Ask most people outside biotech what a good drug discovery program looks like, and they will describe a straight line:. A target gets validated, a molecule gets designed, it clears in vitro assaysworks in the lab, it works in animal modelss,and marches into the clinic it works in people. Zero detours.
That is not how breakthrough biology works. I think the industry does itself a disservice by pretending otherwise. The programs I trust most are almost never the ones with the cleanest pitch decks. They are the ones that rigorously identified the wrong idea early, killed it on solid evidence, and said so out loud.
Why “No-Go” Gets Treated Like Failure
Somewhere along the way, a data-driven no-go call became something teams apologize for instead of something they report with pride. I understand the instinct. Nobody wants to walk into a meeting and say the target they pursued for six months does not hold up. It feels uncomfortably like admitting defeat.
But a no-go call made on solid data is not a defeat. It is the scientific method working precisely as designed. The alternative, pushing a marginal candidate forward because stopping feels uncomfortable, is far more expensive. It consumes budget, burns runway, and eventually destroys trust once the same liability resurfaces two stages later with a bigger price tag attached.
I have sat in reviews where a lead candidate looked strong in every assay we ran, right up until it hit an in vivo model and lost stability almost entirely. That is a gut punch for a team. Scientists care about their work, and as they should. The correct response is never to massage the narrative or explain away the discrepancy. The correct response is to ask what the biology is actually teaching us, extract the design constraints and recalibrate the research campaign.
What High-Performing Teams Do Differently
When evaluating the health of an R&D program, I never look for an unblemished track record free of dead ends. I am looking for three specific operational habits.
The courage to name their own weak data. If every update sounds uniformly positive, something is being smoothed over. Real biological discovery is noisy. High-performing teams identify the most fragile assumption in their data package without flinching.
A clear distinction between “this needs optimization” and “this needs to stop.” Conflating these two calls burns resources in opposite directions. Continuing to fund a program that has hit a biological wall is obvious waste. Shelving a program that simply lacked the right assay window is an invisible waste—you never discover what you abandoned.
Treating failure as input, not an epitaph. When an early candidate fails in vivo, that outcome is critical data about biodistribution, clearance, or matrix stability. Feeding those empirical boundary conditions back into the design cycle separates an organization that learns from one that simply restarts.
The Part People Overlook
Pipeline discipline is rarely a purely scientific problem. It is almost always a leadership problem.
Scientists are well-equipped to handle iambiguity. What they struggle with is carrying bad news into an executive room that has invested capital, timeline pressure, and corporate pride into an asset. The greatest delay in biopharma R&D does not happen at the bench. It happens in the latency between when a bench team first suspects a fatal flaw and when someone feels safe enough to say it out loud in a portfolio review.
Closing that latency gap is a leadership mandate. It requires cultivating an R&D culture culture where surfacing an unviable result earns more organizational credit than quiet compliance. It requires asking, in every portfolio review, “what specific data would change your mind about this asset?,” and celebrating a definitive answer rather than treating it as a setback.
A Better Standard for Discovery
I measure a program by two distinct questions:
- Did achieve the therapeutic outcome we set out to find?
- id we get an honest, definitive answer from the biology?
A program can fail the first question while succeeding completely at the second. That second success rarely makes industry headlines.But it is the quiet, disciplined execution that protects the rest of the portfolio –and ensures the team has the resources ready when the right biology finally arrives.
If I had to give one piece of advice to anyone building a discovery pipeline, it would be this: build the kind of team that is proud of its data driven no-go calls. Every other standard of scientific excellence follows from that.