The biggest barrier to modernizing biopharma?
There's a harmful posture across biopharma that is hampering meaningful progress: a belief that certain categories of work are inherently beneath attention or "someone else's problem". More on this below. As always, if anything we've written about or built at Kaleidoscope resonates, please reach out!
A recent conversation at a conference crystallized something we’ve been circling for a while.
A senior leader at a large pharma company asked what we do. We gave the simple, high-level explanation we usually start with: we help Life Science teams connect their teams, R&D operations, and associated data, in one spot, and create a system that reflects reality closely enough that people can actually operate off of it.
He smiled, said something along the lines of “data stuff doesn’t speak to my soul,” and walked away.
It would be easy to dismiss that as an idiosyncratic interaction with one person. But it’s the type of line that sticks because it’s not really about us. It reveals a posture that exists more broadly in biopharma, especially at large organizations: a belief that certain categories of work are inherently beneath attention. Necessary, maybe. Important, even. But not worth engaging with directly, because they’re not the “real” science.
There’s a charitable way to interpret this. Science does attract curious people. A clinician we spoke with recently described scientists as “curious children who want to get paid to tinker,” and he meant it as a compliment. There’s something true about it. The best scientists are drawn to the sandbox: the hypotheses, the experiments, the sensation of learning something no one knew yesterday. That kind of work does speak to the soul. It’s why many people enter the field.
The problem is what gets left behind when that becomes the only acceptable source of meaning.
In most industries, the boring parts of the job are still understood as part of the job. Finance teams don’t “feel called” to reconcile books. They internalize it as the cost of being a serious operation. Sales teams don’t wake up inspired to log notes in a CRM, but they do it because everyone understands what happens when they don’t: the organization loses memory, coordination breaks, forecasting becomes fiction, and eventually leadership stops trusting the data entirely. Even engineering teams - arguably the most “craft-driven” function in many companies - accept that testing, instrumentation, and maintenance are not optional chores. They’re what makes the work durable.
Biopharma is unusual in the extent to which foundational operational work is often treated as someone else’s problem. Data is messy, systems don’t talk, context gets lost, and everyone agrees this creates friction. And yet, there’s a cultural permission structure that allows people – sometimes very senior people – to opt out of caring. The attitude isn’t always explicit. Sometimes it’s a shrug. Sometimes it’s “that’s just how it is.” Sometimes it’s a vague confidence that research informatics or IT will handle it. But the consequence is the same: problems that are obvious and widely acknowledged remain unsolved for years.
Once you start looking for the mechanism that keeps them unsolved, it isn’t hard to find. It’s a bystander effect, amplified by bureaucracy.
At large companies, there’s often no single person whose job it is to say: this is broken, and it’s going to stay broken unless we fix it. Business leaders assume informatics will drive solutions. Informatics teams often approach the world as a catalog of tools and features, waiting for the business to articulate a clear need. IT assumes the existing vendor should extend their roadmap. The vendor declines. Someone tries to hack something together internally. It works just enough to create hope, then breaks in all the predictable edge cases. Maintenance becomes no one’s job. IT refuses to support it. The “solution” quietly dies. And because no one owns the outcome, no one is accountable for the failure.
This is where large organizations often end up trapped, in a kind of learned helplessness. Not helplessness in the sense that they lack resources (they have more resources than almost anyone!). Helplessness in the sense that after enough cycles of “we tried and it went nowhere”, people stop believing change is possible without risking their reputation. If the safest path professionally is to avoid rocking the boat, then even leaders who privately agree something is broken will default to inaction. The incentives are aligned against progress.
What makes this especially frustrating is that it can coexist with genuine, local urgency. We’ve spoken with senior directors and leaders who can articulate their pain with precision. They know exactly where the process collapses. They’ve tried the obvious routes. They’ve gone to IT. They’ve gone to their incumbent vendor. They’ve experimented with internal builds, including the new version of internal builds where an AI coding assistant helps you get an MVP working in days. And then the same reality asserts itself: an MVP isn’t a maintained product, and “it works” is not the same as “it keeps working”, let alone “we will trust this for our critical infrastructure”.
This is a broader pattern we’re going to see more of over the next few years. AI-assisted coding will make it easier for non-software teams to prototype solutions. It will not make it easier for those teams to maintain them, support edge cases, handle security requirements, survive org churn, and integrate with systems that weren’t designed to be integrated. The first 80% will get dramatically easier. The last 20% will remain the part that determines whether something is real.
And yet, even when teams reach the end of that cycle – when they’ve tried the vendor, tried internal build, learned why it breaks, and then found an alternative, supported solution – progress can still be blocked for reasons that have nothing to do with the problem.
The encouraging counterpoint is that this posture isn’t universal. We’ve met plenty of people, often those who have spent time in pharma and then moved into smaller, faster contexts, who think differently. They break problems down into constituent pieces, roll out systems deliberately, and are comfortable being the person who says “this matters, and we’re fixing it.” Some of them are impressively systematic about it. We met someone who onboarded multiple tools in under a year and a half because they treated infrastructure as part of the work, not as a distraction from it.
That contrast raises an uncomfortable question: how many capable operators exist inside large pharma who would behave like this, but have learned not to? How many people start out as builders and end up as avoiders because the environment punishes initiative? And how much innovation gets quietly selected out of the system because the people who can’t tolerate that posture simply leave?
If that’s true – and we suspect it is – then the biggest barrier to modernizing biopharma isn’t technology. It’s the cultural permission structure around what counts as real work. As long as foundational data and operational infrastructure are treated as peripheral to “real science,” they’ll remain underinvested. And as long as saying no is frictionless while saying yes requires political capital, the default will continue to be stasis.
None of this is an argument that every senior leader needs to care about the details of data infrastructure to solve it themselves. They don’t. But someone has to care enough to insist that it gets solved, because the organizations that treat this as optional end up paying for it in slower decisions, repeated work, lost context, and avoidable risk. Those costs compound quietly until they become normal. And once they’re normal, they’re very hard to reverse.
Kaleidoscope is a software platform for Life Science teams to robustly manage their R&D operations. With Kaleidoscope, teams can plan, monitor, and de-risk their programs with confidence, ensuring that they hit key milestones on time and on budget. By connecting teams, projects, decisions, and underlying data in one spot, Kaleidoscope enables R&D teams to save months each year in their path to market.