TDWI | Training & Research | Business Intelligence, Analytics, Big Data, Data Warehousing

Avoiding the Terminology Trap

The Terminology Trap has the potential to distract even the sharpest of tacks. How can we avoid terminological trivialities and keep the focus on the concrete outcomes that matter to people?

• By Steve Swoyer
• September 23, 2016

What does it mean for a company to digitalize? How is digitalization different from digitization?

The question stimulated a surprisingly animated discussion at the Pacific Northwest BI Summit earlier this year prompted by a slide in Gartner vice president Merv Adrian's presentation. Digitalization, Adrian and Gartner found, will be the number five priority for CIOs this year.

According to Gartner, digitalization continues to move higher in priority, even as the action items ahead of it -- "BI/Analytics," "Infrastructure and Data Center," "Cloud," and "ERP," ranked from one to four, respectively -- continue to trend downward.

This finding prompted a not-unreasonable question from industry luminary Claudia Imhoff, who was also in attendance: "What the heck is 'digitalization' anyway?" she asked.

Imhoff's question wasn't unreasonable for several reasons. The first is that Adrian's slide, sourced from a massive Gartner survey, seemed to conflate digitalization and digital marketing. (The relevant data point: 21 percent of CIOs cited "Digitalization/Digital Marketing" as a critical priority for 2016 -- up from 17 percent last year and just 11 percent in 2014.)

The second reason was that Jill Dyché, vice president of best practices at SAS Institute, had earlier presented on a "digitalization" that didn't seem to have anything to do with digital marketing: viz., the ongoing efforts of organizations -- particularly non-profits, municipalities, government agencies, and other public sector groups -- to digitize their "analog" or paper-based processes. Digitalization a la Dyché seemed to make immediate, even intuitive sense.

Gartner's conflation of the term digitalization with digital marketing? Not so much.

The third reason had to do with ethos: Claudia Imhoff helped define, create, and evangelize the corporate information factory (CIF). She's earned an oracular reputation in data warehousing, business intelligence (BI), and analytics. She doesn't lack for understanding or perspicacity. If the term "digitalization" isn't transparent to Claudia Imhoff, that isn't her fault.

As it happens, Imhoff was on to something. According to Gartner's own IT Glossary, digitization -- not digitalization -- "is the process of changing from analog to digital form." Digitalization, by contrast, "is the use of digital technologies to change a business model and provide new revenue and value-producing opportunities; it is the process of moving to a digital business." Just what is a digital business? Gartner's IT Glossary describes it thusly -- if not transparently: "Digital business is the creation of new business designs by blurring the digital and physical worlds."

Got that?

In his response to Imhoff's question at the BI Summit, Adrian offered a perfect on-the-spot definition. "It's a term that people can decide what it means to them," he joked, before continuing in a more serious vein: "'[They'll say] 'we are making our organization more digital,' 'we're going from paper to electronic, [to] automated decisioning, or to IoT.' Really, a lot of things can fit in here."

And they do, he might have added.

At the very least, the terminology is a distraction. The difference between "digitalization" and "digitization" is tantamount to the difference between "difference" and "différance": the two terms are so morphologically similar as to be identical. Anyone glancing at them on a screen (or, less likely, on a printed page) wouldn't immediately discern a difference. It's only when you look more closely -- e.g., if or when you have cause to pronounce them -- that you notice a difference.

Is it particularly helpful (or necessary) to parse the meanings of digitization, digitalization, digital business, digital marketing, and so on? Isn't it, rather, a productivity-squandering reduction to the absurd? Who, or what, exactly, is the intended audience for such terms?

To successfully codify terminology is to create new sources of revenue -- directly, via trademarks, service marks, licensing, and new business inquiries; indirectly, as a function of increased influence and prestige -- and, in effect, to construct a reality. To the degree that a vendor is able to lard the discourse with its own terms, definitions, and vision, it can shape -- if not control -- that discourse.

Would-be thought-leaders and influencers are desperate to coin their own (viral or meme-tic) terms; some, like Gartner, even maintain their own glossaries of terms or definitions. Recently, one media colleague told me that he, too, is expected to contribute to his organization's glossary of technology terms. The reason? An IT glossary is a reliable source of outside Web traffic. It drives advertising revenues and enhances visibility. Whom does this help, however? Certainly not us.

Dan Soceanu, a senior solutions architect with SAS Institute, picked up on this issue in a separate session at the Pacific Northwest BI Summit. The subject had nominally to do with data preparation; at one point, however, it, too, devolved into an exercise in clarifying the terminology. What is the difference between data prep and data wrangling? How do both terms relate to data engineering? Lastly, how does data prep (or data wrangling or data engineering) relate to traditional data integration? Soceanu had had enough. He expressed his -- and his customers' -- frustration with what he took to be industry navel-gazing and terminological vacuity.

It isn't that this stuff doesn't matter, Soceanu said; it's that it matters for the wrong reasons. We should be focusing on the kinds of concrete outcomes that matter to people, not on abstractions.

"I'm in the trenches with customers and data management [professionals]. To the customer, a lot of these definitions are esoteric and kind of not the point. They're trying to do things. They have a job they're trying to do," he said, not without exasperation. "Trying to finely parse all of these definitions and talk about does this mean this and does that mean that is a distraction. From the ... viewpoint of customers ... the definitions themselves are not really getting at the core need. The core need for customers is 'What am I trying to do in my job and how can I do it? I don't care what you call it!'"

Becky Sharpe, president and CEO of International Scholarship and Tuition Services Inc., a Nashville, Tenn.-based firm that manages scholarships for dozens of Global 2000 firms, says she's seen this same situation play out in countless meetings and face-to-face encounters.

"What I try to do when I'm talking to people is ... I try to lead with, for example, 'My desired outcome of this meeting is to determine how we can most efficiently answer requests for proposals when [customers] ask us about our internal security,'" she says. "That's a lot different than if I start out by asking, 'What kind of security do we have?' Then my IT people will go into our co-location [arrangements] and our highly detailed authentication [mechanisms] -- and before you know it, we've gone down a rabbit hole."

There's a contract that's implicit in discussion at this level, at least in the West: generally speaking, we assume good faith on the part of the other people with whom we're speaking. We assume they're communicating honestly and want to understand us; they assume the same.

If we lead with a question such as the one Sharpe cited ("What kind of security do we have?"), we aren't doing our part: we aren't being honest about what we want and need to know.

"When you don't have this, people will make all of these assumptions. They will assume that everybody is working from their reference point. I was online talking to my chief technical officer earlier today, and he knows me and I know him very well. I said, 'Joe, my goal is to reduce error and redundancy because right now the sales reps are going in and pulling data and doing manual calculations [on their own data sets]," she explained.

"I told him clearly, 'I want them to have a report that's updated dynamically but that they can't change. A report that accurately represents where we are right now with this data.' If I had asked him 'How often are these reports updated?' it would have been a different conversation."