Is Your Org Chart Stuck in a Rut? Try a Scientific Experiment

Harvard Business School’s Organizational Lab helps businesses solve operational problems by changing the way their company is organized. Ethan Bernstein and Clayton Christensen explain how academics and business leaders can work together to create a better org chart.
by Carmen Nobel

If you want to be awed by the pace of technological advancement over the past few decades, compare the capabilities of a bulky PC from 1984 with those of a sleek smartphone in 2016. You’ll find stark differences. But if you want to be underwhelmed, try comparing a company’s organizational structure chart from the same two years. Chances are, those two charts look pretty similar.

Sure, there are organizational outliers, like online shoe-seller Zappos, which famously adopted “holacracy,” a structure of no job titles and no bureaucracy, giving teams the power to reorganize themselves at will. But with few exceptions, organizational structure has failed to keep up with other market-fueled technological innovation. As a result, a lot of companies are stuck in a structure rut, and aren’t operating as smoothly as they could be.

“We have this sort of binge diet kind of process of restructuring organizations that really ends up with no trend line”

“For decades, we’ve been in this cycle where an organization hires a consulting firm to help design and execute a restructuring, and then, as soon as they’re done, they start that process again,” says Ethan Bernstein, an assistant professor in the Organizational Behavior unit at Harvard Business School, who spent five years as a management consultant at the Boston Consulting Group (BCG) before joining the HBS faculty. “And so we have this sort of binge diet kind of process of restructuring organizations that really ends up with no trend line.”

Bernstein believes the key to breaking that cycle may lie in research partnerships that help organizations conduct more field experiments. Just as many companies test out new products with “a/b” testing, they might also take an experimental approach to discovering the best organizational structure for their business.

“I think that most executives look at bold experiments like [CEO] Tony Hsieh’s experiment with holacracy at Zappos and think, wow, I’d really like to experiment with my organizational structure too, maybe not that radically, but with the potential for a similarly large impact on how people work together productively,” Bernstein says. “But they don’t know where to start, or, if they do, all they see are roadblocks in the way.”

To that end, he and several colleagues recently launched the Harvard Business School Organizational Lab. A collaborative effort with the HBS Digital Initiative and the HBX Live virtual classroom, the o-Lab aims to help businesses solve their problems by changing the way their company is organized, and to make that change by introducing a more theoretical and experimental approach to organizational restructuring.

As things stand today, restructuring can feel like a “just because” pursuit among companies, which often coincides with a change in C-suite leadership.

“If you bring in an executive from the outside, it’s very common that, as soon as they get into their chair, they want to reorganize,” says Clayton Christensen, the Kim B. Clark Professor of Business Administration at HBS. “The fact that they keep doing it that way means that there’s not a theory of the organization. Nobody can say, ‘In this situation, based on theory, this is the way we should reorganize.’ So there’s a lot of waste that goes on, and opportunities that are not pursued.”

To be clear, the o-Lab is not out to question the value of business consultants or CEOs. Rather, it aims to illustrate the value of the tools of business research.

“If we, the HBS faculty, are going to claim that we have theories about organizations that are practical and relevant for the parties who are running those organizations, then we should try to see if we can break the cycle with something that lets an organization learn over time,” Bernstein says. “And that would seem to be a more experimental approach than a planned approach.”

Crowdsourcing research

Earlier this year, the o-Lab issued an open call for businesses to share challenges that were difficult to overcome due to their current organizational structures. Some 44 people submitted problems via a platform that let everyone view everyone else’s ideas online.

One submitter explained that her company was transitioning from producing boutique products for a few customers to more mainstream products for a wider customer base, and wondered how to reorganize her company’s team structure accordingly. Another wanted to figure out how to maintain a culture of entrepreneurship among local offices as his company continued to grow globally. Another, a seasoned lawyer, set out to evolve the organizational structure of his firm.

“The emergence of new technologies and the trend toward unbundling legal services into their component parts means the structures traditionally used by large law firms are being increasingly challenged,” says Robert Regan, a partner at Corrs Chambers Westgarth, a law firm in Sydney, Australia. “To respond to this, and deal with falling margins, law firms need to find new organizing principles.”

In March, Bernstein teamed up with Christensen to lead an online discussion, hosted by HBX Live, on how to design an organizational experiment. They started by making sure the participants understood the practical importance of theory. “Behind every massive experiment there’s a theory,” Bernstein says. (He notes that the majority of today’s organizations are largely based on the theory of late nineteenth century economist Max Weber; it’s called bureaucracy.)

At the HBX Live event, Christensen explained why business practitioners should care about theory in the first place.

“We teach our students to be data-driven, fact-based, and analytical in their decision-making processes,” said Christensen. “But if data is only available about the past, we’re essentially teaching our students only to take action when the game is over. A better way to think about the problem is to be able to look into the future and predict what’s going to happen in the future. And if there’s no data, then the only way I can look into the future is if somebody has given us a theory of causality. That allows us to say if we do this, then that will happen. So theory is of immense value to managers.”

Click to watch.

Following the event, six participants submitted experiments for consideration, four of which made it to the “refinement” stage, including Regan’s.

“The hypothesis that we are testing is that if a complex legal brief is viewed firstly as a bundle of tasks to be integrated to create a customer outcome, productivity and collaboration will be improved,” Regan says. “If this is established it could drive a reorganization of our existing organizational structures.”

Other participants in the refinement stage include a health care professional, who proposed an experiment around the idea of “creating cross organizational teams that promote collaboration and alignment in managing patient care,” and an MIT lecturer, who wants to test out expedient ways to combine the best aspects of big data and crowdsourcing. They will be reporting their progress for the next several weeks, inviting input from Bernstein, Christensen, HBS research associate Kathy Qu, and the rest of the o-Lab community.

Problems and potentials

Neither Bernstein nor Christensen is certain that the o-Lab will yield viable new theories about organizational structure. Theories are statements of causality, which help to predict the future outcomes of current actions. The o-Lab experimenters may be more focused on immediate goals that are specific to an individual organization than on predictive outcomes that are generalizable to all organizations.

“That’s the reality of using field experiments for research—it’s never easy,” Bernstein says. “In research, field experiments fail at least as often as they succeed. We’re hoping our partners can help us do more of them and make us better at them, all at the same time.

The challenge about this, to be frank, is that the theoretical reasons for conducting one experiment or another are going to be all over the map,” Bernstein says. “Picking a goal target over a theoretical target always makes for a more difficult synthesis. But that’s the fun part too: taking a bunch of interesting experiments and data and figuring out how to make it useful to us as a research institution and to us as organizational scholars.”

In the process, Bernstein is convinced that business practitioners will benefit from the mere idea of experimenting with organizational structure.

“Theory is of immense value to managers”

“Too many people assume their organization’s structure is a given,” he says, and then gestures to the heavy oak desk in his office. “It’s kind of the way I assume it’s a given that I can’t move this desk. So, I don’t try. But if you were to give me a desk made out of Legos, I might move the Legos around. I’m kind of used to moving Legos around, because I have a three-and-a-half-year-old. So the o-Lab is about that: Gee, if we could break an assumption and provide the tool kit and get a community of support around it, maybe we could get more experiments done, and we could make more progress as a result.”

To learn more about the o-Lab and the creation of business theories:

About the Author

Carmen Nobel is the senior editor of Harvard Business School Working Knowledge.

Post A Comment

In order to be published, comments must be on-topic and civil in tone, with no name calling or personal attacks. Your comment may be edited for clarity and length.