In Tuesday’s Wall Street Journal, I review Loonshots, a new book on innovation written by biotech entrepreneur Safi Bahcall; the author’s key point is that innovation requires attention to organizational structure – it’s critical, Bahcall argues, to separate “artists” and “soldiers” yet allow their ideas to influence each other.
Bahcall is best-known as the co-founder and, for 13 years, the CEO of Synta Pharma, and was profiled for the New Yorker in 2010 by Malcolm Gladwell. Unfortunately, the oncology drugs the company was originally focused on never panned out, Bahcall resigned from the company in 2014, and the company merged with (and was essentially subsumed by) Madrigal in 2016, an example of a private company gaining access to public markets via a type of “reverse merger,” as the law firm Morrison Foerster discusses here.
The WSJ book review stands on it own, but there are three additional categories of observations that may be of interest to regular readers of this column: (1) Bahcall’s explicit swipes at Amgen and Merck, accusing them, essentially, of stolen valor; (2) Bahcall’s more general swipe at decision-making in big companies (which I thought was largely spot-on and is a topic Forbes colleague David Grainger and I have frequently discussed); and (3) a reminder that the best discussion of the challenges of innovation management is still found in Ed Catmull’s Creativity, Inc.; Bahcall, to his great credit, seems to have excellent taste in heroes, including both Catmull and famed Harvard surgeon-researcher Judah Folkman.
Let’s consider these one at a time.
Amgen, Merck: Stolen Valor?
I was struck by the direct and perhaps aggrieved shots Bahcall took on two biopharma companies, Amgen and Merck; if his stolen valor arguments are right, these companies earned praise at a time when they may have been more worthy of scorn.
Amgen: Irritated by a book attributing Amgen’s success to a winning culture, the result of “embracing the myriad of possible dangers,” Bahcall asserts that “the real story with Amgen” is that in the 1980s, the company was going down the tubes, was desperately rushing to develop a drug to stimulate red blood cell growth (erythropoietin), and managed to get there “just ahead of its competitors.” Bahcall states that much of this success was because an academic researcher, Eugene Goldwasser, had a critically important reagent (“an eight-milligram vial of purified protein, painstakingly extracted from 2550 liters of human urine”), and elected to give that to Amgen rather than its main competitor, Biogen, because “Biogen’s CEO had refused to pick up the check for dinner one night.”
Bahcall adds that after “winning the drug-discovery lottery,” Amgen “sued everyone else in the business (including its partner, Johnson & Johnson, which had saved Amgen when it was struggling) to stop them from competing. For the next fifteen years, Amgen was unable to repeat its drug discovery success.”
Notes Bahcall, bitterly, “Amgen may not have had good research but it did have good lawyers. It won every lawsuit, and its competitors gave up. Among insiders, the company was called ‘a law firm with a drug.’”
Bahcall’s point: think twice before “extracting culture tips, after the fact, from its terrific stock price performance.”
Merck: Bahcall narrates the story of the discovery of the statins, and in particular the pioneering work of Japanese researcher Akiro Endo, working at Sankyo. Bahcall notes that using a very similar approach to Endo’s, Merck researchers had discovered a very similar inhibitor to the one Endo identified – one that differed by only four atoms, according to Bahcall, and was discovered within days of starting their program, in contrast to the years it took Endo and colleagues, a happy accident Bahcall views with great suspicion.
Bahcall quotes Roy Vagelos, who was head of Merck research at the time, and later became Merck’s CEO (and is widely beloved and revered in the industry) as noting, humblebrag-style, that this “sudden” discovery was “unbelievable.” According to Vagelos, “as the pace picked up, the excitement steadily mounted. The competition with Sankyo heightened the thrill of discovery.”
Bahcall acidly observes,
“Competition, however, generally requires participants who believe they are competing. Every account that I have read by Merck scientists about the discovery of the statins has omitted a relevant detail: two and half years before Merck’s sudden discovery, Merck approached Endo and his team to collaborate rather than compete, requesting access to the most confidential proprietary data.”
Bahcall goes on to explain that Endo not only “provided samples” of Sankyo’s drug (with Sankyo’s approval) to Merck, but also “shared results from crucial experiments.” In so many words, Bahcall essentially suggests that Merck, led by Vagelos, pretended to collaborate with Endo but were actually competing, and ultimately won, which worked out rather well for both Merck (statins “became the most successful drugs in Merck’s history,” Bahcall tells us, tallying cumulative sales for Merck of over $ 90B) and Vagelos (who became CEO of an organization that was serially honored as “America’s Most Admired Company”); meanwhile, Bahcall says, Endo’s contributions “have gone largely unrecognized.”
As part of his emphasis on the importance of organizational structure (vs culture) in enabling innovation, Bahcall argues that once companies get beyond a certain size (which he pegs at around 150), most employees are more impacted by politics than by performance, an incentive structure that adversely impacts the prosecution of innovative ideas. A physicist by training, Bahcall likens this to a phase transition, the point at which the behavior of matter abruptly changes as it moves from one state (such as liquid) to another (ice).
(The phase transition analogy seems irresistible to physicists; for example, Douglas Robertson used the term in his 2003 book, Phase Change, to describe the impact of transformative new technology, like the microscope, telescope, or, he argues, computation. These technologies effect a phase change in a scientific discipline, fundamentally changing its essential nature. Just as the approaches used to predict the behavior of ice generally won’t be helpful when applied to liquid water, technologies foundationally alter the way a discipline evolves; in both cases, Robertson warns, “any attempt to extrapolate the behavior of a system across a phase change is doomed to failure.”)
To explain the impact of organizational structure on decision-making, Bahcall offers the following example: say you are a middle manager at Pfizer, he begins, and you are on a committee to review early stage projects, and there’s one that seems a bit heretical or unusual, but catches your fancy. What do you do?
Option one, says Bahcall, is “you could pound the table, make the case, and begin the long slog up the ladder of committees.” You “might be turned down” he notes, and even if you’re not, the drug will take years to develop. Moreover, “each time the project stumbles, at it inevitably will, the smiling back-patters, who wish you well at the beginning, will turn around and try to bury you and your project. They want your budget. And they want you out of their way.”
And even “if the project does succeed,” he notes, “you can count on 99,999 other people rushing into claim credit. If it fails, you can count on 99,999 people backing away, pointing at you. They will mention all those early stage warts, which you ignored so recklessly. Your career will be tainted. You might be fired.”
Now consider Bahcall’s option two:
“You could amusingly belittle the loonshot project, highlighting its flaws, poking at the warts…You advertise your wit, breadth of knowledge, and good judgment to everyone in the room. By fortunate coincidence, your summary of where the industry is headed happens to agree precisely with what your boss and maybe her boss believe. They laugh along and nod.”
Instead, Bahcall cynically suggests,
“(you propose) a modest step along a favored research direction. It’s a franchise project, a known quantity. It’s easy to greenlight all the way to the top. Everyone get it. If you continue to play smart politics… you might just be able to get your boss’s job. Perhaps as soon as next year. Which would increase your salary by 30 percent, not to mention double your prestige and influence. And the boost in title and salary could help you land an even high-paying job at another company, which you start looking around. Just in case anything goes wrong.”
Asks Bahcall, “so which do you choose?” As he tartly observes, “dismissing the loonshot in favor of the franchise project is the rational choice.”
In contrast, Bahcall suggests, the incentives are very different if “you work inside a small biotech company” – in this case, you are materially impacted by the success of your endeavor, and “sounding smart in meetings or trying to get your boss’s job is irrelevant. What matters is survival of (the project).” Here, “uniting to support” such a project “is the rational choice.”
While it’s not my experience that large companies have a monopoly on foolish or conservative decision-making, his description of the processes, mindset, and incentives of big pharma rings painfully true, a phenomenon I’ve described as “innovation dissipation” (basically, death by process) here, and here. See also these two pieces focused on the related way David Grainger (now a Forbes colleague) views this challenge: here, here.
Articulating The Creativity Challenge
Bahcall affectionately and appropriately cites both the late Harvard researcher Judah Folkman and the longtime head of Pixar, Ed Catmull, for their perspectives on innovation and its challenges – including Folkman’s famous comment that “you can tell a leader by counting the number of arrows in his ass.”
To my mind, the very best business book ever written about the management of innovative enterprises remains Ed Catmull’s Creativity, Inc., which I’ve discussed at length here, and in the context of biopharma here.
What I appreciated most about Catmull’s book is his deep recognition that in service of narratives, we tend not only to overestimate our ability to predict the future, but also consistently misinterpret the past, often because we insist on seeing patterns that may not actually be there.
Most importantly, Catmull recognized the unique and irreproducible nature of creative success – and how there isn’t a formula for achieving it. You need to figure it out anew each time. This contrasts profoundly with what I’ve seen repeatedly in pharma, where a narrative is developed around a blockbuster success, and senior management earnestly seeks to apply the “lessons learned” to future innovative activities.
“People want to hang on to things that work – stories that work, methods that work, strategies that work. You figure something out, it works, so you keep doing it – that’s what an organization that is committed to learning does. And as we become successful, our approaches are reinforced, and we become even more resistant to change.”
“I know that a lot of our successes came because we had pure intentions and great talent, and we did a lot of things right, but I also believe that attributing our successes solely to our own intelligence, without acknowledging the role of accidental events, diminishes us. We must acknowledge the random events that went our way, because acknowledging our good fortune—and not telling ourselves that everything we did was some stroke of genius—lets us make more realistic assessments and decisions. The existence of luck also reminds us that our activities are less repeatable.”
A related challenge, faced by those developing innovative drugs and those developing innovative films, is the difficulty (and often, the impossibility) of knowing in advance whether the innovative project you’re working on is actually going to pan out. Imagine the frustration of not knowing if the captivating new drug you’re championing is going to be the next cancer blockbuster or a costly, late-stage failure, or not knowing if the imaginative original film you’re give your life to will be the next Inside Out or the next Good Dinosaur. “The most creative people,” Catmull observed, “are willing to work in the shadow of uncertainty.
But uncertainty is an uneasy fit in the C-suite, as Nassim Taleb and I discussed in the Financial Times over a decade ago, and the abundance of data and metrics have, in some ways (as I’ve recently argued) only exacerbated the false precision and increased the sense of confidence that uncertainty can be domesticated, if only you had one more metric visualized on one more dashboard.
Given the structural impediments Bahcall describes, pharma’s often-expressed need for innovation will continue to struggle without the emergence of more leaders like Catmull, with the wisdom to dismiss tidy narratives, and the courage, and humility, to embrace uncertainty rather than insist on false precision.
As Catmull puts it,
“Creative people discover and realize their visions over time and through dedicated, protracted struggle…. I’ve known many people I consider to be creative geniuses, and not just at Pixar and Disney, yet I can’t remember a single one who could articulate exactly what this vision was that they were striving for when they started.”
Pharma desperately needs innovation; the question Bahcall’s book begs is whether senior pharma executives can structure their organizations in a fashion can liberate the creativity, and tolerate the uncertainty, innovation requires.
WSJ book review of Loonshots.
Forbes on ibrutinib discovery (also discussed in Loonshots)
Forbes on pembrolizumab discovery (referenced in WSJ review)
“Drug Research Needs Serendipity” – Commentary with Taleb in Financial Times.