rid of our VOIP with them and went with an $8/month outfit out of India.

As soon as we switched to a new VOIP provider, we started getting inquiries from the Cable TV company to win us back. We got new offers that would only expire after a time and would mean we’d be on the phone arguing with them in the future. Stop. The. Madness!

When I started penning this article, my wife was on the phone arguing with a customer representative about their service and costs again. We decided to cut the Cable TV cord once and for all and just keep their $60/month Internet service – also very costly if you ask me.

They tried to lure her back with more offers but she crossed a threshold, the deals stopped and there was no attempt to keep her. The customer representative did what he had to do and asked her to stay because with of all the great offered packages but she retorted with a single statement that summarizes what true competition and customer service should be, she said: The packages you offer for the price is not good enough. You need to give me what I want for less, and not at teaser rates.”

I don’t know a darn thing about the Cable TVs churn model, but I like to believe that my wife is in a special subset of customers that demand good service and cheap prices – at all times! She was probably flagged as one of those pain in the butt customers that you should let go, but in this particular case she’s really telling them something valuable. You’re fat, lazy, and missing a new trend.