Insight, Observations, and and Updates on the Big Data Landscape

Jeffrey Abbott

Subscribe to Jeffrey Abbott: eMailAlertsEmail Alerts
Get Jeffrey Abbott: homepageHomepage mobileMobile rssRSS facebookFacebook twitterTwitter linkedinLinkedIn

Related Topics: EMC Journal

Blog Feed Post

Data Science says: “You don’t want a Ferrari.”

Jeffrey Abbott

I recently read an article about a guy named Doug DeMuro, who, like many of us, always wanted a Ferrari. That may not sound newsworthy, but he actually went out and bought one. And it was a big disappointment. Data Science would have told him to buy a Porsche.


Figure A: Doug’s Ferrari 360 Modena

Doug wanted a car that would do everything a 15-year old boy’s 1985 Ferrari poster promised it would. However, he knew that analyzing this choice too carefully could result in a minivan. So he chose to blindly base his decision on “I want it. I want it. I want it.” And I can respect this. Sometimes, that’s the only way. He bought a used Ferrari 360 Modena. The Ferrari did everything it’s supposed to do. It was red. It looked awesome. It sounded awesome. It went fast. It made other people jealous.

Measuring the benefits of the Ferrari reveals that it excels in certain areas, such as appearance, quality of materials, sound, road/track performance, and the prestige of the “Ferrari” brand. These are the reasons we all want Ferraris, and Doug chose to put logic aside, and went for it! Boo-yah Doug! Doug for President!

But what went wrong that left him regretting his choice?

  • Did Doug overlook significant metrics?
  • Did Doug emphasize the right metrics (for Doug)?
  • Did Doug assign the right performance value (for the Ferrari) to each of the metrics?
  • How well would other cars perform across those metrics?

It’s extremely difficult for a person to determine if they are making the best car decision, and that’s why Doug intentionally ignored most categories. Doug made an emotional decision. But in business, we don’t have this “luxury.”

Businesses are at risk of making similar mistakes when deciding how to advance a business initiative by tapping into their data. The business would seek to discover some insight that can be leveraged to develop a new product, improve an internal process, reach a new market, or enhance the customer experience. They’d first need to determine the analytic questions that could be asked of the data to derive the insight they need. They’d need to determine the best analytics use case for their initiative. And they need the right data to analyze, and the capabilities to do it. EMC Global Services offers a Big Data Vision Workshop that focuses precisely on finding the ideal analytics use case for a business initiative. We examine your top-level business initiatives and help your IT and business groups focus on one. In Doug’s case, the initiative was to own and drive his childhood dream car.

Applying this 3-4 week workshop engagement to Doug, we would use data science approaches to consider all the questions that could be asked to determine the right car for Doug’s initiative.

A data scientist would research Doug to determine the types of car metrics that should be analyzed in context to others, and measure the performance values across many choices of automobiles. And importantly, the data scientist would find deeper correlations and relationships (insight) than his Car & Driver magazine’s “10 Best Super Cars” issue would provide, because the analysis would be in context to Doug’s persona and Doug’s initiative.

For example, part of driving a Ferrari is driving. If Doug had a Ferrari, how much would he drive it? Well, as it turns out:

  • Ferraris are difficult to drive
  • Ferraris are uncomfortable on all but the smoothest of roads
  • Ferraris are not driven in rain, snow, off road, or on muddy, bumpy, or even wet pavement
  • Ferraris are not parked at the super market parking lot
  • Ferraris do not hold more than 2 people
  • Ferraris have very little storage
  • Ferraris are not left overnight in public or generally left unattended for more than 10 minutes
  • Ferraris are not used for long daily commutes or in heavy traffic

So how do those limitations match up against Doug’s persona?

  • Where does Doug live?
  • How much free time does Doug have?
  • Where does Doug go on a regular basis?
  • With whom does Doug travel?

Given that Doug’s initiative was based on “driving” (not just owning), the Ferrari turns out to be an impractical choice for Doug’s busy and urban lifestyle in Philadelphia. Bill Schmarzo of EMC explains the role of analytic profiles in this blog. You may notice that I’m leaving out metrics around price and maintenance costs (for Doug, buying an economical, reliable, and easily repairable car were not part of the initiative). So to what extent do the positive attributes of the Ferrari outweigh the drawbacks and how does that measure up against other cars, based on what’s important to Doug and his usage opportunities? Fortunately for Doug, making the wrong choice was a tolerable outcome.

Businesses, on the other hand, are making bigger, more financially significant decisions that can’t simply be undone, like selling a car. They have more analytic use cases to consider, and more data to analyze.

Had Doug applied our Big Data Vision Workshop approach, he’d have discovered that when he combined the benefits of the Ferrari with the feasibility of driving it, he would have deprioritized the Ferrari, and… bought a Porsche (okay there I said it).

Unfortunately, Doug found that the only place he could take his Ferrari for a drive was right back to his garage that he just pulled out of. As he put it, it was a “point A to point A” car. In other words, whenever he could carve out 30 minutes to accomplish nothing, he’d take it for a drive and then take it right back. Many organizations that dabble with big data start by going after a specific use case but they fail to adequately consider:

  • To what extent can this big data use case help me advance my most important business initiative?
  • How valuable is this big data use case (car) compared to others?
  • How can this approach be applied to additional business initiatives?

EMC-GS-May-the-4th1-300x145In our Global Services organization, we help IT and the business come together on a business initiative to explore how big data approaches can be applied. Then we help them consider individual analytic use cases – leveraging sample data to examine the benefits and feasibility, to form a priority matrix and plan. We even offer a Proof of Value service to let customers test drive their analytic use case in their environment to prove out how the use case can deliver the value that it needs to. And when the customer is ready to deploy, we configure and deliver a big data platform, such as the recently announced Federation Business Data Lake Platform, and customize it to the customer’s use case (showcased at EMC World 2015).

Doug got the formula wrong. I suspect that the Big Data Vision Workshop would reveal that Doug should have bought a Porsche. He would have gotten slightly less of the “gotta have it” feeling, but he’d have been able to drive the car far more, resulting in higher overall contribution to his initiative to own and drive a dream car.

Porsche 911:

  • Still a poster-worthy super-car
  • Easy to drive anywhere
  • All-wheel drive, 4-season usage
  • Enough storage for weekend trips
  • Inconspicuous enough to leave unattended
  • Roadworthy enough to handle foul weather, bumps, potholes
  • Reasonable as a commuter car
  • 4 seats can accommodate 2 people in back (it’s tight, but possible)

Figure B: Me, searching for a use case…

Had Doug considered the all the metrics that actually affected him, he’d have been led to a Porsche (for the record, I’m more of a large-displacement V8 guy (see figure B on right – my guilty pleasure) but the 911 is the most successful sports car of all time for a reason – many reasons). The result of Doug’s lack of deep analysis is that he sold the Ferrari, and abandoned the very thing that he had lusted over since childhood. He should have bought a Porsche. Don’t feel bad Doug, I did the same thing for the same reason, but I’m not disappointed.

Data Science says: “You don’t want a Ferrari.”
Jeffrey Abbott

Read the original blog entry...

More Stories By Jeffrey Abbott

Jeff is part of Tata Consultancy Services Digital Software and Solutions group, as a lead evangelist for its IoT analytics platform solutions for smart cities, smart retail, smart banking, smart comms, and other areas.

Prior to TCS, Jeff was part of EMC’s Global Services division, helping customers understand how to identify, and take advantage of opportunities in Big Data, IoT, and digital transformation. Jeff helped build and promote a cloud-based ecosystem for CA Technologies that combined an online community, cloud development platform, and e-commerce site for cloud services and spent several years within CA’s Thought Leadership group, developing and promoting content and programs around disruptive trends in IT. Prior to this, Jeff spent 3 years product marketing EMC, as well as a tenure Citrix, and numerous hi-tech marketing firms – one of which he founded with 2 former colleagues in 1999. Jeff lives in Sudbury, MA, with his wife, 2 boys, and dog. Jeff enjoys skiing, backpacking, photography, and classic cars.