CIOs and CTOs share the IT culture misconceptions that are holding some IT organizations back. Do people on your teams need to dump any of these?
How to explain big data in plain English
You may think everyone knows what big data is by now, but misconceptions remain. Get expert advice for discussing big data in plain terms with colleagues, customers, or any audience
Big data has been a boardroom buzzword for some time now. Despite its widespread use, however, it can still be wildly misunderstood.
Technology leaders know that big data alone has no inherent worth. “People sometimes think all they need are large datasets, but large datasets aren’t intrinsically valuable,” says Hadayat Seddiqi, director of machine learning at legal tech company InCloudCounsel. “Big data’s true value lies in the information you can extract to answer a specific business question.”
Nor is “big data” a terribly precise term. Some use it to refer to the data itself, while others employ it when talking about the analysis of, or insight derived from, that data.
[ Are you skipping important data decisions? Read also: 4 bad data habits that devour value. ]
Let’s explore some starting points for a conversation with any audience about what big data is and is not, where it might deliver new insights or opportunities for the organization, and what a big data strategy should have.
5 big data definitions, in plain terms
Nitin Aggarwal, vice president of data analytics for The Smart Cube, keeps his explanation of big data basic: “If your enterprise data cannot be stored, accessed, and processed effectively in your existing data warehouse or storage, it’s called big data.” The volume of data may be too big, for example, or the rate of data growth will outpace the rate of storage you can economically add, or the types of data cannot be managed with current technology.
We asked some other experts for their best plain English explanations for kick-starting a big data discussion:
- “Big data refers to the ability to access and use data – data that was never available in the past – to make more educated decisions and predictions.” –Todd Wright, head of data management, SAS
- “Big data refers to extremely large volumes of disparate data that can be used for analysis, insights, and predictions.” – Phil Rodoni, CTO, Rubicon Global
- “Big data is high-volume, high-velocity, and/or high-variety information assets that demand cost-effective, innovative forms of information processing that enable enhanced insight, decision making, and process automation.” –Gartner IT Glossary
- “Big data is a relative term and depends on who is using it. Broadly, it refers to the data which is significantly [greater] in size than most enterprises are accustomed to, generally changes faster than usual data, and typically is needed to be analyzed in a shorter time to derive business value.” –Yugal Joshi, vice president, Everest Group
Bonus big data analogy: Think shopping
When all else fails, an Amazon online shopping explainer usually does the trick, says Christopher Rafter, COO of Inzata. “Every product you click on, review you read, item you put in your cart, and what you eventually purchase, is captured. All of those individual data points come together to paint a picture about what happened, what you shopped for, what you browsed, and what you ultimately purchased,” he explains. Captured from thousands of shoppers and millions of purchases, the resulting big data is analyzed for patterns and trends to drive better decisions about pricing, product suggestions, and more.
How to clear up common big data misconceptions
Most business leaders have a reasonable understanding of big data, but some significant misunderstandings persist. The first, and perhaps most damaging, is the assumption that all big data has business value.
“The term ‘big data’ leads many to assume that value is derived simply from the sheer amount of data that an organization holds, and the organization that has the most data wins,” says Wright of SAS. Not so. “The true value comes from how an organization can get a broader view of their customer and business by tapping into different and previously unused data sources,” he explains. “That in turns leads to more educated and informed decisions with the use of analytics.”
Volume ultimately matters much less than the quality, cleanliness, usability, and accessibility of data, adds Aggarwal. What’s more, not every company needs big data. “In our experience, a majority of business questions do not require big data,” Aggarwal notes. “Big data isn’t the cure for all business problems.”
Some people also assume that big data is like regular data – but yields more detailed insight. “That is not necessarily true,” says Polina Reshetova, data scientist with EastBanc Technologies. “Big data often brings new questions. It has its own statistical properties and it requires a new way of thinking about results and asking questions.”
In addition, not all big data initiatives require massive amounts of input. “Projects can be surprisingly small,” says Wolf Ruzicka, chairman of EastBanc Technologies. “Our smallest big data project deals with one terabyte of data. It started in the gigabyte range. The key is to have the right type of data: clean, accurate, relevant, timely, and rich enough.”
That’s why big data efforts don’t have to be huge investments – another incorrect assumption. “One does not need to wait for years and spend millions of dollars to set up an enterprise-level big data platform,” says Aggarwal. “There is a lot that can be done at a smaller level.”
How do you construct a smart big data strategy? Let’s delve into that question: