Episode 16: Technology and AI Thought Leader Tom Davenport

Transcript: Tom Davenport and Alan Alper

Alan Alper: Tom, why don’t we start with something pretty straightforward: How did you get started in management consulting research with a focus on IT?

Tom Davenport: Well, I was an academic. And I didn’t really like the fact that the stuff that I wrote would be read by five or six people around the globe. So I got more and more interested in consulting, I had been a user assistant for statistical computing in graduate school, and then actually became head of end user computing for Harvard’s Computing Center for a little while. And so that was why I was interested in the technology. And I remember I got a Harvard Business School guide to management consulting, and I thought it was really interesting. I had my eye on one company in particular company called Index, where your colleague Bob Buday worked. Nobody was taking me terribly seriously as a PhD in sociology in the consulting world, so I finally decided I better go to business school. So I applied to Harvard and MIT. I got rejected from Harvard Business School, even though I had a PhD from Harvard in four years, which is about as quick as anybody gets one. But I did get into MIT and was ready to go…and a couple of weeks before I was supposed to start, I got a job offer from Index. I had done a little bit of consulting so I took it, and but then I immediately was assigned to a project working with Mike Hammer and Walter Popper on end user computing. I really liked that. And I realized I liked research and thought leadership better than I liked mainstream consulting. So that’s basically what I’ve done, either in consulting firms or business schools, for almost 40 years.

Alan: So you did have a real solid background, at least to some degree, in IT. You just didn’t glom on to the topic because it was so hot in those days; you actually had some hands-on experience.

Tom: Yeah. And it wasn’t as hot then as it is now. It was starting to get hot in the kind of popular imagination. But I had a lot of exposure to one aspect of IT — not so much the transactional side, but what you do with data. That’s always been more interesting to me anyway.

From Consulting to Thought Leadership

Alan: So how did you make the transition from consulting to thought leadership? Can you talk about your formative experiences here?

Tom: I worked a lot with the late Mike Hammer, and he was certainly a good mentor. I remember going to my first meeting with Harvard Business Review with Mike, and we had a list of things we wanted to write about. One of them was about this reengineering idea. HBR wasn’t interested. But they were interested in an article about how senior executives can get involved in decision making through what we called a principles-based approach to information architecture. So that was my first HBR article. They stuck a guy’s name on it who had nothing to do with it — as consulting firms often do. But I thought, this is fun. I like this and I’m going to keep on doing it. And then I wrote this article on reengineering when I left Index — the first article, by a few weeks anyway, but certainly not the best-known article. A few months later I decided I should probably write a book on this. So I did. And it was the first book on reengineering – but not the most popular one either.

Alan: So it’s interesting. While HBR rejected the whole concept of business process reengineering, it obviously caught on pretty quickly. Can you talk a little bit about how it started to germinate and take hold, and the role the thought leadership played in kind of getting out in front of the demand for this kind of service?

Tom: I think the timing was just fortunate. We were very concerned at the time that Japanese companies were going to eat our lunch with quality and the fifth-generation computing Initiative and so on. There had been a lot of interest in quality management, but Americans typically are not patient enough for long-term quality management. They prefer home runs to walks or singles. So I think reengineering was very appealing in that regard. And so often these thought leadership ideas, their success and failure, is result of timing, and the timing was really good for that. I think we were probably entering a recession too; and people were looking for ways to get out of it. Index wasn’t terribly well known, but they did a good job of harnessing publicity. They got some good ghost writers to write the book, which became a good introduction to get people excited.

A Crowded Marketplace for Ideas

Alan: So how has thought leadership changed over the years in which you’ve been involved? What kind of skills, structured thinking, creative instincts are really required to be successful as a thought leader?

Tom: I don’t think that the skills have necessarily changed, except that being a traditional sort of researcher, writer, or academic is probably less valuable now just because the distribution channels for management ideas have exploded. I’m not on TikTok yet, but I sometimes wonder if I should be. In fiction, for example, you have the Colleen Hoover phenomenon, where most of her fame and fortune and very high positions on the New York Times bestseller lists are basically attributable to TikTok. I don’t think anybody’s exploited that too much yet on the management idea side, but, it’s probably only a matter of time. So I think you have to be open to those new channels and think about how you get the message out in all sorts of ways. I remember a former editor of the Harvard Business Review, writing in a book called “The Lords of Strategy,” who wrote that writing one Harvard Business Review article would get you enough consulting work to keep your firm going for a couple of years. And that’s hardly true anymore. You write one HB article, and then write a couple of other digital articles for HBR and Sloan Management Review and various trade press publications, and 14 LinkedIn posts and 36 podcasts? Maybe you’ll get somebody’s attention.

How Big Ideas Are Hatched

Alan: So you’ve talked a lot about the dissemination and the publicity for your thinking. How about the actual crafting of your big ideas, and how you research and validate your notions and how you then bring them to life through books and other media?

Tom: I think that at least in my case, you really have to get out into the world and talk to a lot of people. I try to adopt and research business ideas at a relatively early stage. I don’t think of myself as inventing them. Business process reengineering wasn’t invented by me or Mike Hammer or Jim Champy, but by companies that had problems to solve. So I think getting close to practice is really important.  At Index, Mike Hammer and I co-founded a research program called PRISM, which stands for Partnership for Research and Information Systems Management. It was research funded by a bunch of companies, and we found out about what they were doing in different areas. This was great fodder for writing about them. And I’ve had a number of those programs since then. They’re frankly a lot of work to manage, and I’m not doing it now; I just call up companies and talk to them. But when they’re paying you to do research, every six months or so you’d have face-to-face meetings where you’d have to come up with something interesting to say. Fear of embarrassment is a powerful motivator. So it did drive a lot of thinking about how we package up these ideas in a way that’s interesting and helpful to these companies. Beyond the money, it’s the contact with companies that that really helps.

Getting Companies to Talk Candidly

Alan: You do a lot of qualitative interviews in your books. You have to go out and find the people who are the movers and the shakers and get them to talk to you, and be able to explore to a level of depth that can help to reinforce some of your ideas. How do you do that? It can’t be easy to do.

Tom: It’s easier now than it used to be, because I get a lot of emails from PR firms telling me about a startup that has done something really fantastic and asking if I’d write about them. I always say that maybe I’ll write about them, but it has to be in the context of what one of their customers is doing with their fantastic invention. If there’s any benefit to having written 23 books and maybe 220 Harvard Business Review articles, it’s that most people have some idea of who I am. It still takes a big effort to talk to these companies and synthesize what they say. I always try to get into an area early, when the concept is not so well understood that you could easily do a survey or something like that. But I have done a number of surveys, and I’m happy to do that if we want to get a broader reading on what’s happening in the world at large.

Alan: In your latest book, “All In On AI,” you are talking to established businesses about the use of AI. In some cases, it may be considered confidential or sensitive how they’re applying the various algorithms and thinking and acquiring the human resources necessary to make it work. So you have to skirt a fine line between really revealing enough to show your thinking, without disclosing anything that your interview subject may feel is going to put them in a bad situation because it shouldn’t be public domain.

Tom: I always say that tell them that I won’t write anything that they can’t see ahead of time. I know this violates journalistic principles, but I’m not a journalist. Nobody wants to be embarrassed in print. Sometimes it does mean that you have to cut out something that is essential to really, truly understanding what’s going on in a company. But there is also the mechanism of describing companies anonymously, and I do that a fair amount too. If you’re going to say something bad about a company, you probably will have to anonymize it. I do try to mention actual companies as often as I can.

Alan: I think at the end of the day, it’s more credible and believable for the reader when it’s a real-world scenario, not just some hypothetical thing. You’ve done your research and you’re not going to put your name on something that’s made up.  

Tom:  It’s amazing how many things I see from consulting firms without a single real-life example. It drives me crazy. I don’t know how they think they can get away with that.

Working With Co-Authors

Alan:  Tom, in your latest book about AI you worked with a co-author who’s a partner over at Deloitte. You’ve co-authored lots of books over the years — what’s it like co-authoring with somebody? How do you find that right balance between your ideas and your co-author’s ideas, and how do you consolidate them into one point of view?

Tom: Writing a book is a fairly lonely activity, and I generally like it better when I have somebody to talk to about the ideas. Some co-author relationships are very productive; they do at least half of the work. Some don’t do that much, but they supply some ideas. I don’t think I’ve ever had a co-author who didn’t at least talk to me occasionally about what I was writing. But writing comes pretty easily to me; I’ve never had writer’s block. It’s worth it to me to have somebody who is engaged in the topic and willing to read it.

Alan: So jamming can unlock even better and bigger ideas by just bouncing things off of people.

Tom: In Deloitte, for example, I got access to some companies that I wouldn’t have known about. This does mean you have to have everything reviewed by the risk department and the client service partner, or whoever controls that relationship. Fortunately other people handle most of that process for me.

Hits and Misses

Alan: When we spoke last October, you were prepping for your presentation at our Profiting From Thought Leadership conference. You were quite upfront about certain concepts that weren’t as successful as you would have liked or thought they might be — such as the attention economy, knowledge worker productivity and performance, how to make good decisions, things like that. Why didn’t those things work out? And what would you have done differently or better if you could do it over again?

Tom: Almost everybody has some offerings that don’t succeed as well as others. Sometimes it’s a matter of timing. “The Attention Economy” was one of the most fun books I had worked on. We tried to be very attention-getting with how we wrote the book, and we included lots of stories. But it came out on September 10, 2001 and there wasn’t a whole lot we could do to get the world’s attention when something like 911 was going on. “Judgment Calls” talked about how organizations make good decisions. We found that people would rather read about bad decisions than good decisions. Maybe that’s just human nature. Still, even the less-successful books were not total disasters; they might have sold 10,000 copies. I don’t regret writing any of them.

Alan: So let’s talk about the successes and what made them that way. Why do you think they stood out and they resonated with the target audience?

Tom: I think the timing was very good for the reengineering book. My next big best seller was the knowledge management book, “Working Knowledge,” and people were very interested in the new technologies that had emerged around that time, such as Lotus Notes, that promised to capture and distribute knowledge much more effectively around an organization. We were starting to realize that the value of a company was much more in its intangible assets than in its tangible assets, certainly at tech companies. The timing was also right for my book about competing on analytics. Data was popping up all over the place, and companies had to figure out something to do with it.

I think timing is probably a more critical factor in the success of books than anything else. Sometimes when you start writing about something the timing may not be right, but it’s the right timing by the time the book comes out. I didn’t write a book on E-commerce, because I knew other people had moved more quickly into that space than I did. I did just write an article on generative AI in HBR, but it was published a week or two before chat GPT came out, and suddenly it was totally swamped by millions of words on chat GPT. Somebody recently proposed to me that we co-author a book about generative AI. There aren’t great books on that topic now, but by the time we finish there will be a lot of books and it’ll be a very crowded field.

Working With Top-Tier Publications

Alan: You’ve written so many articles for HBR and other publications of its ilk. I think if HBR had an in-house columnist, you would be among them. How did you crack that nut? It’s the mecca of business management journals.

Tom: Getting to know the editors in your space helps a lot. And also writing relatively high-quality drafts that they don’t have to do a lot of work with. I recently wrote an article for print – I don’t write as many print articles as I used to; I think I’ve done 35 print articles. It takes so much longer to get print articles published, and the acceptance rate is lower. But the editor immediately asked me what examples I was going to use, and that’s great. I do write a column now for MIT Sloan Management Review. Unfortunately, there aren’t that many places for articles that are semi-respectable from an academic standpoint, and also read by practitioners. California Management Review is one of them, but hardly anybody knows about it or reads it. And there’s this new one called Management and Business Review, which is sponsored by a lot of different business schools. I hope that one catches on, and I’ve written a couple of pieces for it. But HBR is still at the top, as a focus of interest by business people. It’s probably less than it used to be since there’s just so much content out there, but it’s still the probably the single best place to publish.

Alan: You told me that you sometimes use HBR as a testbed for ideas, things you’re considering maybe writing longer-form content on like books. Can you give me a sense of how that works?

Tom: Well, I don’t know if it works anymore, frankly. I am co-authoring a book now about the citizen revolution: technology citizen, data science citizen, automation citizen…low code, no code use. We decided that we’d write an HBR article on it. But sometimes you write things in HBR and nobody ever says anything about them. So while it’s quite hard to gauge the reaction, it certainly helps to get a smart editor involved early on to give you some feedback on it.

Generative AI and Thought Leadership

Alan: So we’ve talked a lot about generative AI, and so much has been written about it already. I’m wondering: What role do you see such tools playing in thought leadership over the near and midterm? Are they going to be part of the research, writing and validation process for big ideas, and how they are developed and brought to the marketplace?

Tom: I think it’s going to take over most of the lousy thought leadership out there. It’s not good at all at describing things that haven’t already been written about, only things that have already been written about. So if you want to write about leadership…sure, why not, because there are millions of articles about leadership already and it probably could do a fairly good job of that. If you want to write about something that hasn’t really been discussed before, it’s probably a bad bet. My co-author on this citizen development book asked chat GPT to describe the literature already in that space, and it was really quite helpful. So I think it can accelerate the research process a little bit, and at least look at what’s been published already, but it’s not terribly good at coming up with new ideas. I’m sure there will be entire books written by chat GPT. But my guess is they won’t be terribly successful.

Alan: Maybe it will augment human capability, and take over some of the the manual, labor-intensive work of doing the desk research that is necessary to figure out what’s been written and, and where your ideas are maybe a little different. But I think if you’re a thought leader, and you’re quality thinker and writer, you’re probably not going to to worry about chat GPT taking your job anytime soon. Tom, before you go: How do you see the thought leadership profession evolving over the next few years?

Tom: Your company, Buday Thought Leadership Partners, has done some interesting research about the number of different organizations who are now interested in pursuing thought leadership. Even consulting isn’t at the top of the list anymore at all. I was reading an article about how Google has fallen behind in the generative AI space, even though they were doing research on it a long time ago. The article mentioned the race for thought leadership in that space. So I think thought leadership is becoming a very broadly understood concept that applies to almost every business. It’s a way to get people interested in what you’re doing – what your services are about, and how organizations are using them in interesting, innovative ways.

Nature versus Nurture

Alan: Do you think thought leaders are born? Or can they be developed?

Tom: I would like to think that a certain ability to express oneself is necessary. None of us are born with that, so I think that can be developed. I think some people are more intellectually curious than others, and this skill is harder to inculcate. It’s essential to have a basic set of basic capabilities: clear expression, good thinking and intellectual curiosity.

Future Plans

Alan: What’s next for Tom Davenport? In thought leadership and beyond?

Tom: I’ve written five books on AI, and I’m thinking maybe that’s enough. But organizations and individuals are always changing how they use information and technology. If generative AI really takes off as a business capability, I suppose I could be persuaded to write something else about it. I’m quite interested in the whole profession of data science, which is still new. How much does it require professionals? What does it mean for a company to really commit wholeheartedly to data science? Those are interesting issues to me, and I’ll continue to explore them. So you know, I want to write at least 25 books. Maybe I’ll quit after that.

Alan: Thanks, Tom, for a wonderful conversation. Appreciate you taking the time to speak with us today.