New research finds systemic lack of visibility across pharma and medtech
According to Model N’s 2026 State of Revenue Report, 97% of life sciences companies are using AI for revenue management. Within two years, that number is expected to reach over 99%. Revenue management in this industry is complex work. It covers how companies set prices, calculate rebates owed to payers and partners, manage contracts and stay compliant with government programs. With the high adoption of AI, why is the sector still losing more than $50 billion a year in revenue it should be keeping?
The numbers make even less sense when you look closer at the report statistics. Organizations report doing more with manual processes, and systems are becoming more disconnected. Something isn’t adding up.
How AI is Being Used in Life Sciences
The report surveyed more than 400 pharmaceutical and medtech leaders, all from companies with annual revenue exceeding $250 million. It found an industry that is investing heavily in AI while still operating on slow, fragmented and siloed data. Individual teams and departments are using AI tools independently, often without a shared strategy for how those tools connect to one another or the organization’s central revenue data.
However, confidence in AI across the life sciences is high: 90% of leaders say AI is already delivering results, and 98% expect positive returns within five years. What solutions are they using?
- 64% use AI capabilities available in existing revenue management platforms.
- 62% use GenAI.
- 39% use agentic AI.
Despite high AI adoption, 83% of leaders say they face at least one significant barrier to automating their workflows. The most common barrier, cited by half of respondents, is simply not being able to access the data they need. Revenue teams know that AI tools need accurate, current data to generate reliable outputs. The problem is the infrastructure behind it.
The $50 Billion Problem Hiding in Plain Sight
Fewer than 30% of life sciences companies have fully integrated their gross-to-net data across their pricing and reimbursement programs. Gross-to-net (GTN) refers to the difference between a drug or device’s list price and what a company actually collects after rebates, discounts and fees are paid out. Only 1% of surveyed executives have real-time visibility into what those payouts look like across their government and commercial programs.
Model N estimates that gaps in pricing, rebates and compliance cost U.S. life sciences companies well over $50 billion in lost revenue every year. The leakage takes several forms, including double paying fees to distribution partners, applying discounts to customers who do not qualify for them and paying rebates based on contract terms that are mismatched or out of date.
To understand the scale, Drug Channels Institute estimates total GTN reductions for all brand-name drugs reached $356 billion in 2024. That is the total amount paid out in rebates and discounts across the industry, a figure that grew 7% over the prior year.
Nearly all life sciences leaders surveyed agreed that managing GTN has become more complex due to several market factors.
Managing this complicated issue makes clean, integrated revenue data all the more critical. The report data shows the use of purpose-built commercial software grew by 46%, which means many companies are now running several different tools in parallel with little or no connection between them. Adding more tools to an already fragmented environment compounds the visibility problem.
The use of manual tools grew 43% as GTN complexity increased. Manual processes are prone to inconsistencies that require extra time to reconcile and raise the risk of overpayments and compliance errors.
These statistics, combined with the extremely low real-time visibility, suggest departments and even individuals are applying AI to data in isolation. They may be extracting pieces of data into spreadsheets or separate tools and analyzing them outside centralized systems. The issue is not data flow or processing speed. It is that revenue management data is being pulled into disconnected workflows shaped by individual objectives rather than managed through a structured, centralized source of truth. The results at the individual level can be useful, but the cumulative effect is worsening the problem of siloed data.
Where Leaders See Opportunity
Looking ahead, 90% of leaders expect to be using AI embedded into their revenue management platforms by 2028.
When asked which areas of revenue management could benefit the most from AI, 65% named deal analytics and insights. That is a 63% increase over the previous year, making it the single fastest-growing AI priority in the survey. Process automation came in second at 56%, also up sharply from 2025, and revenue forecasting and compliance monitoring each came in at 48%.
These findings reflect where leaders believe AI can produce solid returns, as long as the data supporting those tools is reliable.
The operational gaps behind those priorities are equally clear. When asked to name the areas within their revenue operations where they see the greatest room for improvement, data analytics was at the top with 64%, up 10 percentage points from last year. That increase signals that leaders see data as the lever with the most upside. Revenue forecasting followed at 42%, then pricing and quoting at 41%.
This is consistent with last year’s Model N research, which found that 87% of leaders said their primary innovation focus was on automating revenue management operations. The ambition to modernize has been building for years, but the data infrastructure isn’t there to fully support it.
The Remedy
Life sciences companies expect AI to deliver big results, but before they can achieve those goals, they need a deliberate strategy for adoption. This includes:
- Defining clear use cases across revenue management workflows.
- Establishing which tools departments and individual users should use.
- Building integration points between those systems and connecting pricing, rebate, contract and compliance data into a single shared view.
- Appointing a top-level champion to drive execution and measure outcomes.
Implementing AI for the sake of AI is not a long-term solution. Instead of leaning into individual AI applications, companies should invest in agentic AI that can move across all point solutions to bring together data without creating more data silos.
About the Author
Michael Grosberg is the Vice President of Product Management at Model N, responsible for life sciences products across Model N’s portfolio. Michael joined Model N as part of the acquisition of Deloitte’s Pricing and Contracting Solutions business in 2021 and continues to lead a team focused on regulatory compliance, revenue management and analytics for pharmaceutical manufacturers. A data scientist and a policy wonk, Michael views the highly complex spaces of pharmaceutical revenue management, market access, and government pricing through the underlying data, pursuing accessibility and quality that drive commercial insights. His 15-year career spans roles in public policy, analytics, systems implementation and change management.
