The cliff was never the crisis: Pharma’s strategy problem
Let me start with something most strategy decks dance around: the patent cliff isn't the problem. It's the symptom.
Every commercial leader knows the headlines. Evaluate estimates that more than $300 billion in branded drug sales will lose patent protection between 2025 and 2030, roughly a sixth of the industry's revenue base. The first round of Medicare-negotiated prices took effect on 1st January, with cuts ranging from 38% to 79% below list prices. Most-Favoured-Nation pricing, tariffs, even the idea of government-led drug purchasing moved from policy debate to operational reality in what felt like the blink of an eye. By 2029, Medicare negotiation will expand to twenty drugs a year.
The natural reaction is to treat 2026 as a revenue problem. Defend price. Cut costs. Protect margins. Wait for conditions to improve.
That instinct may end up costing more than the patent cliff itself.
Because this was never really a pricing issue. It's a strategy issue.
What's actually changing
For the better part of two decades, pharma operated under a comfortable assumption: develop a strong asset, get it approved, and then optimise the commercial strategy over time.
The IRA didn't just lower prices. It changed the economics that shape investment decisions long before a product reaches the market.
A small-molecule drug becomes eligible for Medicare negotiation roughly nine years after launch. A biologic gets around thirteen years. Four years may not sound like much, but it becomes highly significant when portfolio teams are deciding where to invest. Suddenly, oral therapies look different. The revenue horizon changes. Expected returns change. Capital starts moving accordingly.
And there's another shift that receives far less attention.
The clock now starts at launch.
Questions that companies once answered years into a product's lifecycle now have to be addressed much earlier. Which indication comes first? How should expansion opportunities be sequenced? What price position can survive future negotiation pressure? What evidence will matter most to payers?
Those aren't lifecycle decisions anymore. They're launch decisions.
And in many organisations, the people responsible for answering those questions still aren't aligned early enough.
That's not simply a margin challenge. It's a fundamental shift in who makes critical decisions, when they make them, and how early they need to commit.
Many operating models were built for a different reality.
Speed isn't an advantage anymore - it's oxygen
For years, speed-to-market was considered a competitive edge. Beat the next entrant. Beat the generic. Today, it's becoming something more basic: survival.
As commercial windows compress, the period between approval and peak adoption matters more than ever. What used to be a long runway is becoming a much shorter sprint.
Lose six months to a slow launch, a delayed market access strategy, or payer conversations that should have started years earlier, and you're not trimming the edges of the business case. You're giving away value you'll never recover.
But launch speed is only part of the story. The more important capability is how quickly an organisation can change course once new information arrives. A competitor releases data. A formulary decision shifts access. A safety signal emerges. Market conditions change.
The companies pulling ahead aren't necessarily the ones with the best forecasts. They're the ones that can update those forecasts quickly and act on them immediately.
They can reallocate resources, adjust messaging, and revisit assumptions within days or weeks.
Most organisations still can't.
Too often, the commercial plan is frozen in a deck approved months earlier. By the time planning cycles catch up, the market has already moved. Speed isn't something you switch on at launch. It's built into how quickly an organisation can make decisions, learn, and make better decisions again.
The AI conversation has moved on, but much of pharma hasn't
This is usually where someone says, "We'll use AI."
Fair enough.
But AI isn't the hard part anymore. A 2025 MIT Technology Review Insights and Globant survey found that roughly three-quarters of pharma leaders were already planning, piloting, or deploying agentic AI initiatives. The challenge wasn't the technology itself. The challenge was workflow design, governance, and compliance. In other words, the intelligence isn't the bottleneck. Knowing what to do with it is.
Even the conversation around AI has changed dramatically in the past year. Twelve months ago, most discussions focused on copilots, systems that assist human users. Today, attention is shifting towards orchestration: multiple AI agents working together across workflows.
One agent gathers evidence. Another drafts a response. A third validates it against regulatory and MLR requirements before routing it for approval. That's a meaningful leap forward. It's also where governance becomes far more complicated.
What happens when an agent drafts a payer narrative? Who is accountable if an AI system incorrectly prioritises a safety signal? When multiple agents contribute to a decision, where does responsibility ultimately sit?
Pharma built its control structures around people: review committees, SOPs, approval chains, and documented accountability. Those frameworks don't automatically translate to autonomous software.
The next competitive advantage won't come from a slightly better model. It'll come from the less glamorous foundations that allow organisations to trust these systems at scale: audit trails, observability, governance frameworks, clear guardrails, and defined accountability.
When AI initiatives disappoint, the problem is rarely the model. It's fragmented data. Unclear ownership. Decision-making processes that still require six committees and three planning cycles. You can't layer autonomous systems onto an organisation designed to move slowly and expect transformation.
What leading companies are rebuilding
The organisations gaining ground aren't necessarily the ones with the flashiest AI pilots or the longest list of innovation programmes.
They're treating 2026 as an opportunity to redesign how they operate.
Three moves stand out.
1.Make commercial decisions before the science is finished
In many companies, commercial strategy, market access, and evidence planning arrive late in development, often after key R&D decisions have already been made.
That sequence no longer works.
Commercial leaders, access teams, and evidence experts need to be involved while options still exist, when indication prioritisation, modality choices, pricing assumptions, and launch sequencing can still be influenced.
Waiting until approval to finalise commercial strategy used to be common practice. Today, it's simply expensive.
2.Build one source of truth, not another layer of dashboards
Every company claims to be data driven. Yet, many struggle to answer the same business question consistently across teams. An AI-ready data foundation isn't an IT initiative sitting on the side of the business. It's core infrastructure.
Treat it the way you would any critical operational asset. Fund it properly. Assign ownership. Maintain it continuously. Hold leaders accountable for its performance.
Without that foundation, every AI investment becomes harder to scale and harder to trust.
3.Put an owner and a deadline on every major decision
What slows organisations down isn't usually technology. It's ambiguity.
Nobody knows who owns the decision. Nobody knows what evidence is required. Nobody knows when the decision must be made. The result is endless escalation and committee-driven paralysis. The organisations moving fastest tend to be remarkably simple in this regard. Every major decision has a clear owner, a defined timeline, and explicit accountability.
AI systems should be governed the same way.
Every autonomous workflow needs an owner. Every decision needs an audit trail. Every system needs clear boundaries.
An autonomous system without accountability isn't a capability. It's a compliance problem waiting to happen. None of this is fundamentally a technology challenge. It's a strategy and operating model challenge that happens to be accelerated by technology.
Get the sequence wrong, putting tools ahead of the operating model, and you end up with impressive demonstrations and disappointing business outcomes.
The real divide
At its core, the constraint facing pharma isn't scientific. The pipeline remains strong. The real question is whether commercial organisations can move quickly enough to capture value before the window narrows.
The patent cliff will divide the industry into two groups.
One group will treat it like bad weather. They'll cut costs, defend margins, and wait for conditions to improve. The other group will recognise what's really happening. The old commercial operating model is reaching the end of its useful life. Those companies won't just survive the next decade; they'll define it.
The cliff was never the crisis. It was the alarm.
The only question left is, who's listening?
References
- Patent-cliff exposure (more than $300bn in branded sales losing exclusivity, 2025–2030; roughly one-sixth of industry revenue): Evaluate. Analyst estimates range from about $236bn (PharmaVoice) to $400bn (DrugPatentWatch) depending on methodology.
- Medicare negotiated-price discounts of 38–79% off list for the first ten drugs (effective 1 January 2026), the expansion to up to 20 drugs a year from 2029, and the earlier negotiation-eligibility timing for small molecules versus biologics: Centers for Medicare & Medicaid Services (CMS); summarised in KFF, "Key Facts About Medicare Drug Price Negotiation."
- Agentic-AI adoption among senior pharma executives (~three-quarters planning, piloting or deploying; workflow design and compliance cited as top barriers): MIT Technology Review Insights and Globant, "Transforming Commercial Pharma with Agentic AI" (2025).
About the author
Amanjeet Singh Saluja is a seasoned leader in AI, analytics, and cloud software. He currently heads a Strategic Business Unit at Axtria Inc., a global provider of AI and cloud solutions to the life sciences industry. Singh Saluja has built, scaled, and exited three successful ventures, and is the original inventor of a US patent for collection cycle optimisation through advanced analytics. He brings 26 years of experience advising Fortune 500 clients in financial services, life sciences, and MedTech on risk management, commercial strategies, and artificial intelligence. He has been recognised in Marquis Who’s Who in 2025. Singh Saluja began his career in process re-engineering and strategy roles at KPMG and Andersen. He holds a degree from the Indian Institute of Technology Kharagpur and an MBA from the Indian Institute of Management, Ahmedabad. He is passionate about leveraging AI and analytics to drive business success.
