September 20, 2014

Understanding Big Data with “Number Sense”

number senseOne challenge the Internet presents is too much choice when one has a preference for simple entertainment.  For example, when I want a mini–break from writing or analyzing data, I casually watch YouTube reviews of new vehicles. Reviewing cars is part of my car enthusiast childhood. It helps me recall the three automobile magazines subscriptions my dad would bring home.

But the volume of YouTube auto review sites can eat up my time.  Too much choice can overwhelm, forcing me or anyone to applying judgment to not overindulge in variety.

I share that personal aside to frame the purpose behind the book, Number Sense: How To Use Big Data To Your Advantage by Kaiser Fung. Fung is a professional Harvard educated statistician and author of Numbers Rule Your World.  

He wrote Number Sense to heighten the ability of readers to discern when the plethora of choices of data has value and when it does not.   I learned about the book through NetGalley and sought an advance review copy before its July publication.

Number Sense: Making Sense of Numbers

Number Sense is written for organizations in general, not just businesses.  All types of organizations are facing more data-based decisions by the day.  In establishing a wide scope Fung makes a solid claim that big data increases the operations possible in an organization:

“The reason why we should care is not more data, but more data analyses. We deploy more people producing more analysis more quickly…. With so much free and easy data, there’s bound to be more analysis.”

Fung goes on to note how there is now more chance for error:

“Data gives theory legitimacy. But every analysis also sits on top of theory.  Bad theory cannot be saved by data. Worse, bad theory and bad analysis form a combustible mix.”

To enlighten us on how the chances for error are occurring, Fung supports the details with a case study approach in the text.  The book lays out its chapters into four segments – Social Data, Marketing Data, Economic Data, and the intriguingly titled Sporting Data. Each segment contains enterprise and societal viewpoints to highlight how data-derived decisions can misinterpret the models an organization creates from the data.

A superb early example is a Republican model for the 2012 presidential election.  Fung notes a set of polls enhanced with a data model that predicted a last minute groundswell of Republican support and a Romney victory.  It’s a poignant example of how competitors can shoot themselves in the foot even with equal access to the same data.

Readers who own small restaurants, retail shops and services reliant on Groupon-like digital services will benefit from the two chapters examining personalization within – wait for it – Groupon.  The takeaway is that personalization extends the value of customer segmentation, but not always as expected from a planned strategy. Fung explains how tech can go beyond intention:

“Target technology is one tool that can strengthen the economics of a Groupon merchant. But the punditry fails to comprehend how. Targeting as described is not so much concerned with sending more relevant deals to subscribers. It works by directing coupons to profitable segments of customers, away from the free riders and towards the first timers.”

Ideas like this can enhance small business actions when it comes to strategy (see Megan Totka’s  article “How A Small Business Can Use Big Data” for more big data suggestions).

Examples in Number Sense read like cases, so business readers should not rush into the text for answers, or else they’ll overlook the salient points.  The end chapter on data and fantasy football surprises in its presentation, but its takeaways were simpler than the text describing the example.

In some instances I did prefer asides to explore the details.  When Fung notes that analytic solutions “can have a gap as high as 20-30%” he gets the number right, but the book does not deeply explore the current state of technology solutions being applied to big data challenges.

Read Number Sense to Learn, But Be Cautious to Apply

With Number Sense, Fung seemed ambitious in expanding on McKinsey Global Institute’s definition of big data. And I appreciated Fung’s humility and fairness in mentioning that he can be in error as well – “Even experts sometimes fall into data traps.”  The tone shows an eye for continual learning and inspired exploration – the ultimate takeaway readers should have for business intelligence books.

Ultimately readers enamored with stats-books like Moneyball and Big Data will not be disappointed.  In narrowing the choice of where to starting researching the concepts behind big data, Number Sense makes a credible read that can show where the big data story is heading.

As for small entertainment until my next book review … I am still stuck choosing which auto review video to watch (smile).

4 Comments ▼

Pierre DeBois - Associate Book Editor


Pierre DeBois Pierre Debois is Associate Book Editor for Small Business Trends. He is the Founder of Zimana, a consultancy providing strategic analysis to small and medium sized businesses that rely on web analytics data. A Gary, Indiana native, Pierre is currently based in Brooklyn. He blogs about marketing, finance, social media, and analytics at Zimana blog.

4 Reactions

Leave a Reply

Your email address will not be published. Required fields are marked *

*

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>



Compare your business to the industry - Try our new tool