This is a topic that came up in the comments section of a post last week, but it’s important enough that I wanted to give it some exposure here on the front page. It was a question from someone outside the industry, who asked about how companies can retain “institutional memory”, what with all the well-documented instability (mergers, layoffs, reorganizations and so on).
I think that these effects are simultaneously immediately apparent to all of us who are actually out in the R&D organizations, but very much underappreciated by the business side of the industry. In fact, I don’t recall ever seeing it officially mentioned. Now, you don’t expect anyone to say “Despite the damage that this will do to our corporate expertise. . .” or anything like that, but there are (as far as I’ve seen) no mentions about how the proposed merger or re-org will preserve that expertise, either. It’s as if it doesn’t exist.
But it sure does. And drug discovery is a field that (as fate would have it) lends itself to this sort of institutional memory being even more valuable than usual. There are several effects leading to that. A big one is that every drug discovery and development project is its own little world with its own specific problems. All those factors that go into a successful drug (potency, selectivity, pharmacokinetics, formulation, and the assays that determine them) are being optimized at the same time, and every project has its own problems and its own ways of balancing them and trading off one for another. This leads to a vast amount of detail, which develops in odd discontinuous ways with detours, sudden bursts of enthusiasm and/or panic, the contest of strong opinions and the need for teamwork, and so on – so you can imagine that assembling a definitive history of just how a particular project went is not an easy job.
Make that “more or less impossible”. At the end of the project, everyone involved will have a slightly different (perhaps very different!) idea of what happened and why. Now, you do sometimes see big wrap-up papers (or a series of them) published in J. Med. Chem. and other journals, and those are valuable. But everyone on the project will tell you that these articles do not (and cannot) tell the whole story. Narrative bias and the needs of scientific publishing can make things look more rational than they were at the time. Side quests get left out of the story or ret-conned to appear in a different guise. Mistakes (and there are always mistakes) are smoothed over – this isn’t deceit so much as it is human nature.
But those same humans remember what doesn’t go into the journal reports, or even into the internal company reports. They remember the quirks of particular assays, of particular chemical series, animal models, cell lines, and all the rest of it. And as people gain experience, they gain the ability to sit up during later projects and say “Hey, you know, when we had the such-and-such project, we ran into something like this, and. . .” That applies to all stages of discovery and development, and you never know where an insight is going to come from. But the key thing to note is that a lot of these things are either never really written down, or are never available in a form that allows people to make the connections. Those connections are made in the heads of the people involved, who saw these things happen and remembered them.
And when they’re laid off, that knowledge goes with them. It’s not like anyone, after being told that the company no longers sees a reason to employ them, immediately starts writing down all the unusual knowledge that they might have. It’s not possible to call that stuff up that way, for one thing, and it’s not like humans work that way when they’ve lost their jobs, either. No, one’s first thoughts are not of random experiences that might turn out to be handy in the future for their former employer. That knowledge departs, never to be seen again.
Now, companies lose experienced employees to retirement all the time, and in that more orderly process some of them do attempt to pick up on that experience beforehand, if possible. But a big merger or re-org clears a lot of people out at the same time, and it tends to hit a disproportionate number of experienced scientists and managers. They’re more expensive and more likely to be made redundant, at least on the org chart. A comment in last week’s post summed it up perfectly: “If many employees leave at once (as in a merger), it’s like the company has a stroke. It can take years to recover from it.” A followup comment noted that in some mergers, it’s like both companies had strokes.
I can’t put it better than that.Even among the employees that remain there’s a fog of uncertainly that slows progress in all of R&D during a merger, but this is another effect on top of that one. And no one putting together M&A deals ever mentions either problem, or even acts as if they exist.