Here is an excerpt from an article written by Rebecca Knight for Harvard Business Review and the HBR Blog Network. To read the complete article, check out the wealth of free resources, obtain subscription information, and receive HBR email alerts, please click here.

* * *

In the era of Big Data, analytics are becoming a competitive necessity for many managers. And even if it’s not an explicit part of your job description, chances are you need to understand at least something about data and analytics to be successful. So, where should you start? What’s the best way to get a handle on the basics? How should you get to know the quants in your company? And how can you be sure you’re asking them the right questions?

What the Experts Say: Whether you’re in financial services, consumer goods, advertising, hi-tech, or public health, the nature of your day-to-day job is likely changing to include data. For some managers, this is intimidating. “A lot of people want to hide because they weren’t particularly quantitative in school or school was a long time ago, but given the number of executives that want to use data-driven insights to propel their decision-making, you can only hide for so long,” says Thomas H. Davenport, a professor at Babson College and the author of Big Data at Work. Besides, hiding is bad for your business and your career, says Joe Knight, a partner and senior consultant at the Business Literacy Institute, and the coauthor of Financial Intelligence. Your goal, after all, is to be the kind of manager who not only understands formulas and analytic methods, but also knows how to “interpret [results] to make better decisions and improve efficiency.” Here are some strategies to help you.

Get educated: If you remember your statistics classes from college or business school, you may be okay. But for those who don’t or, ahem, are somewhat numerically challenged, a refresher is in order. Fortunately, there are myriad ways to get back up to speed: Enroll in an executive education class; read books and articles on the subject; take an online course. Your goal, according to Knight, is to become data literate. “It’s not hard to get the fundamentals down,” he says. “You need to have a sound baseline understanding of business analysis—including the four families of financial ratios and tools for calculating return on investment.” Davenport highly recommends also getting a solid grasp on the basics of regression analysis. “The vast majority of statistical analysis approaches are based in regression analysis,” he says. If that term makes you want to turn and run, don’t. You don’t have to know how to run a regression analysis but simply what it means and how it is performed.

Form relationships: In some organizations, the quant team and the business executives have an adversarial relationship, says Knight. Managers may feel, “Those numbers guys are always ruining our projects.” While the analysts think, “Those business people don’t understand the analysis.” It’s counterproductive. You will get much better business results if you form deep, trusting ties with the quants in your organization, according to Davenport. Don’t write off your analysts as “geeky” number crunchers. Get to know them and treat them with respect. Perhaps the best way to develop mutual trust is to make clear that you value their skills and want to learn from their expertise. Simply saying, “I have a lot to learn from you” can go a long way toward forming a partnership. And if you have a say in hiring decisions, be on the lookout for people who “want to solve business—not just mathematical—problems,” says Davenport. This will help prevent antagonism in the first place.

Include analysts in decision-making: If you manage a team that includes quants, don’t let them sit on the sidelines. “Make the quant a full-fledged member of your team,” Davenport says. “Expose them to the business problems, so they can see them with their own eyes.” Set up frequent meetings and project-reporting sessions so that they fully understand the business. But don’t just ask them to observe; include them in the decision-making process, especially when you’re using the data they’ve delivered to inform the outcome. “Let them know you’ll be working together to make good decisions,” suggests Knight. Ask for their opinion. Often they might have insight that no one else on your team does.

Establish open communication: Maintaining and nurturing your relationships depends on honest, “open communication,” according to Knight. “The quants must be willing to help you understand the estimates and assumptions in the numbers and help you figure out where the risks lie.” Along the way, you need to ask a lot of questions, he says. “There is a tendency to think of numbers as hard and inflexible but that’s not true. The data are inexact.” Davenport suggests pressing the quants in your organization about their models by asking questions like, What are the assumptions you’re using in this model? Under what conditions might those assumptions become invalid? And, how well do the sample data represent the population? “Don’t be afraid to admit what you don’t know,” he says. Ask for help in understanding anything that’s not clear to you.

Don’t get overwhelmed by the volume of data: Often, sifting through the sheer volume of data available is daunting. Fortunately, “all businesses run on a few key metrics,” Knight says. “You don’t need to be looking at 10 to 20. That’s too many. You need to understand the two or three that drive your organization’s profitability and cash flow.” To figure out which ones are most relevant to your business, he suggests enlisting the help of your quants to, “look at the cause-and-effect of all your metrics, and then weed through the ones that don’t have very much impact.” When you discover which metrics are most important, zero in. “You need to understand those in great detail so you can immediately see when there is a problem,” he says.

Respect the data: Data—not instincts or opinions—should drive decision-making, says Knight. The goal of any analytics team worth its salt is to provide an objective, unbiased perspective. So while “a good operations executive might have an intuitive feel for what could work in the business,” that manager should never pressure an analyst to seek out numbers to support his or her opinion. Not only does that pressure defeat the purpose of rigorous analysis, it also creates an environment where the data team feels it exists merely to please the leader. “Most people who do analytics want to see what the numbers reveal about a truth in the world, not support an executive hunch or prejudice,” says Davenport. Be willing to run experiments and trials to test your ideas and gut feelings. Of course there may be some reasons or “situations where you may want to overrule the data, but if you just cynically ignore what the numbers say, or ask an analyst to prove whatever you think is correct, it can be very damaging for your organization.”

* * *

Rebecca Knight then identifies “Principles to Remember,” suggests several dos and don’ts, and provides some mini-case studies that illustrate her key points.

Here is a direct link to the complete article.

She is a freelance journalist in Boston and a lecturer at Wesleyan University. Her work has been published in The New York Times, USA Today, and The Financial Times.