Building Something Worthy

This is the first in a series. The second, "A Role That Doesn't Fit Neatly Into Any Box," examines why clinical-stage biotech has no financial operating system. The third, "What the Financial Layer Should Have Been All Along," lays out the design principles behind TrialCast.

I used to joke that I pray at the altar of Excel, because I love Excel that much and have found it to be an incredible force multiplier throughout my time in school and my career. So it shouldn't surprise you to learn that I build financial models for clinical-stage biotech companies. The sophisticated kind: rolling forecasts that stretch five to twenty years, with study-level cost assumptions driven by enrollment curves, CRO milestone payments, site budgets across multiple countries, and pass-through costs that shift every time a protocol amendment lands.

These models work. They serve my clients well. And every one of them lives in Excel.

I say this not as a confession but as context. The people doing FP&A work at clinical-stage biotechs are building something genuinely complex, and they're doing it with remarkable skill. The question I keep coming back to isn't why we're still using spreadsheets; it's what it would take to build something worthy of replacing them.

That's a higher bar than most people realize.

---

The Work Deserves Recognition

Before talking about where the industry is headed, it's worth pausing to appreciate what biotech FP&A professionals actually do, because this work is rarely visible outside the finance function and almost never discussed at the level it deserves.

A clinical-stage biotech might have six studies in various stages of planning and execution, each with its own CRO contract, its own enrollment assumptions, its own site budget structure, and its own regulatory timeline. The FP&A function is responsible for translating all of that clinical complexity into a coherent financial picture: how much cash do we have, how fast are we burning it, and how long does it last under different scenarios?

That translation requires a rare combination of skills. You need to understand clinical operations well enough to know that a Phase 3 trial burns $2 million a month during enrollment and $200,000 during follow-up, and that the transition between those two states depends on site activation rates, screening failure rates, and enrollment velocity across multiple geographies. You need to understand accounting well enough to reconcile accruals against invoices when vendor names are buried in journal entry memo fields. And you need to understand corporate finance well enough to model multi-tranche capital raises against a cash runway that depends on clinical milestones that haven't happened yet.

The people who can do all of this, who can sit in a room with the Chief Medical Officer and the CFO and translate between them in real time, are extraordinarily valuable. And extraordinarily rare.

---

Why Excel Has Persisted

Excel persists in biotech finance for a reason that deserves respect rather than criticism: it is the only tool flexible enough to model the full complexity of clinical-stage operations.

Traditional FP&A platforms (Anaplan, Workday Adaptive Planning, Planful) were designed for companies with revenue, seasonal patterns, and cost structures that follow predictable cycles. They're excellent at what they do. But they were built for a world where last year's actuals are a meaningful starting point for next year's plan. In clinical-stage biotech, last year's actuals might tell you almost nothing about next year, because next year you might launch a Phase 3 trial that didn't exist in anyone's plan eighteen months ago.

The clinical operations side has its own systems: Medidata, Veeva, and newer entrants building trial budgeting capabilities. These tools do important work within the clinical domain. But they live on one side of a divide. They can model what a trial should cost based on protocol design. They generally can't tell you how that cost interacts with your CMC manufacturing timeline, your headcount plan, your G&A overhead, your pre-launch investment, and your cash runway, all viewed together, all responsive to the same set of milestone assumptions.

That integrated view, the one that connects clinical timelines to vendor contracts to cash position to board-ready scenarios, is what FP&A builds. And they build it in Excel because nothing else lets them hold all of those relationships in one place, shaped exactly the way each company needs them shaped.

That flexibility is Excel's greatest strength. It's also, increasingly, its constraint.

---

66 Tasks. Every Month. Every Company.

Everyone in biotech understands what clinical development costs. The $155 million pivotal trial. The $25 million supplemental studies. The CRO contracts, the site budgets, the CMC manufacturing campaigns. These are the numbers on the slide when the CEO presents the development plan to the board.

What almost nobody talks about is what it costs to manage those numbers.

When you map the full FP&A operating cycle at a clinical-stage biotech (monthly close, accrual reconciliation, invoice review, forecast updates, variance analysis, board preparation, investor reporting, scenario modeling, vendor management, capital planning, budget development, audit support, and the constant flow of ad-hoc requests) you arrive at 66 distinct recurring tasks.

The monthly cycle alone starts with closing the prior month's books within the first five to ten business days. For a company with 30 to 50 active vendors, this involves reviewing dozens of invoices, validating coding against budgets, posting accruals for work performed but not yet billed, and following up on missing accruals from CROs and clinical vendors who don't always deliver on time. The accrual process is where clinical operations and finance intersect most directly; estimating them accurately requires understanding where each study is in its lifecycle, how many patients are enrolled, how many sites are active, and what milestones have been achieved. This isn't a simple calculation. It requires clinical context that most accounting professionals don't have, and accounting precision that most clinical professionals don't practice.

Then comes variance analysis: understanding why actual spend deviated from plan, and the answer is almost always clinical. Enrollment ran ahead at three sites, increasing per-patient costs. A protocol amendment triggered an unbudgeted change order. A manufacturing run was accelerated, pulling CMC costs forward by a quarter. Each variance tells a story, and the FP&A professional's job is to understand it well enough to explain it to the CFO, who will explain it to the board.

Everything intensifies on a quarterly cycle. Board materials, investor reporting, and forecast updates layer on top of the standard monthly workload. The board package alone is an integrated narrative requiring inputs from clinical operations, manufacturing, regulatory, and finance, assembled by FP&A. Investor updates have hard deadlines, typically 45 days after quarter-end, with consequences ranging from uncomfortable conversations to covenant violations.

And on top of the monthly and quarterly rhythms, the annual budget cycle adds six to eight weeks of intensive work that overlaps with everything else. The annual budget for a clinical-stage company requires rebuilding the long-range forecast, incorporating updated clinical timelines, reforecasting vendor contracts, and slicing a continuous, milestone-driven program into 12 monthly boxes while preserving the underlying clinical logic.

Of those 66 tasks, fewer than a dozen require the kind of human judgment that makes a senior FP&A professional irreplaceable: interpreting variances, advising on clinical-financial trade-offs, presenting to the board, counseling the CEO on capital strategy. The rest are mechanical but essential: pulling data from multiple systems, reformatting reports, reconciling invoices against contracts, updating spreadsheets that track information the accounting system can't.

---

What's Starting to Change

Several things are converging that make this moment different from previous cycles of "technology will transform finance."

API integrations with platforms like QuickBooks and BILL.com now make it possible to build live data connections without asking companies to change their existing accounting or payment systems. You don't have to rip out the foundation to improve the structure above it.

Machine learning has reached a point where it can handle the kind of messy, unstructured data that has historically required human pattern recognition, like resolving vendor names from free-text memo fields in journal entries, or matching invoices against contract terms across inconsistent coding schemes.

And the economic pressure on clinical-stage biotechs, particularly in the current fundraising environment, has elevated operational efficiency from a back-office concern to a board-level priority. When capital is harder to raise and the timeline to revenue is measured in years, the ability to model scenarios quickly, track vendor spend precisely, and present a clean financial picture to investors is a strategic capability.

None of this means Excel disappears. It won't, and it shouldn't, not entirely. What it means is that the mechanical scaffolding I described earlier (the 50-plus repetitive tasks that consume so much of a skilled FP&A professional's time) is increasingly automatable. And as it becomes automated, the people doing this work get to spend more of their time where they add the most value: in the room with the CEO, interpreting the numbers, advising on strategy, and helping the company navigate the most capital-intensive period of its life.

---

Building Something Worthy

I think about this constantly, what it would mean to build financial tools that actually match the sophistication of the work being done in biotech FP&A. Not tools that simplify the work, because the work isn't simple. Tools that honor its complexity while removing the parts that don't require human judgment.

I currently serve as an embedded fractional FP&A leader for my clients, doing the work of a full-time FP&A Director on a part-time basis. If I'm honest about how I spend my time, a meaningful portion goes to the mechanical scaffolding (the data gathering, the reconciliation, the reformatting) rather than the strategic counsel that actually moves the needle for my clients. If the mechanical work were handled by purpose-built systems, I could focus on the half-dozen tasks where my judgment genuinely matters. I could serve five times the clients at the same depth of insight.

That's not a hypothetical. That's what my co-founder and I are building.

The standard for any new tool in this space should be high. It has to understand clinical trial economics at the study level. It has to connect the financial model to the clinical timeline so that when one changes, the other responds. It has to serve the FP&A professional, the program manager, the CEO, and the board, each of whom needs a different view of the same underlying reality. And it has to be built by people who have actually done the work, because the nuance of this domain doesn't reveal itself to outsiders.

The people doing FP&A at clinical-stage biotechs have been building extraordinary things with the tools available to them. They've earned something better. And the conditions are finally right to build it.

If you're a biotech finance leader who recognizes this world, I'd love to hear from you. We're building the FP&A operating system for clinical-stage biotech at CaladanAI.