Does Clinical Development Really Need An S&OP?

By Tom Walls, principal and founder, Axon Bridge Consulting

The series closes on the question in the original pitch, and makes the case that the answer to "do we need the meeting?" is yes, even if the answer to "do we need the name?" is not. It also proposes a better name.

Three articles in, the case has been building. Demand in a clinical-stage company comes from four functions that do not report to supply chain. Supply is mostly outsourced, yield-dependent, and constrained by shelf life. The forecast changes monthly and the consequences of getting it wrong run from a halted randomization to a seven-figure batch in the incinerator. The question this series set out to answer was whether the sales and operations planning process, built for companies that sell things, bends far enough to fit that world. Here is the answer, in three parts: yes, the meeting is necessary; no, the commercial template does not fit unmodified; and the part that matters most, the process has to be owned centrally, or it will not work at all.

The Case For The Forum

Start with what happens without one. In an emerging biotech with two or three programs, the demand for a given drug substance is owned by clinical operations for the trials, by analytical development for stability and methods, by regulatory for submission and retains, and by development for the next study in the plan. Supply is owned by CMC, usually program by program, each with its own CDMO relationship. Without a standing cross-functional meeting, every one of those functions plans in its own spreadsheet, on its own horizon, in its own unit.

The results are predictable. The CDMO receives a forecast that nobody in the company would stand behind if asked, because it was assembled from three documents of different vintages. Material expires in one program while another program is short of the same drug product, because nobody was looking at both lots on one page. A supply decision gets made in a hallway, by whoever escalated loudest that week, and the function that was not in the hallway finds out when the batch is already allocated. Finance builds a budget on batch counts that planning has already revised. Leadership, asked whether the program can be supplied through the next readout, gets an answer that depends on who they ask.

A monthly forum that puts one set of numbers in front of all the demand owners and forces a decision is not commercial overhead imported prematurely. It is the only mechanism that gives a leadership team a real view of whether their programs can be supplied, and it is the only place where the trade-off between one program's risk and another's can be made explicitly rather than by accident. That is the affirmative case, and it does not depend on what the meeting is called.

What The Commercial Template Gets Right

Give the traditional S&OP process its due. It got four things right that transfer directly.

The monthly rhythm. A fixed cadence of demand review, supply review, reconciliation, and an executive decision meeting, each in its own week, is exactly what a clinical program needs, and the planner should feel free to cancel a meeting with nothing on it or accelerate the cycle when a batch fails. The cadence is the discipline; the calendar is a tool.

Cross-functional ownership. S&OP was invented because sales and operations were each planning in isolation and the result was chaos. Replace sales with clinical operations, regulatory, and analytical development and the diagnosis is identical.

One set of numbers. The single most valuable output of a mature S&OP is that finance, operations, and the commercial organization stop arguing about whose forecast is right and start arguing about what to do. A clinical-stage company that gets its four demand streams into one unit, on one horizon, in one plan, has achieved the same thing.

An executive decision forum. The strategic tier of S&OP exists to put the trade-offs that operational people cannot resolve in front of the people who can. In a clinical program, the decision to commit a drug substance campaign inside a cancellation window, against demand that may not materialize, is exactly that kind of trade-off. It should be made by executives, on a schedule, with the risk on both sides quantified.

Where It Strains

Four things do not transfer, and each of them is structural rather than cosmetic.

There is no sales number to reconcile. The commercial demand review is a negotiation between a statistical forecast and a sales organization with a quota; the tension between the two is what produces a consensus number. In a clinical company there is no quota, no history, and no function whose job is to want more demand. The demand review is instead an interrogation of assumptions, and it needs a planner who knows which questions to ask rather than a facilitator who can broker a compromise.

The demand owners do not report to operations. In most commercial companies S&OP lives inside the supply chain organization and the demand side attends because the chief commercial officer has signed up. In a clinical company, clinical operations reports to development, regulatory reports to its own head, and neither has any reason to show up to a supply meeting unless it is visibly making their programs safer. Attendance has to be earned with outputs those functions value, which is why the change-over-change report from the previous article matters so much.

The horizon outlasts the trial. Commercial S&OP typically looks 18 to 24 months out and rolls. Clinical planning needs 36 months minimum, which means forecasting studies that have no protocol yet, from a clinical development plan that development may not want to share widely. The forum has to be trusted with that.

The supply side is mostly CDMOs. The supply review in a commercial S&OP is a conversation with your own plants. Here it is a conversation with partners that have other clients, their own capacity constraints, and contractual commitment windows that do not care about your enrollment curve. That supply review has to be fed by a CDMO governance process that runs on its own cadence and feeds the forum, rather than a plant manager reporting on a line.

Who Owns It: The Central Planning Argument

This is the part of the argument I want to press hardest, because it is where good intentions most often fail in practice.

The natural instinct in an emerging biotech is to let each program plan itself. The CMC lead for Program A knows the process, the CDMO, and the study. The CMC lead for Program B knows theirs. Each runs a planning conversation with their own program team, in their own spreadsheet, and reports upward. It feels efficient, it respects expertise, and it is exactly how the failures in the second section of this article come about.

The reason is contention. The moment a company has more than one therapy, or more than one indication for the same molecule, the programs compete for the same things: the same drug substance lots, the same fill slot at the CDMO, the same packaging campaign, the same analytical capacity, the same cash. A planning process owned by the program streams has no place where that contention becomes visible, let alone resolved. Program A will plan as if it has the lot. Program B will plan as if it has the lot. The CDMO will receive two forecasts that together exceed the campaign, and the conflict will be discovered at the point of allocation, which is the worst possible moment.

A central planning function fixes this by construction rather than by goodwill. One function owns the demand picture across every program, the supply picture across every CDMO, and the balance between them. It is the only function whose job is to see the contention, and it is the only one that can present a trade-off between programs without being a party to it. When Program A and Program B both need the same drug product in the same quarter, the question goes to the forum with both options quantified, and leadership decides. That is not possible when the two programs are represented by two advocates each holding their own numbers.

Central ownership also carries the two things that make the numbers comparable in the first place: standardized data and standardized process. Every program forecasts in the same unit, on the same horizon, in the same buckets, with the same definitions of safety stock, weeks of supply, and yield. Every program refreshes on the same cadence, using the same actualization, requirements review and reconciliation steps, and reports in the same change-over-change format. Standardization is not bureaucracy. It is what allows a leadership team to look at three programs on one page and believe the comparison. The moment one program measures inventory in vials and another in weeks of supply, the page is decorative.

None of this requires a large team or an ERP. Two disciplined planners with a well-run spreadsheet and an agreed process will outperform six program-embedded analysts with a planning system each. Process discipline and master data standardization come first; the system, when it comes, should be specified by the planning function that already runs the process, not imposed on it.

Nor does central ownership mean removing the program teams from the conversation. It means the opposite: the program teams own their demand assumptions and show up to defend them, the CMC leads own their supply assumptions and their CDMO relationships, and the central function owns the plan that nets them together and the meeting where the trade-offs are made. Ownership of the process is what makes participation by everyone else worthwhile.

The Naming Question

Which leaves the name. "Sales and operations planning" is precise about a thing a clinical-stage company does not have. That is not merely awkward; it sends a signal. A clinical operations lead invited to a meeting called S&OP may reasonably conclude it is a supply chain meeting that does not concern her, and a regulatory lead will assume it is about the commercial product that does not yet exist. The name, in other words, can quietly exclude the very people the forum exists to include.

There are three reasonable options. Keep the name S&OP and modify the content, on the grounds that the executives and board members who will sponsor it know what S&OP is and why it works. Adopt integrated business planning, which has the advantage of being broader and the disadvantage of being a consulting label that means different things in different companies. Or use a development-specific name that says what the meeting is actually about.

My proposal is the third, and a specific version of it: demand and operations planning, or D&OP. The change is one letter, and it is the right letter. What a clinical-stage company lacks is not the operations half of the process. Manufacturing, CDMOs, packaging, depots, and yields are all present and all need planning. What it lacks is sales, and what stands in for sales is demand from four owners who do not think of themselves as demand owners at all. Putting demand in the name tells clinical operations, analytical development, and regulatory that the meeting is about their assumptions and their programs, which is exactly the audience that is hardest to keep in the room. It is close enough to S&OP that an executive or board member recognizes the process on sight and knows why it works. And it does not expire: when commercial demand finally arrives, it becomes one more stream in a forum already built to carry several, and nothing needs renaming.

The name should be chosen for the audience that is hardest to keep in the room, which is the demand owners rather than the executives. Executives will attend a decision meeting whatever it is called. Clinical operations and regulatory will attend a meeting that is visibly about their programs, and a name that says so helps. That is the case for D&OP. I hold it more loosely than the rest of the argument, because the name is the least important decision on this list, but it is the one a company can make today.

Working Conclusion

The process is about communication and decisions, not the spreadsheet. A planner in a clinical-stage company is a communicator first: the best plan in the world fails if clinical operations, finance, and the CDMO do not believe it, and the cleverest model is worthless if the people who own the assumptions never see the output. The right answer to "does clinical development need an S&OP?" is therefore the pragmatic one. It needs a monthly forum where every owner of demand and every owner of supply looks at one set of standardized numbers, sees what changed, and makes the decisions that constrained material forces. It needs that forum to be owned by a central planning function that sees across programs, so that contention is resolved on purpose rather than by accident. And it needs a name that gets the demand owners into the room and keeps them there.

Call it S&OP if that is what your board understands. Call it D&OP if you want the name to tell the demand owners the meeting is theirs. Just do not let each program run its own, because a company with three therapies and three planning processes does not have a plan. It has three arguments waiting to happen.

Editor’s Note: You can read Part 1 here, Part 2 here, and Part 3 here.