In-Depth

Big Questions Simmered for Data Analysts in 2011

The role of data analyst has struggled for clarity on standards this year as the importance of analytics advances.

• By Ted Cuzzillo
• 12/13/2011

The business intelligence industry smiled at itself in the mirror again this year and said, "What big data we have!" -- while more interesting trends simmered among those who make sense of that data.

The role of data analyst seems to have edged forward this year and struggled for clarification. I see four trends.

Trend #1: Confusion about what, exactly, a data analyst is

They're everywhere yet hard to recruit. Some recruiters wouldn't know a real analyst from an adminstrative assistant, and some candidates don't know mean from median. They have been called business analysts, "doc," fakers, triage nurses, and battleground medics -- while increasingly more don't call themselves data analysts any more than they call themselves "telephonists" because they use a phone.

Meanwhile, software vendors talk about analytics as if no human hand needs to so much as touch the dial. It's all automatic! Some companies -- including a few producers of analytics software -- have tried to combine data analysis with a loose collection of other, unrelated tasks. Is the job really so easy?

"Many people have created fantasies of what analytic talent is or should be," wrote Meta S. Brown, general manager of analytics at the language technology firm LinguaSys, in an e-mail to me, "fantasies that bear only as much resemblance to real analysts as the cover of a fashion magazine does to a real woman."

Trend #2: Nascent standards

In a more hopeful trend, Brown and other members of the analytical community hope to bring standards to bear. Brown reports that she's watching an effort in the UK that she and others may promote in the U.S.

An attempt at best practices, meanwhile, comes from Blake Johnson, a consulting professor at Stanford University. I wrote about his ideas here in early October. [

Most interesting are his practices, include forming groups of data analysts who would take over some data management functions from IT and some strategy and metrics functions from business. That is, a new formation in what had been the no-man's-land between IT and business that would be a power all its own. You can imagine what trends that might ignite.

Trend #3: Instant analysts

Like instant pudding, their analysis is usually good enough. They know their business, they know their data -- and they know their desktop tool. Is what they call analysis really just reporting? Who cares? They get results, and that's proof enough.

Their success varies. Seattle Children's Hospital, for example, had success this year when more than 100 staff quickly adopted one tool and now analyze data routinely, according to Tom Corbett, director of knowledge management.

Of course, Seattle Children's is like any modern hospital in that it bursts with metrics, high stakes, and smart people -- most or all of whom are not professional data analysts. Among the results of Tableau's permeation was a 30 percent reduction in patient identification errors.

The instant analysts include semi-pros and those whose other training has prepared them for analysis.

Other parts of the class, however, include "fakers," as they're called by Theresa Doyon, marketing statistical analyst at World Travel Holdings. They can't hold up for even five minutes with a real analyst, she e-mailed me, but unlike many professional analysts, they tend to have good people skills.

Trend #4: IT walls continued to fall down

Much of the data that analysts require to do their jobs comes from official sources, but some doesn't. The determined analyst -- especially the so-called renegades, the "cowboys" -- have been known to do whatever is necessary to access the data they need. This year, the trend to ease barriers seems to have continued -- thanks mostly to a small crowd of intelligent, creative, collaborative analysts who've sustained their focus on the ultimate work of business intelligence. All the better to guide a business with.

Ted Cuzzillo is an industry analyst and journalist with more than 20 years' experience explaining, analyzing, and researching how people use technology. To paraphrase Lance Armstrong, who says "It's not about the bike," Ted says, "It's not about the computer." His current research focuses on business analysts, including the tools they use, the roles they play, and their careers. He can be reached at analysts@datadoodle.com.