Why Digital Is a State of Mind, Not Just a Skill Set

You attend webinars about artificial intelligence, follow machine learning gurus on Twitter, and scrutinize your company’s analytics.

And yet, something hasn’t quite clicked. Digital technology hasn’t infiltrated every aspect of your organization, and it feels like you have a long way to go to catch up to peers.

"To be successful, it is necessary to think differently."

You’re not alone. Today, thanks to the digital zeitgeist, business leaders have much more to learn and master to be successful in a modern organization, according to a new book by Harvard Business School Professor Tsedal Neeley and digital transformation scholar Paul Leonardi.

“Thriving in the digital age requires more than simply acquiring skills to work with digital technologies. To be successful, it is necessary to think differently,” write the authors of The Digital Mindset: What It Really Takes to Thrive in the Age of Data, Algorithms, and AI.

For example, business leaders must understand:

• The foundational principles of new technology. This includes the basic tenets of coding, programming languages, scripts, algorithms, compiling, and machine language. This knowledge is crucial for understanding how digital applications are programmed and how computers are made to execute.
• Biases in the technology. It’s important to be aware of the biases of artificial intelligence (AI) and how machines can learn without being explicitly programmed by humans.
• How to challenge data. Leaders need to ask how data was produced, who had access to it, and how well it represents the behavior organizations hope to understand.

It’s important to have a keen digital awareness, say Neeley, the Naylor Fitzhugh Professor of Business Administration and senior associate dean of faculty development and research strategy at Harvard Business School, and Leonardi, a professor at the University of California, Santa Barbara.

Lacking a digital awareness would make it difficult to participate in the digital economy, to take advantage of the data-intensive ways of informing key decision making. “This also means,” Neeley adds, “we don't have the capability of running organizations that are impacted by digital technology.”

"You only need to develop 30 percent fluency in a handful of technical topics to cultivate a digital mindset."

Here’s the good news: Attaining a digital mindset does not mean learning to program in Ruby on Rails or training a machine-learning application, she says. Rather, she recommends applying the 30 percent rule. Instead of needing to gain mastery over every digital skill, “you only need to develop 30 percent fluency in a handful of technical topics to cultivate a digital mindset,” says Neeley.

Why develop a digital mindset?

According to Neeley, a digital mindset is not just thinking about what digital is; it involves a wholesale revision of how we think.

“We need to see through the lens of data. We need to see through the lens of technology. We need to see through the lens of organizational design and change. It's to begin to see through the prism of all that digital brings when it comes to your own work. With that, you need a whole new approach and a set of skills and tools.”

For example, how do you collaborate successfully with machines? That’s important because, thanks to AI, machines are increasingly becoming our workmates, maintaining our calendars, and guiding our decisions, Neeley says. Although machines are becoming adept at imitating human interaction—it’s entirely possible to “talk” to a robot without knowing it—the authors say we should treat them as machines, not humans.

The book offers the example of University of California, Los Angeles professor Burt Swanson, who tried to schedule a meeting time with a colleague using the popular automated scheduling agent “Amy.” In dealing with a robot driven by AI, Swanson communicated using colloquialisms and conventions from informal conversation. The result: Swanson was frustrated by Amy’s rapid-fire dismissal of his available times—and she even cancelled a confirmed meeting when the colleague had a new conflict.

“Thoroughly frustrated, Swanson typed out a lengthy response to Amy expressing how unpleasant the experience … had been …,” the book relates. “Amy never wrote back.”

“Recognizing that an AI agent cannot accurately infer your intentions means that it’s important to spell out each step of the process and be clear about what you want to accomplish,” the book says.

Creating a digital presence

The ascendancy of remote work has raised a new question among the current workforce: How do work-at-home employees ensure their colleagues feel their presence when they are not in the same room? The authors say workers must develop a “digital presence.”

“It's about using tools. It's about understanding the use of synchronous versus asynchronous communication, and it's also figuring out, how do you get to know people formally and informally,” says Neeley.

The authors advocate using social tools to connect on a personal as well as social level. Studies show that social exchanges on business channels like Slack—about hobbies, pets, and vacation experiences—lead people to get a sense of who you are, to feel comfortable about you, and to trust you.

“If you are mostly engaging with others through digital tools … it's important to introduce structured-unstructured time so that people are able to engage in the non-work elements of their world, which always makes people feel more connected and cohesive,” Neeley says.

Remote workers must also be cognizant of the “Mutual Knowledge Problem,” which is lack of shared understanding of place, context, emotionality, and other things that help people get on the same page.

Three techniques the authors recommend to combat MKP are to send frequent updates, create a sense of curiosity, and communicate on the other person’s timeline, not yours.

Data is not truth

In many ways, data is the DNA of organizational decision-making. But data is not necessarily truth; it’s information that must be analyzed and challenged, the authors say. Someone lacking a digital mindset can easily be fooled into accepting data as gospel.

Data will never be unbiased, Neeley says, because biased humans gather data, interpret data, and sometimes build models that don't take into account potential risks and harms from technologies derived from misunderstood or incomplete data.

“A digital mindset requires us to fully understand how to think about data, how to analyze data, and how to ask all of the right questions to ensure that no harms or risks are embedded in them as well,” she says.

"How you present the results of your analytics matters immensely."

One potential remedy is to embed ethical or bias reviews when big data is being used and to determine which group will police the data so harmful outcomes are not being created.

Another key to using data effectively is to understand how to present it. Too much information, or too little, can cause viewers to draw the wrong conclusions.

For example, towns use AI during public forums with residents about future development plans. When potential projects are depicted with a lot of detail, like 3D visualizations of building shapes and locations, residents tend to consider those depictions as final solutions rather than just concepts to be debated. Conversely, presentations that are rich in data but light in visuals make it difficult for participants to reach consensus.

The lesson for practitioners, according to the authors, is that data needs to be interpreted and explained. “How you present the results of your analytics matters immensely,” Neeley says.

The digital leader

Digital leaders must be in a perpetual state of inventing, reinventing, and transitioning, the authors stress.

Perhaps most of all, achieving a digital mindset means overcoming a fear of technology, Neeley says.

“People cannot be afraid of technology,” says Neeley. “They cannot be afraid of data work. They cannot be afraid of entering an era where they have to learn something new every day. You have to understand how machines learn because otherwise, you won't be the one leading your organization.”

Related reading from the Working Knowledge Archives

Developing the Global Leader

Feedback or ideas to share? Email the Working Knowledge team at workingknowledge@hbs.edu.

[Unsplash/Maximalfocus]

Without a properly developed digital mindset, even the people who know the advantages that digital experimentation can bring—like the senior manager of the auto company—don’t see the world through an experimentation frame. We’ve also seen that managers and employees like Balaji and Davide who intuitively embrace experiments don’t always have the right vocabulary or frame of reference to successfully push their ideas. Both are reasons why a digital mindset for everyone on a team or organization is essential for creating a healthy culture of digital experimentation.

The idea of experimentation as a core business practice isn’t new, but the complex, data-intensive nature of experimentation is. With digital innovations requiring the integration of software, hardware, and organizational layers, products and services are becoming highly integrated and computationally driven. Such complexity makes the experimentation process both more complicated and more essential for success in the digital age. Recent research suggests that organizations that can create a healthy culture of digital experimentation grow at least eight times faster than global GDP.  Microsoft attributes annual revenue increases of 10 percent to 25 percent to monthly improvements that it makes on its Bing search engine through constant experimentation. One large bank that we worked with saw a 30 percent increase in revenue over a two-year period from the introduction of new financial products honed through digital experimentation.

The capacity to reap the benefits of rapid prototyping and data analysis to improve internal work processes, products, and services comes from a culture that embraces the digital mindset in experimentation. Developing that mindset requires understanding four core approaches:

1. Develop a learning agenda that will set the tone for experiments and provide a framework for making sense out of them.
2. Turn the digital exhaust generated by user behavior and systems logs into digital footprints that can be systematically analyzed
   to create the design of future experiments.
3. Prevent current organizational structures, such as how work is divided among people or departments, from stymieing experimentation efforts.
4. Build a culture of psychological safety in which experimentation is not just encouraged but expected (even when it’s experimentation about experimentation).

These approaches are significant whether or not you are actually doing digital experiments. Understanding their basic concepts plays into key technological processes that affect our work and our lives.

Reprinted by permission of Harvard Business Review Press. Excerpted from THE DIGITAL MINDSET: What It Really Takes to Thrive in the Age of Data, Algorithms, and AI by Paul Leonardi and Tsedal Neeley. Copyright 2022 Harvard Business School Publishing Corporation. All rights reserved.