The tough news for Big Pharma in 2026 is that its incredible scientific achievements have run into the financial realities of employer-provided healthcare. Millions of patients are being put on GLP-1 drugs. Cancer medications are becoming more effective — and expensive. Specialty drugs, cell and gene therapies keep getting launched. At the same time, employers are staring at healthcare costs that are increasingly out of control. And they’re starting to push back. Last week Disney announced it will no longer cover working spouses who can get insurance through their own employers. The reason? Healthcare costs. Other large employers are grappling with the same dynamics. Here’s your reward, pharma. You invented some great drugs. Now American employers are deciding what benefits they can no longer afford.

Start With a Number: $26,993

That’s the yearly family premium cost for employer-sponsored health insurance in 2025. $26,993. Employees, on average, paid $6,850 of that premium cost. Average premiums for single coverage hit $9,325. That doesn’t take into account all the money you spend when you use your insurance, either.

The average deductible for workers with a general annual deductible was $1,886. Average deductibles climbed to $2,631 for workers at small businesses. One in three covered workers, 34%, had insurance with a deductible of at least $2,000 for single coverage. One in five covered workers with an out-of-pocket maximum had to pay over $6,000 before their insurance kicked in with any real gusto. Welcome to health care in America. First you fork over thousands on insurance. Then you happily fork over thousands more before your insurance does much of anything. Let’s pretend this is “affordable.”

It Gets Better in 2026

Aon estimates that the average employer health-plan cost increased from $16,212 per employee in 2025 to $17,562 in 2026. Employer cost increased 8.8%. Employee payroll premiums increased 6.4%. Employee out-of-pocket spending increased 10.2%. Put premiums and out-of-pocket costs together and the average employee burden reached approximately $5,297 in 2026, up 7.9% in a single year.

Employees aren’t imagining healthcare inflation. They’re experiencing it in their paychecks and at the doctor’s office. And the trajectory isn’t improving.

Business Group on Health’s newest survey projects a median 9.2% increase in employer healthcare costs for 2027 before plan changes, following an estimated 8.5% trend in 2026. It estimates that healthcare costs could rise a cumulative 76% over ten years, including the 2026 and 2027 forecasts.

Eventually, CFOs notice numbers like that. And when CFOs start managing employee healthcare, “comprehensive benefits” quickly becomes “cost optimization.”

Enter the GLP-1 Gold Rush

GLP-1 medications are amazing drugs. They are also something of the ideal pharmaceutical business model. Pick a huge chronic disease. Create a highly effective medication. Broaden the eligible population. Broaden the indications. Push chronic utilization. And let somebody else pay the bill.

For employer plans, that “somebody else” is you. In a recent Business Group on Health survey, nearly 8 in 10 employers reported that GLP-1 drugs were driving up their healthcare costs. Two-thirds covered the medications for weight loss at the time of the survey.

But only 72% of employers who were covering the medications said they were likely to continue doing so in 2027. Employers are already backing away from the drugs. Business Group on Health’s newer 2027 survey showed GLP-1 obesity coverage declining from 72% of employers in 2025 to 60% in 2026.

Apparently, “patient access” sounds wonderful until the invoice reaches the benefits department.

Here’s the Part Pharma Doesn’t Put in the Investor Deck

IQVIA projects U.S. net medicine spending will increase 10.6% in 2025, reaching $606 billion from $548 billion. That’s $58 billion more than we spent last year. GLP-1 drugs added roughly $14 billion of that growth figure ($9.6 billion from therapies approved for weight loss and associated conditions).

Commercial insurer patient pharmacy deductibles and co-pays grew 5% in 2025 and were 37% higher than they were five years ago. There’s your innovation paradox. Pharma churns out better drugs. Spending soars. Payers restrict access. Patients pay higher out-of-pocket costs. And we pat ourselves on the back for ‘making healthcare better’.

Pharmacy Is Eating the Healthcare Budget

Prescription drugs aren’t just another cost center for employers. They are rapidly becoming one of the largest expenses for employers that offer health insurance to their employees. Pharmacy made up 25% of total employer healthcare expenditures, according to Business Group on Health. That’s right, one out of every four healthcare dollars went toward drugs, and it was expected to rise another 12% in 2026.

Employers aren’t simply purchasing doctor visits and hospital stays. They are subsidizing one of the most profitable pharmaceutical industries in the world. And employers can’t print money like the federal government. Something’s gotta give.

Then There’s Cancer

If GLP-1s are the high-volume budget buster, oncology drugs are the high-cost clinical conundrum. Cancer is eating employers’ healthcare budgets. Not just sometimes.

For the past five years running. In Business Group on Health’s latest annual survey, 70% of employers said cancer was their top healthcare cost driver in 2026, compared with 58% in 2025. A whopping 92% of employers said cancer was one of their top three cost drivers.

IQVIA recently reported that “‘protected’ branded drugs,” which usually refers to oncology drugs and immunology products, were the main drivers of the substantial increase in U.S. drug spending in 2025. And now we get to the ethically awkward part. What do you tell an employee facing metastatic cancer? That their newest treatment option is too expensive? How about the employee’s spouse? Kids? Please. Like most expensive medical scenarios, employers just pick up the tab. And since health plans don’t have unlimited funds, that money has to come from somewhere. Enter the rest of your workforce.

Congratulations. You’re Helping Pay for Someone Else’s $200,000 Drug.

This is the part nobody wants to talk about. Most employees will never use a $200,000 oncology medication. Most will never qualify for gene therapy. Many will never be on a GLP-1. Doesn’t matter. You’re still footing the bill. Directly through premiums. Out of pocket through deductibles. Out of pocket through coinsurance. Out of pocket through higher out-of-pocket maximums. Out of pocket through tighter formularies. And time wasted through prior authorization. You may even pay out of pocket when your spouse is kicked off of your company plan because they “technically have coverage” elsewhere. And yes, you may pay through compensation. Employer-provided healthcare is compensation. Every dollar above the previous year that an employer spends on healthcare has to compete with wages, bonuses, new hiring, retirement matching, and every other benefit. The pharmaceutical industry doesn’t send you that bill. Your employer does.

Pharma Loves Talking About “Value”

Ok this is my favorite one. “This drug adds value to the healthcare system.” Ok it may. A GLP-1 may help prevent diabetes, CV events, sleep apnea, and other expensive downstream complications. A cancer drug may add years to life. That is real value. But there’s a problem. Money. Value does not equal affordable. Sure, a Ferrari is worth $300k. Doesn’t mean I can afford it. Pharma loves pricing drugs based on theoretical value. Employers work on annual budgets. Try explaining this to a CFO. “Hey, it’s ok if you spend $15 million more on GLP-1s this year. You’ll save on CV events in 2041.” I can guarantee you that will go over very well. Half your employees will be working elsewhere by then.

The Most Absurd Part: We Built an Entire Industry Between Pharma and the Employer

Oh wait, it’s not just drug companies that are screwing us. We made the PBMs too. Because heaven forbid we should just be able to buy drugs without an advanced degree in Price Canyon. Now employers have to deal with: List prices. Rebates. Administrative fees. Formulary payments. Specialty pharmacy margins. Spread pricing. Rebate guarantees. Prior authorization. Step therapy. Manufacturer “patient” assistance programs. Contracts so Byzantine they require consultants to interpret to the folks who negotiated said contracts. The drug company tells you the PBM is gouging you. The PBM tells you the manufacturer is price gouging. The Insurer cries high utilization. The Hospital cries low reimbursement. Blame seems to abound for everyone. Miraculously, everyone also appears to be getting paid. Except for the one person who actually needs the drug. The employee is forced to pay a $2,000 deductible.

The Great American Healthcare Shell Game

Here’s how it works. Pharma ramps up spending. PBMs renegotiate. Insurers shift prices. Hospitals consolidate. Employers absorb it. Until they can’t. Suddenly it shifts. The premium rises. Maybe. The deductible rises. Maybe. The formulary narrows. Maybe you need pre-authorization for your GLP-1 now. Maybe your specialty drug is mailed from a different pharmacy. Maybe your spouse is ineligible anymore. Maybe your company switches networks entirely. No one ever “cut healthcare”. They “optimized the benefit design.” Corporate America is filled with euphemisms for “we’re making you pay more”.

And Now the CFO Has Entered the Room

This may be THE most important trend of all. Almost 8 out of 10 employers say exec leadership is involved in GLP-1 coverage decisions, according to Business Group on Health’s 2026 GLP-1 survey. You should care about that. It’s no longer just medical directors and benefits executives deciding if your drug gets covered. People who are responsible for corporate financial performance are paying attention. And they will ask questions that are considerably less romantic than the ones pharma is used to answering. They will not ask: “How innovative is the mechanism of action?” They will ask: How many employees are eligible? What’s the net cost? How fast is utilization growing? How long do patients stay on therapy? Where are the medical savings? When do we see those savings? How much of those savings do we capture before the employee leaves the company? And eventually: Why the hell are we paying this much?

The Numbers Are Becoming Impossible to Ignore

Put the numbers next to each other:

• $26,993: average employer-sponsored family premium in 2025.
• $6,850: average employee contribution toward family coverage.
• $1,886: average single deductible among workers in plans with a deductible.
• $5,297: Aon’s estimated average employee premium-plus-out-of-pocket burden in 2026.
• $606 billion: U.S. net medicine spending in 2025.
• $14 billion: 2025 spending growth attributable to GIP/GLP-1 medicines.
• 25%: employer healthcare spending now attributable to pharmacy.
• 12%: estimated employer pharmacy-cost increase in 2026.
• 70%: employers identifying cancer as their No. 1 healthcare cost driver in 2026.
• 60%: employers covering GLP-1s for obesity in 2026, down from 72% in 2025.
• 9.2%: projected employer healthcare cost trend for 2027 before plan changes.

This isn’t sustainable mathematics. It’s a transfer mechanism. Money moves from employers and employees into an extraordinarily expensive healthcare ecosystem. And when employers finally say “enough,” the healthcare industry doesn’t absorb the pain. Employees do.

Thank You, Pharma

Thanks, pharma. Really. Thanks for developing drugs that can make patients lose 20% of their body weight. Thanks for cancer drugs that can buy patients extra months or years with their loved ones. Thanks for being able to treat diseases that doctors didn’t know how to treat just one generation ago. I’m impressed by the science stuff. But please stop patting yourselves on the back for “patient access” while you build pricing models that make access impossible to pay for.

Drug access doesn’t happen when it’s approved by the FDA. Access happens when someone can actually afford to purchase it.

And increasingly, that someone is not an employer. So they’re doing what any logical company does when a cost outpaces every other expense: They’re trimming. Constraining. Shifting costs. Switching vendors. Restricting eligibility. Adding authorization requirements. Cutting coverage. And making their employees pay more.

Pharma companies may not be directly raising your health insurance deductible, but they don’t need to. Drug companies simply raise the price of the system that sets your deductible. That’s the bit they leave out of the beautiful photos of smiling patients playing in meadows and chasing grandchildren. They should probably include one more frame at the end. The employee sitting at their kitchen table staring at next year’s benefits packet. Higher premium. Higher deductible. New formulary. Increased prior authorization. Pay-more-if-you’re-married penalty. And maybe no coverage of the drug their doctor prescribed.

Fade to black: Ask your employer if you can afford our innovations. Because that, my friends, is where we’re headed. If we don’t fix it. We have engineered a drug innovation wonderland where medicine is made that our grandparents would have thought miraculous. We just haven’t engineered a way to pay for it all. And until pharmaceutical companies, PBMs, insurers, hospitals, and legislators care about that consequence, employers will continue to “solve” the problem by making employees pay part of the bill. So thank you for the drugs, pharma. They are truly amazing. Too bad we’re working hard at making the health insurance we need to pay for them unaffordable.

Drug companies have made amazing innovations that allow us to live longer, healthier lives. They should be celebrated for the drugs that treat cancer. Diabetes. Heart disease. Infectious diseases. So many things. But there comes a point when innovating and trying to squeeze every last dime become two very different priorities.

When pharma profiteering through higher and higher margins, aggressive pricing tactics, patent extension strategies, and blockbuster drug revenues causes American employers to scale back benefits, force workers to pay more premiums and higher deductibles, and price patients out of their own medicines, we have a problem.

Perhaps the greatest challenge in confronting elevated drug prices is the narrative employees of pharma companies tell themselves: “research and development costs are too high,” “insurance companies and PBMs are the true villains,” “patients never pay the sticker price anyway,” “we need high margins to invest in future discoveries,” or “the market sets the price.” While each of these statements holds some validity, they can also serve as a defense mechanism to avoid facing an inconvenient truth:

Pharma companies consciously decide how much to charge for their drugs and how fiercely they will defend those profits. PBMs, insurers, hospitals, and government policy are all definitely at fault for America’s broken healthcare system, but blaming others does not absolve drug makers of blame. When an industry whose purpose is to better humanity starts justifying prices that price patients out of treatment, pharma employees should feel empowered to question the status quo instead of echoing it.

Yes, companies need to turn a profit in order to fund new discoveries and reward risk. But if putting shareholders first starts harming patients’ access to healthcare, it’s fair for society to question if the system works for patients. Or if patients are working for the system.

During a usability testing session, I uncovered something interesting. Users usually scrolled past the fair balance at the bottom of the page and abandoned it after just a few seconds. When we did a deeper dive, we found they read only the first few lines because they thought it was too complicated.

There is surprisingly little published research that directly measures what percentage of visitors to a pharma brand website actually read the full fair-balance/Important Safety Information (ISI). But the available evidence strongly suggests that visibility is not the same as readership, and placement has a major effect.

There is little evidence that consumers actually read lengthy pharma fair-balance disclosures in their entirety. Research shows that people scan rather than read most webpage content, and that moving drug-risk information away from the main page significantly reduces consumers’ ability to recall those risks.

Fair balance shouldn’t just satisfy regulators. It should inform patients.

The FDA mandates that prescription drug advertising disclose a fair balance of information about effectiveness and risk. In TV and radio ads directed at consumers, the most significant risks are conveyed in what’s known as the major statement. Recently, the FDA has increased its requirements. Its Clear, Conspicuous, and Neutral rule requires the major statement in DTC TV and radio ads to be presented clearly, conspicuously, and neutrally. The rule recently became effective on May 20th, 24, with compliance required beginning November 20th, 20. That’s progress. But we can do better. The standard shouldn’t just be: “Did we disclose the risk?” The standard should be: “Did the patient understand the risk well enough to have an informed conversation with their healthcare provider?”

What if the FDA simplified fair balance on all websites?

The problem with today’s risk communication

It’s understandable that pharma companies are gun-shy about fair balance. Legal, regulatory and medical affairs teams want to ensure that all material risks are disclosed. Nobody wants to receive an FDA enforcement letter because a risk was minimized or omitted. But the result is often risk information that sounds like a shortened package insert rather than something an average person could use to make a healthcare decision. And there’s irony in that. More risk information doesn’t necessarily lead to better patient understanding of the risks.

In fact, the FDA has studied alternative ways to present risk information because lengthy major statements could diminish consumer understanding and obscure the information that is most important for patients to know. FDA has said that risk information directed to consumers should be written in consumer-friendly language, and has encouraged sponsors to use language appropriate for the intended audience. So maybe it’s time for the industry to reassess what “good” fair balance sounds like.

What if fair balance sounded like this?

Imagine a hypothetical prescription drug called Cardiovexa, used to reduce cardiovascular risk in certain adults. The following is deliberately simplified and illustrative only. It is not proposed regulatory language for a real product.

Traditional-style risk communication

Cardiovexa may cause serious adverse reactions including severe bleeding, liver injury and allergic reactions. Do not use Cardiovexa if you have active pathological bleeding or have had a serious hypersensitivity reaction to Cardiovexa. Tell your healthcare provider about all medications you take. The most common adverse reactions include headache, nausea, dizziness and bruising.

Now imagine communicating essentially the same concepts around the questions a patient is actually likely to have.

A patient-centered version

What is the most important risk I should know about?

Cardiovexa can increase your risk of serious bleeding. In rare cases, serious bleeding can require hospitalization and may be life-threatening.

Who should not take it?

You should not take Cardiovexa if you are currently experiencing serious bleeding or if you’ve previously had a serious allergic reaction to the medication.

What symptoms should make me call my doctor right away?

Contact your healthcare provider immediately if you experience unusual or persistent bleeding, blood in your urine or stool, vomiting or coughing up blood, or signs of a serious allergic reaction.

What side effects are more common?

Some patients may experience headache, nausea, dizziness or bruising.

What should I tell my doctor before taking it?

Tell your doctor about your medical conditions and every prescription drug, over-the-counter medication and supplement you take because some medicines could increase your risk of bleeding.

Where can I learn about all of the risks?

Ask your healthcare provider or pharmacist and review the FDA-approved medication information for the complete list of warnings, precautions and side effects.

Same basic purpose. Very different communication experience. Instead of forcing patients to decode regulatory language, we organize information around what they actually want to know:

What could happen to me?

How serious is it?

How likely is it?

What should I watch for?

What should I do if it happens?

Those are actionable questions.

Numbers could make fair balance even better

There is one more chance. Whenever possible, patients should be provided understandable absolute risk information, rather than vague terms like “common”, “uncommon” or “increased risk”. How about hearing: “Approximately 3 out of every 100 patients taking this drug experienced this side effect, versus about 1 out of 100 patients taking placebo.” That doesn’t sound like “This drug may increase your risk of this side effect.” In fact, the FDA has issued guidance specifically addressing quantitative efficacy and risk information in consumer-directed promotion. Numbers won’t work for every label, and they can be misleading if used improperly. But when you have good information, it can help patients understand risk.

Stop confusing completeness with comprehension

I think this may be THE problem with how pharmaceutical companies communicate risk. We think that if we just give people MORE information, they are BETTER informed. That’s not true.

Ask anyone who’s had to sit through 15 side effects read so fast you can’t understand what they’re saying. They’ll remember none of them. But they may remember the three or four risks that matter most to them– if those are explained clearly, with instructions on what to do.

THAT is the difference. Communicating fair balance shouldn’t be about cramming as much information as possible into 30 seconds by a pharm exec. It should be about helping patients understand their benefit-risk equation. The FDA itself says benefit-and-risk communication is key to helping patients make informed decisions with their doctors.

Pharma should test comprehension, not just recall

Drug companies spend enormous resources testing advertising. Does the message resonate? Does the patient remember the brand? Does the advertisement motivate someone to ask their doctor about the medication? Here’s another metric that should matter:

Did the patient understand the risks?

Before approving consumer advertising, companies could test whether representative patients can answer a handful of questions after seeing the communication:

• What is the most serious risk?
• How likely is it, if that information is known?
• What symptoms should you watch for?
• When should you contact a healthcare professional?
• Who shouldn’t take the drug?
• Where can you obtain complete safety information?

If consumers can’t answer those questions, the risk communication failed—regardless of whether every required sentence appeared on the screen.

Medical, regulatory, and marketing need to solve this together

Making fair balance easier to understand doesn’t mean that marketing can eliminate awkward warnings. Far from it. It doesn’t even mean downplaying risk. It means that Medical Affairs, regulatory, legal, patient-experience and marketing teams work together to communicate scientifically accurate safety information in a way that patients can understand. That’s much more difficult than regurgitating phrases from the prescribing information. But it’s also far more useful. Key points should be retained through translation. The language itself need not.

There is also a role for AI

Generative AI has the potential to be a remarkably helpful tool for communicating risk to patients. Picture FDA-approved prescribing information as the source, with compliant outputs including: Clinical language for doctors. Plain language for patients. Large type for older adults. Spanish and other language translations. Q&A layouts. Visual risk comparisons. The underlying info stays the same. Only the communication changes. Naturally, any AI-generated drug communications would need thorough review by medical, legal, and regulatory staff. However, leveraging tech to aid comprehension should be explored.

Fair balance needs a new objective

Pharma shouldn’t argue for less disclosure. It should argue for better communication. The FDA’s movement toward clear, conspicuous, and neutral presentation is an important step. But the industry should embrace a larger principle:

Risk information should be judged not only by whether it was presented, but by whether a reasonable patient can understand what the risk means and what they should do about it.

That’s the opportunity. Simplified fair balance doesn’t mean hiding risk. It means making risk impossible to misunderstand. And if we’re serious about patients making informed decisions about their medications, comprehension—not regulatory word count—should be the ultimate measure of successful risk communication.

Being Able to Advertise a Drug Doesn’t Mean You Should

Television can be a wonderfully effective marketing tool. But it’s also expensive, noisy, hard to target, and poorly suited to communicating nuanced clinical information. Which means that TV should have to pass a very high strategic test.

We shouldn’t be asking “Can we advertise X on TV?” We should be asking “Why is TV the best use of this brand’s next $20 million, $50 million, or $100 million marketing budget?” For some products, that question will have an obvious answer. The disease affects many people. Awareness is low. Patients are highly involved in their care. Symptoms are easily recognizable. The treatment benefit can be clearly communicated. And consumers can influence their physicians’ prescribing decisions. Great. Market away. Buy a bunch of television.

Unfortunately, those attributes don’t apply to every pharma product advertised on television. Some products treat complex diseases where only specialists can prescribe the medicine. Some have small populations. Some have clinical drivers that consumers can’t begin to evaluate. Some have problematic safety profiles. And some brands don’t have enough of a consumer value proposition to warrant advertising, period. But they end up on TV anyway. How does that happen? Simple. Somewhere along the way, marketing got confused with advertising.

Agencies Are Supposed to Execute Strategy, Not Become the Strategy

Some fantastic advertising agencies work for pharma companies. The issue isn’t agencies. The issue is when pharma companies relinquish too much marketing decision-making to agencies.

Ask an advertising agency if your brand needs advertising, and you shouldn’t be surprised when advertising is part of the solution they recommend. That’s not an attack. That’s their job. Pharma should have savvy marketers who aren’t afraid to challenge that recommendation.

An experienced brand leader should be able to look an agency in the eye and say, “Sell me on television being more effective than every other venue we could spend this money on,” and then keep asking questions. What patient behavior do we want to change? Where do patients actually consume information? What is the barrier to treatment initiation? Is it awareness? What happens after patients see a commercial? Will doctors welcome the resulting patient conversations? What percentage of the audience has any realistic chance of becoming a patient? What is the incremental return? And most importantly: What opportunities are we losing because we’re investing this money in television? Experienced marketers ask tough questions. Inexperienced marketers often just ask their agencies for a recommendation. There’s a huge difference.

The Best Marketers Know When NOT to Advertise

Marketing is not about making campaigns. Marketing is about investing limited resources to change behavior. Sometimes that requires advertising. Sometimes it doesn’t.

Maybe the greater opportunity is patient education. Maybe patients are dropping off during diagnosis. Maybe reimbursement friction is stifling prescriptions. Maybe physicians fail to grasp the clinical differentiation. Maybe the website isn’t answering patients’ questions. Maybe search behavior is showing consumers are seeking information the brand isn’t addressing. Maybe adherence is poor. Maybe there is no patient-support program. Maybe physicians just don’t believe the product offers sufficient incremental value.

A great TV commercial doesn’t fix those problems. Smart marketers know this. Because they’ve lived through campaigns that failed. Campaigns that earned gigantic media budgets and dismal business results. They know awareness does not inevitably lead to prescriptions. And they know what drug companies sometimes forget: Activity and effectiveness are different things.

Pharma Has Become Very Good at Producing Marketing

This is critical.

Big Pharma is very good at creating marketing documents. Brand plans. Creative briefs. Message platforms. Segmentation studies. Market research. Patient journeys. Advisory boards. Media plans. Agency presentations. Dashboards. PowerPoint after PowerPoint covering every aspect of every disease state on the walls of conference rooms. Endless meetings to discuss each of them. Creating marketing documents and meeting about them is not the same as understanding a market.

The best marketers have far more difficult-to-produce skill sets: judgment. Judgment is earned by experiencing setbacks. Launching products. Missing forecasts. Having competitors catch you by surprise. Listening to customers. Battling for budgets. Making poor decisions. Making brilliant decisions. Then living through the results long enough to know which were which. That learned marketing intelligence is invaluable. The problem is that pharmaceutical companies have inadvertently become experts at pushing away the people who acquire that intelligence.

Why Experienced Pharma Marketers Keep Leaving

Sit down with frustrated veterans who have worked for big pharma companies, and certain patterns emerge.

The bureaucracy drains you. Decision-making feels glacial. Internal politics suck up incredible amounts of energy. Playing it safe becomes more prized than innovating. Compliance programs balloon. Meetings happen more frequently. Agencies cluster. Sign-offs increase. And sadly, marketers can slowly morph into managers of the marketing process rather than practitioners of marketing.

What was once a senior marketer thinking about customers, competitors, positioning, and market opportunities can turn into spending most of the week focused on managing agency relationships, budgets, internal reviews, procurement, legal and regulatory approvals, cross-functional meetings, dashboards, and senior-management presentations. All of those things are important. But when they become your full-time job, good marketers will eventually wonder: “When did I stop doing marketing?” And walk away.

Pharma’s Promotion Culture Doesn’t Help

Here’s another issue. Big organizations tend to promote those who become adept at navigating the organization.

Understanding how to operate inside a large organization is a skill. However, organizational skills and marketing acumen are different. You can be a master at managing up, building consensus, creating slick slides, and avoiding risky decisions. It doesn’t necessarily mean you know customers. The marketer who questions the annual plan may actually be the more valuable employee.

The individual who says, “We spent $70 million on this last year. Show me why we should do it again,” could be exactly what your organization needs. But questioning $70 million in established spending is difficult. Copying last year’s plan is safe. And “safe” is rewarded in corporate America. Thus, institutionalized marketing is born. We replicate last year’s TV campaign this year because we already have TV in the plan. We have an agency relationship in place. We have a budget parked in a spreadsheet somewhere. Management “gets” it. Everyone knows what to do. And no one will get fired for rolling out another Pharma TV spot. But just because it’s safe doesn’t mean it’s good marketing.

Then Pharma Wonders Why Agencies Have So Much Power

When seasoned internal marketing expertise leaves, a void is created. Enter a replacement. Oftentimes, agencies. Now agencies are doing more than just creating work. They are influencing positioning. They are reading research. They are suggesting channels. They are building strategy. They are defining customer journeys. They are delivering competitive reviews. They are proposing how budgets should be spent.

Soon your organization relies on external resources for skills that should be nurtured in-house. When that happens, an inherent conflict of interest emerges. The consultants who are recommending how you should spend your money may also profit from where you spend your money. This doesn’t make agencies bad people. It makes strong internal marketing leadership critical. The drug company needs to own the strategy. Let the agency help you make it shine. They are very different roles.

The Cost of Losing Experienced Marketers Is Bigger Than Salaries

Companies tend to look at experienced employees through a compensation lens. They shouldn’t. Sure, that veteran marketer is likely going to cost you much more than someone early in their career. But that experienced marketer can ask the one question that saves you millions of dollars.

There’s the economics companies should be looking at. Let’s say you have a brand with a $100 million promotional budget. If that experienced marketer improves how that budget is allocated by just 5% you’ll see a huge economic impact. Far larger than the difference in pay between that marketer and a younger employee you might hire to replace them.

Marketing experience is not expensive. Ineffective marketing decisions are expensive. The issue is that compensation shows up neatly on a financial statement. Bad judgment gets lost in the marketing budget.

Pharma Needs Marketers Who Are Willing to Say “No”

We don’t need another television commercial. Awareness is not our biggest problem. The agency recommendation doesn’t make sense. Physicians will not believe that message. Patients aren’t behaving the way our segmentation deck says they are. Another $30 million in media won’t fix weak differentiation. Copying the competitor is not a strategy. And every once in a while we have to say: No, this product probably shouldn’t be advertised on television at all.

Saying those things requires confidence. Confidence is usually derived from experience. Experience comes from spending years making real marketing decisions and being responsible for the results.

Stop Treating Marketing Like Project Management

Pharma needs to define what they want their marketers to be. If they want project managers to coordinate agencies, route materials through review, manage timelines, and prepare management presentations, then call it project management. Let marketing be something else. Marketing needs to understand customers. Competitors. Behavioral economics. Media. Data. Positioning.

The clinical environment enough to know when commercial ambition crashes into medical reality. And most importantly, they need to understand the business enough to know WHERE NOT to spend money. That’s where experience is valuable. AI will analyze data. Agencies will create campaigns. Consultants will build frameworks. Research companies will deliver insights. Media agencies will buy audiences. But someone in Pharma will need to connect it all and make the call. That’s marketing leadership.

Maybe That TV Commercial Is Telling Us Something

The next time you watch a pharma commercial and ask yourself, “Why are they advertising that drug to me?” You may not be watching a bad advertisement. You may be witnessing the downstream effect of deeper organizational issues: Too much dependence on agencies. Too much focus on execution. Too much bureaucracy. Too little appetite for questioning sunk costs. And not enough experienced marketers with the intestinal fortitude to say: “This doesn’t make sense.” Pharma doesn’t need more creative agencies. It needs more marketers. And pharma will know it needs more marketers when it recognizes that seasoned marketers are not interchangeable cogs in a corporate machine. They have institutional knowledge, perspective and scar tissue that cannot be outsourced to an agency or learned from a research brief.

Continue marginalizing those employees and agencies will gladly continue to perpetuate themselves. And consumers will continue to ask themselves: Why are they advertising that drug to me?

The data is becoming harder to ignore with each passing year. Recently, the Kaiser Family Foundation (“KFF”) reported that the average annual premium for employer-sponsored family health coverage hit $26,993 in 2025. Employees are shouldering an average of $6,850 of the total premium cost. Year-over-year family premiums rose 6% and are now 53% higher than 2015 levels.

Healthcare costs are rising, and they are expected to rise even faster going forward. According to the Business Group on Health’s annual survey, employers project a median healthcare cost increase of 9.2% in 2027 before plan changes and 8% after plan changes. If the trends we are seeing for 2026 and 2027 continue, Health care costs could increase by 76% over the next decade. Employers can keep whining. Or they can start acting like buyers.

1ne. Audit Your PBM Contract—Really Audit It

For many employers, pharmacy has become one of the fastest-growing components of healthcare spending. Business Group on Health reports that pharmacy now represents approximately 25% of employers’ total healthcare spending, with employer drug costs estimated to increase 12% in 2026. Yet too many employers still don’t understand exactly how their pharmacy benefit manager makes money.

Employers should know:

• What rebates are being collected?
• What percentage actually reaches the employer?
• Are there spread-pricing arrangements?
• How are specialty pharmacies compensated?
• Are there administrative fees hidden elsewhere in the contract?
• Who receives manufacturer fees and other payments?
• How is “generic” defined?
• How is “specialty drug” defined?
• Are drugs being selected because they produce the lowest net cost—or the largest rebate?

Employers should consider periodic independent PBM audits and competitive bidding rather than automatically renewing incumbent arrangements. The question shouldn’t be, “How big is our rebate check?” It should be: “What is our total net pharmacy cost?” Those are very different questions.

2wo. Stop Treating Hospital Prices as Untouchable

One of the largest opportunities may be staring you right in the face: hospital prices.

Recently, researchers at RAND found that employers and private insurers paid hospitals roughly 254% of Medicare prices on average for comparable inpatient and outpatient services in 2022.

Let that sink in for a moment. Employers often aggressively negotiate office leases, cloud-computing contracts, and airline travel… yet they may be paying more than twice Medicare’s rate for hospital services. Employers need to see what they are really paying individual health systems.

Pull the claims. See which hospitals are costing you the most money. Compare what you’re actually being charged to what you think you’re being charged. Evaluate price against outcomes. Then direct employees to hospitals that provide better value. Finally, larger employers should explore direct contracting, high-performance networks, and other models that eliminate pieces of the traditional healthcare purchasing system.

3hree. Stop Assuming Your Health Plan Is Negotiating Aggressively for You

Employers need to ask themselves an uncomfortable question: Does my insurer have any financial incentive to lower hospital charges paid to me? Many employers turn over massive segments of their health care purchasing strategy to carriers and consultants. That does not absolve the employer of accountability. Insist on seeing how much you’re paying after negotiated rates, utilization, outcomes and total cost are factored in. Then benchmark them. Ask your carrier to explain why its network provides a good value—instead of just telling you that its discounts are deep. A 50% “discount” off of an insanely inflated hospital charge isn’t necessarily a good deal.

4our. Identify the 20 or 30 Things Actually Driving Your Healthcare Spending

Stop trying to manage healthcare by averages. Drill down into the claims. What conditions are costing the most? What hospitals? What physicians? What drugs? What procedures? What sites of care? Who are the employees or dependents with catastrophic claims?

The latest Business Group on Health survey revealed cancer was the No. 1 cost driver among the 70% of employers that reported it as such in 2026. Musculoskeletal and cardiovascular conditions were also commonly cited as cost drivers.

So instead of spending money on another cookie-cutter “wellness solution,” employers should create strategies targeted at their unique set of cost drivers. If spending on musculoskeletal issues is out of control, tackle musculoskeletal care. If cancer costs are spiraling, look at oncology CoEs and cancer navigation. If specialty drugs are your biggest driver, crack down on specialty pharmacy pricing. Look where the money is going.

5ive. Get Much More Aggressive About Centers of Excellence

Employees understandably assume that the closest hospital or the biggest hospital brand provides the best care. That’s not necessarily true. Employers can identify high-quality providers for expensive procedures and conditions and encourage employees to use them.

That might include:

• Cancer
• Joint replacement
• Spine surgery
• Cardiac procedures
• Transplants
• Bariatric surgery
• Complex maternity care

Employers can make Centers of Excellence significantly more attractive by reducing or eliminating employee cost sharing and, when appropriate, covering travel expenses. Business Group on Health reports that 84% of surveyed employers plan to offer at least one Center of Excellence in 2027. The objective isn’t simply lower prices. It’s avoiding unnecessary procedures, complications, readmissions and poor outcomes.

6ix. Attack Site-of-Care Costs

Prices can vary wildly based on where a drug, infusion, scan or procedure is delivered. Treatment received in a hospital outpatient department, for example, can cost significantly more than the same treatment provided in a physician office, freestanding infusion center or sometimes even the patient’s home. Employers should pinpoint specific procedures and treatments with wide site-of-care price variations and structure benefits to steer employees toward low-cost, clinically suitable alternatives. That’s healthcare purchasing 101: If you can safely get something for $2,000, don’t pay $5,000.

7even. Stop Adding Vendors Without Proving They Save Money

Benefits departments at employers have built up a ridiculous vendor zoo over the years. Mental health platforms. Diabetes programs. Weight-management programs. Navigation companies. Second opinion services. Musculoskeletal vendors. Virtual primary care. Cancer navigation. Fertility benefits. Sleep programs. Wellness apps. Some are great. Some probably save money. But many will just charge another per-member-per-month fee.

Every vendor should be held to objective performance metrics. Employers should be asking vendors: Did you reduce total medical spending or materially improve outcomes? I’m not talking about engagement. App downloads. Or employee satisfaction surveys. I’m talking about outcomes. If a vendor can’t back up their value with hard numbers, employers should consider cutting them loose. Seriously. 58% of employers surveyed by Business Group on Health say they’ve already cut underperforming vendors.

8ight. Treat GLP-1 Drugs as a Healthcare Strategy, Not Just a Pharmacy Problem

GLP-1 drugs provide a perfect example of why benefit silos fail. The pharmacy benefit will experience significant drug spending. The medical plan has the potential to see decreased spending on diabetes complications, cardiovascular events and other obesity-related conditions.

Covering the drugs with no plan in place could be incredibly costly. Refusing coverage could also be short-sighted. Employers need eligibility criteria, clinical management, negotiated pricing, outcome tracking and longitudinal analysis of total medical spend. What’s interesting is employers are starting to rethink their strategy: Business Group on Health found GLP-1 obesity coverage among employers they surveyed decreased from 72% in 2025 to 60% in 2026.

Instead of asking: “How much are we spending on GLP-1s?” Employers should be asking: “What is happening to the total healthcare cost and health outcomes of employees receiving them?”

9ine. Make Primary Care Easy—and Valuable

Workers shouldn’t have to wait three weeks just to see a primary-care doctor and then give up and visit urgent care or the ER. Employers can look at enhanced primary-care models, virtual PC, near-site clinics and other ways to increase access. But remember: employers should track outcomes. Did visits to the ER decrease? Were chronic diseases caught earlier? Did patients take their medications as directed? Did visits to specialists become more appropriate? Did overall cost go down? Primary care should be the front door to the health system, not just another disjointed benefit.

10en. Put Healthcare Spending on the CFO’s Dashboard

This may require the largest shift in thinking for many employers. Historically, healthcare benefits have been treated as solely an HR issue. At current spending levels, that approach is no longer feasible.

Consider that a business with 10,000 employees can have annual healthcare costs in the tens or even hundreds of millions of dollars when you include dependents. That should be on the CFO’s radar. Track healthcare spend like any other significant business expenditure. Create benchmarks (KPIs) for: PMM cost. Medical trend. Pharmacy trend. Specialty drug usage. Hospital price variation. ER visits. Preventable admissions. Cancer cases. Musculoskeletal cases. High-cost claimants. Vendor ROI. Employee out-of-pocket costs. Then hold management accountable by reviewing them frequently.

Stop Passing Every Increase to Employees

Wait, there’s one more tactic employers have turned to for years: Raise deductibles. Raise copays. Increase employee premium contributions. Problem solved. Not so fast. Employees are healthcare consumers navigating a system they don’t understand, with hidden prices and life-altering clinical decisions. Transferring more expenses to employees may lower your costs in the short-term, but it also leads to employees skipping needed care. That’s cost shifting – not cost management.

Employers Have More Power Than They Think

Many employers act like victims in the healthcare arena. They’re not victims. Employers provide health insurance for most Americans. Employers spend hundreds of billions of dollars on hospitals, health insurers, PBMs, pharmaceutical companies, consultants, and healthcare vendors. That’s leverage.

Recent data shows that some employers are starting to use their leverage. In a recent Business Group on Health survey, 71% of employers said they used the RFP process to negotiate lower prices, 60% said they were increasing their focus on prevention and primary care, and 58% said they had cut contracts with poor-performing vendors.

More employers should feel empowered to do those things. Demand transparent contracts. Audit your PBMs. Benchmark your hospitals. Measure health outcomes. Reduce your vendors. Pay providers that deliver high value. Question consultants. Analyze your claims data. Manage your specialty pharmacy. And don’t settle for 8%, 9%, or 10% annual healthcare inflation. Employers will never be able to fully control healthcare costs. But there is a big difference between being unable to control healthcare inflation and not trying your hardest to control it.

Before you accept next year’s healthcare increase, ask yourself this: What have we done to earn a different result?

America’s biggest self-insured employers may want to take matters into their own hands. Instead of waiting for PBMs to fix the nation’s prescription-drug problem, they could form their own drug purchasing coalition.

The concept isn’t complicated: Pool together enough employees and dependents to represent millions, if not tens of millions, of insured lives. Determine which drugs drive the majority of your pharmacy spend. Then invite drug manufacturers to bid on that business.

Negotiate transparent net prices directly with them, and then contract with vendors to handle only the administrative functions you really need. Put simply: Purchase drugs separately from managing the pharmacy benefit. It would upend the employer-sponsored pharmacy business as we know it.

Employers Have a Drug-Cost Problem

Prescription drugs are devouring employer health care budgets.

According to a recent survey from Business Group on Health, pharmacy now accounts for roughly 25% of employers’ health care spend, and will grow another ~12% in 2027. No surprise, GLP-1 obesity drugs have dominated the conversation lately. Confronted with skyrocketing utilization and cost, 14% of employers surveyed have already dropped GLP-1s for weight management or plan to do so this year. But GLP-1s are just the tip of the iceberg. Employers are also facing huge bills for cancer drugs, immunology therapies, specialty pharmacy, gene therapy, rare disease – you name it.

The traditional solution has been to squeeze the PBM. Consider a different approach: What if employers negotiated directly with drug manufacturers? Buying Clout Is Already Concentrated. Let’s look at America’s largest companies. How much purchasing clout do they represent collectively? It’s hard to know exactly, but consider a hypothetical coalition of 20, 50, or 100 large self-insured employers. Between them, could they represent millions of covered lives? Sure. Would that make them a coveted customer? You betcha. Rather than having each employer negotiate a contract with a PBM (which then negotiates manufacturer rebates on the PBM’s behalf), the group could deal directly with drug manufacturers.

Imagine that the group sits down together and realizes that across all of its members, it spends billions of dollars on a portfolio of costly drugs. It writes up a document that’s functionally very similar to a request for proposal, and sends it to pharma companies: Here are our members. Here is our utilization. Here is the value of the market we represent. What are your best net prices? Drug manufacturers would then compete to win access to that population. That is buying power. Corporate America knows a thing or two about buying power.

Start With the Drugs That Cost Employers the Most

That agreement wouldn’t have to cover every drug either. Begin with maybe the 25 or 50 drugs that account for the most total spending by the member companies. The buying entity could assess claims across its membership and pinpoint where pooled bargaining has the most potential. Within competitive classes of therapeutics, manufacturers could compete against each other for preferred position. For non-competitive drugs, the alliance could negotiate bulk pricing, outcome-based agreements, utilization guarantees or some combination. The key distinction would be visibility. The employer group would see:

-The manufacturer’s price. Not the sticker price minus a rebate minus PBM fees minus other payments cleverly disguised to flow through multiple parties. The real negotiated net price.

GLP-1s Show That This Isn’t Just Theory

The pharma market is already moving toward direct employer relationships. Eli Lilly says it launched a commercial platform for employers in 2026 and describes Employer Connect as an access pathway that complements traditional insurance coverage. Lilly also says it intends to keep developing value-based arrangements that link medicine costs to outcomes.

Healthcare Dive reported earlier this year that Lilly and Novo Nordisk were testing direct-to-employer approaches that could reduce employers’ dependence on PBMs. Perhaps even more interesting is the emergence of cooperative purchasing.

Andel launched an employer cooperative in March built around obtaining medications directly from manufacturers, initially offering Lilly’s Zepbound. According to the company, its model doesn’t require a PBM, TPA or insurance arrangement for the medication program.

The infrastructure is starting to appear. The next logical step is scale.

Here’s How an Employer Drug Alliance Could Work

Imagine 50 major corporations establishing the Employer Pharmaceutical Purchasing Alliance, or EPPA. It would not be another PBM. It would be a purchasing organization owned or controlled by the employers themselves.

Step 1ne: Aggregate the Lives

Participating employers contribute anonymized utilization and spending data. The alliance might discover that collectively its members cover, for example, 8 million employees and dependents. Now those companies aren’t negotiating as 50 separate healthcare purchasers.

They’re negotiating as one purchaser representing 8 million potential patients.

Step 2wo: Identify the Biggest Opportunities

The alliance ranks medications according to total spend rather than list price. Perhaps the top categories include: GLP-1 obesity medications, oncology drugs, autoimmune therapies, diabetes drugs and specialty medications. The alliance focuses its negotiating resources where the dollars are.

Step 3hree: Invite Manufacturers to Bid

Manufacturers would receive RFPs for therapeutic categories in which competition exists. For obesity, for example, manufacturers could compete on far more than price. The RFP might evaluate:

• guaranteed net price;
• price protection for future increases;
• clinical outcomes;
• patient persistence;
• supply guarantees;
• adherence programs;
• real-world outcomes data; and
• performance guarantees.

The winning manufacturer could receive preferred formulary positioning across millions of covered lives. That’s something worth negotiating for.

Step 4our: Eliminate the Rebate Game

Here’s where the model gets interesting. The alliance wouldn’t ask: “What’s the rebate?” It would ask: “What’s the price?” That distinction matters. The FTC has alleged that PBM rebate structures can encourage competition based on rebates tied to higher list prices rather than simply lower net prices.

The FTC’s subsequent settlements with Express Scripts and Caremark have included requirements intended to move standard offerings away from rebate guarantees and spread pricing and toward greater drug-level transparency.

Employers could take the concept further. Forget rebate guarantees. Negotiate the acquisition economics directly.

But Someone Still Has to Process the Prescriptions

I’m not advocating eliminating every task PBMs perform today if we bypass them on drug-price negotiations. Employers will still require: claims processing, pharmacy networks, eligibility verification, prior authorization, specialty drug management, formulary administration and data analytics. Those services can be unbundled and purchased separately. The employer alliance could competitively bid contracts for each. A claims administrator could be paid a transparent, per-member-per-month fee. A specialty pharmacy could be paid acquisition cost plus a transparent dispensing and service fee. A clinical management company could be paid for utilization management. Those are defined services. Pay vendors for services rendered, not hidden in the cost of the drug.

The PBM Becomes a Vendor, Not the Purchasing Gatekeeper

That’s the fundamental shift in thinking. Employers don’t have to get rid of PBMs altogether. They just need to unbundle them. PBMs could still bid for the business of processing claims or managing a pharmacy network. But the PBM would not automatically get to handle manufacturer negotiations + rebates + formulary placement + pharmacy reimbursement + specialty pharmacy + claims processing. Those could all be unbundled and competitively bid. Corporate America buys literally everything else this way. Why not drug benefits?

Employers Should Own Their Data

One more membership requirement: Employers have full access to their pharmacy data. All prescriptions. All manufacturer payments. All administrative fees. All pharmacy reimbursements. All clinical interventions. All dollars. Without full transparency into that data, employers have no way to know whether they are saving money. This matters because the PBM industry is extremely concentrated. The FTC found that the top six PBMs processed nearly 95% of all U.S. prescriptions last year. Consolidation gives even more reason to form purchasing groups outside of the traditional model.

Manufacturers Might Actually Like It

You might think Big Pharma would hate the idea of negotiating directly with employers. You’d be right…to an extent. Some pharmacy benefit managers would certainly oppose it. But others might view direct negotiation with employers as a huge business opportunity.

Offering preferred access to several million commercially insured patients in return for a guaranteed transparent price could be very attractive to drug companies. No negotiating rebates through a bunch of intermediaries. No risk of not making formulary. No messy rebate waterfall. Price + volume + access. Drug manufacturers already transact with sophisticated purchasers this way all across healthcare. Why not large employers? One indication that pharmacy benefit managers may face real competition is that drug manufacturers are willing to experiment.

Lilly’s recent employer initiative, combined with several new direct-to-employer platforms entering the market, suggests that the traditional PBM middleman isn’t the only way to reach commercially insured patients anymore.

There Is One Major Caveat: Antitrust

Yes, but an employer purchasing alliance couldn’t simply be a bunch of corporate buyers sitting in a room together deciding what pharma companies should pay. It would have to be designed to avoid running afoul of antitrust laws. It would have to have its own independent governance, legal counsel and tightly constructed joint-purchasing agreements.

The participating employers would also have to refrain from exchanging information that could be considered competitively sensitive except as it relates to healthcare purchasing. That isn’t a reason to throw up our hands and not explore the concept. It’s a reason to explore how to do it the right way. Healthcare already has purchasing groups and coalitions. The trick is forming one whose interests align solely with the employers paying for the benefit.

Imagine the Negotiating Table

Picture the same drug maker across the table from negotiators for an alliance representing 10 million Americans. The dynamics of that negotiation change drastically. The alliance can say to the manufacturer: “We want your medicine covered for our employees. We also negotiate on behalf of 10 million covered lives. Provide us your best transparent net price. Guarantee us supply. Show us outcomes. And protect our members from unreasonable price hikes. If not, we will look to your competitor as a viable alternative.” That is a far cry from the leverage of an individual employer questioning whether the rebate guarantee it’s getting from its PBM is a good price.

Corporations Negotiate Everything Else

Big businesses haggle over the price of technology. Transportation. Raw materials. Energy. Consulting. Cloud services. Advertising. Virtually every other major expense line. They consolidate purchasing volume. They send out RFPs. They insist on competing bids. They negotiate multi-year contracts. They demand performance guarantees. They audit their suppliers. But when it comes to one of their fastest-growing costs—prescription drugs—many just hand over their negotiating leverage to intermediaries with extremely opaque economics. We think it’s time to push back.

The Biggest Employers Don’t Need More Rebates. They Need Buying Power.

PBMs justified their existence in part by arguing that they could pool together millions of lives which would give them scale to negotiate with manufacturers. That’s fine. But America’s largest employers collectively control millions and millions of lives too. Why lease someone else’s purchasing power when employers can wield their own? Welcome to the new world of direct employer pharmaceutical programs.

The wall between manufacturers and employers is already starting to crumble. Why not tear it down? Form an independent employer-owned pharma purchasing coalition.

Pool lives. Focus on the costliest drugs. Force manufacturers to compete against each other. Negotiate transparent net prices. Unbundle pharmacy admin fees. Pay transparent fees for actual services rendered. Audit every penny spent. And split the savings with employees. America’s prescription-drug supply chain has built up layer upon layer of middlemen, rebates, fees, and non-transparent contracts. Maybe the country’s largest corporations don’t need another guy walking into their offices telling them how to get a better deal from their PBMs. Maybe they need to simply sit at the table across from pharma without PBMs.

I knew something was wrong. My coworker was always the first one in the office and often stayed late. At first I thought it was her work level, so I asked what projects she was working on. I soon found out it wasn’t the amount of work she had. She had just exited a long-term relationship and was feeling lonely. More and more I could see the impact it had on her life, and it was hard to see such a great person go through this.

There’s one potential cost driver we don’t talk about enough: Loneliness.

More and more, loneliness and social isolation are being linked to worse health outcomes and outcomes, higher economic costs, and—even in some populations—increased healthcare utilization. These numbers are big.

Last year, the U.S. Surgeon General’s office issued a report stating that social isolation in older adults contributes to about $6.7 billion in excess Medicare spending annually. And remember, that just covers Medicare expenditures. One recent systematic review of the economic literature found that loneliness and social isolation are associated with annual costs to society ranging from about $2 billion to $25.2 billion—mostly from healthcare spend and lost productivity.

The message for healthcare leaders should be clear: Social connection may be something we will need to manage as part of the healthcare delivery process.

Why Would Loneliness Increase Healthcare Costs?

The answer starts with a crucial qualifier. Loneliness and social isolation are not the same thing. Loneliness is subjective: a person doesn’t feel connected or satisfied with their relationships. Social isolation is objective: a person lacks social contacts, relationships or support. Because of this, a patient may be socially isolated but not feel lonely—or be surrounded by others but feel extremely lonely. Both matter because social connection affects health in many ways.

The CDC notes loneliness and isolation are linked to higher risks of heart disease, stroke, type 2 diabetes, depression and anxiety, dementia, and premature death. It can also affect health through day-to-day healthcare activities. Imagine a 72-year-old patient with diabetes, hypertension and heart failure who lives by themselves. Who detects that patient’s breathing has worsened? Who reminds them to refill their medications? Who takes them to doctor’s appointments? Who realizes they aren’t eating properly? Who can encourage them to call the doctor before a problem worsens into an emergency? Families and friends can provide social support that acts as an extra layer of healthcare safety net. When that safety net unravels, health systems may feel the impact down the road.

Medicare Provides a Window Into the Potential Cost

One of the most commonly referenced U.S. studies looked at social isolation among people on Medicare. The study found that about 14 percent of older adults enrolled in traditional Medicare—about four million people at the time—had few, if any, social connections.

Medicare spending for socially isolated enrollees was about $1,608 higher per person per year, resulting in approximately $6.7 billion in extra Medicare spending annually. $6.7 billion is a lot of money for something that rarely, if ever, shows up in bold letters on a patient’s problem list.

If a medication added billions to the Medicare program, we would expect policymakers to take notice. If unnecessary procedures added billions of dollars to health plans’ costs, we would expect health plans to develop utilization management programs to reduce those services. For the most part, the healthcare system has failed to look at social isolation through an economic lens. It’s time that changed.

But There’s an Important Caveat

Healthcare leaders shouldn’t make loneliness the next reductionist healthcare battle cry. While the link between social connection and healthcare spending is intuitive, the data is somewhat inconclusive.

One 2025 systematic review examined 53 studies, including eight that estimated healthcare costs/expenditures; 33 looked at healthcare resource utilization, and 19 were economic evaluations of loneliness interventions. Results were mixed, with some studies showing higher cost/utilization, others showing no significant difference, and some showing lower expenditures for lonely populations. “The current evidence is insufficient to accurately estimate the total direct healthcare-system cost that can be attributed to loneliness.” This matters because we shouldn’t claim that every lonely patient is a high-cost patient.

However, just because some of the cost estimates are all over the place doesn’t negate the mountain of evidence connecting disconnection and health – it shows we have work to do on measurement.

Healthcare Organizations Measure Almost Everything—Except Social Connection

Health systems and insurers routinely collect enormous amounts of information.

A1C.

Blood pressure.

BMI.

Medication adherence.

Smoking.

Alcohol consumption.

Depression.

Hospitalizations.

Emergency department visits.

Social connection rarely receives comparable attention.

That creates an obvious question:

Should loneliness and social isolation become routinely measured healthcare risk factors?

For Medicare Advantage plans, health systems, accountable care organizations and other organizations assuming financial risk for populations, the answer deserves serious consideration.

A short loneliness or social-connection assessment could potentially become another data point within population-health analytics.

Organizations could then examine whether socially disconnected patients demonstrate differences in:

hospital admissions,

emergency department utilization,

medication adherence,

missed appointments,

behavioral-health utilization,

post-discharge outcomes,

skilled nursing utilization,

and total cost of care.

Instead of assuming loneliness increases healthcare costs, organizations could determine whether it actually predicts costs within their own populations.

There May Also Be an ROI Opportunity

This idea gets particularly interesting when you start looking at intervention studies. A recent systematic review of loneliness and social-isolation economics was updated to include studies estimating annual economic costs of loneliness/social isolation ranging from about $2 billion to $25.2 billion.

More significantly, all five social-return-on-investment studies included in the update found positive returns, with estimates ranging from about $2.28 to $13.72 returned for every $1 invested.

Once again, those numbers should be taken with a grain of salt. The researchers point out limitations in the studies they reviewed, including challenges in proving that the intervention caused all the observed benefits. But they do open the door for an interesting possibility. Some interventions that promote social connection can create economic value. Healthcare organizations don’t necessarily need to start running social clubs. They do need to start experimenting.

Treat Loneliness Like a Population-Health Problem

Healthcare organizations could begin by identifying populations where loneliness and isolation potentially overlap with high medical risk. For example, patients who are older, living alone, recently widowed, chronically ill, frequently hospitalized, or experiencing transportation barriers may warrant additional assessment.

Interventions could then range from community programs and peer-support networks to social prescribing, transportation assistance, volunteer programs, caregiver engagement, and technology designed to facilitate meaningful human interaction.

But every program should have measurable outcomes.

Healthcare organizations should ask:

Did loneliness decline?

Did medication adherence improve?

Did appointment attendance improve?

Did emergency department utilization change?

Did hospitalizations decline?

Did quality of life improve?

And ultimately, did total healthcare spending change?

That is how loneliness moves from a well-intentioned social initiative into a serious population-health strategy.

Employers Should Pay Attention Too

The economic consequences aren’t limited to insurers and Medicare. The Surgeon General’s advisory cites research estimating that stress-related absenteeism associated with loneliness costs U.S. employers approximately $154 billion annually. That changes the discussion considerably.

Employers evaluating healthcare costs traditionally concentrate on medical claims, pharmacy benefits, PBM contracts and high-cost claimants. They may also need to think about social connection as part of workforce health. Because the economic impact of poor health doesn’t stop when an employee leaves the doctor’s office. It appears in absenteeism, productivity, disability and turnover.

The Next Healthcare Cost Frontier May Be Social

For decades, American healthcare has largely been organized around what happens after someone becomes sick.

We diagnose disease.

We prescribe medications.

We perform procedures.

We hospitalize patients.

Then we wonder why healthcare expenditures continue increasing.

Loneliness exposes one of the limitations of that model.

Some determinants of healthcare utilization may begin long before a patient enters a hospital or physician’s office. The CDC describes social connection as an important contributor to better physical and mental health. That means healthcare executives, employers and policymakers may need to broaden their definition of healthcare infrastructure.

Doctors matter.

Hospitals matter.

Medicines matter.

Insurance matters.

But families, friendships, caregivers, neighbors and communities may matter economically too.

The healthcare system does not yet have a definitive dollar figure for the total cost of loneliness—and the newest systematic evidence cautions against pretending that it does.

But we already know enough to start asking a different question.

Instead of asking:

“Is loneliness really a healthcare issue?”

Healthcare leaders should begin asking:

“How much is social disconnection costing our population—and what happens if we actually do something about it?”

That could become one of the most consequential population-health questions of the next decade.

Over the last couple of weeks, I have been busy leading usability research for the launch of a newly approved brand. The feedback we received was great, and online health seekers really know what they want from a branded pharma website. During the research, AI kept coming up when we asked, “Where would you go to get…”. Unfortunately, too many pharma companies are afraid of integrating AI, but it can be done within FDA guidelines.

AI is poised to transform the consumer health information experience. However, the typical pharmaceutical brand website hasn’t changed much in the past ten years. Visitors to these sites have clear questions: How effective is this medicine? What are the side effects? How should I take this medication? What if I miss a dose? Should I take this drug? Answers may be readily available on the site, but patients are often required to sift through numerous pages, Important Safety Information, FAQs, and an encyclopedic Prescribing Information document to find what they need. There is likely a better way.

Don’t Let the AI Invent the Answer

The biggest mistake would be giving a generative AI model internet access and allowing it to formulate answers about a prescription drug. That’s an unnecessary regulatory risk. Instead, AI should primarily determine what the visitor is asking. The actual answer should come from a controlled and approved content environment.

The architecture could look something like this:

Patient Question → AI Intent Detection → Approved Content Retrieval → Regulatory/Safety Rules → Response → Audit Log

Suppose someone asks: “What are the most common side effects?” The AI recognizes that the visitor is asking about adverse reactions. Instead of generating an answer, the system retrieves an MLR-approved response based on the current FDA-approved Prescribing Information. That difference is crucial.

The system isn’t asking AI to practice medicine or create promotional claims. It’s using AI to make approved pharmaceutical information easier to access.

Think “Conversational Prescribing Information”

Perhaps the best way to describe the concept isn’t an AI chatbot at all.

Think of it as:

Conversational access to approved product information.

A patient shouldn’t need to understand how pharma websites are organized to find an answer. They should simply be able to ask the question. For example:

Patient: “How well does Drug X work?”

The system recognizes an efficacy question and retrieves the appropriate approved clinical-trial information along with the required safety context.

Patient: “How often do I take Drug X?”

The system retrieves approved dosing information.

Patient: “Who shouldn’t take this drug?”

The system retrieves contraindications and relevant safety information.

The experience feels conversational to the consumer, but underneath it is a highly controlled information-retrieval system.

An example of AI integrated into a fictitious pharma website

Six Guardrails Would Be Essential

A pharma AI query engine would need significant regulatory engineering behind the interface.

1ne. A Closed Knowledge Environment

The system shouldn’t search the open internet for product answers. Its primary sources should include materials such as:

• FDA-approved Prescribing Information
• Medication Guides and patient labeling
• FDA-approved indication and dosing information
• Approved clinical-study data
• MLR-approved promotional claims and responses
• Approved patient-support information

Every answer should be traceable to an approved source.

2wo. Fair Balance Should Be Engineered Into the System

Imagine a patient asks: “How effective is Drug X?” Simply displaying an impressive efficacy statistic could create promotional risk if material risk information isn’t presented appropriately. The system therefore needs regulatory rules governing when safety information must accompany efficacy information and how prominently that information is displayed.

Fair balance shouldn’t be something added after the AI system is developed. It should be part of the software architecture.

3hree. Build an Off-Label Firewall

Consider another question: “Does Drug X work for Condition Y?” Suppose Condition Y isn’t an FDA-approved indication. An unrestricted generative AI system might attempt to answer using published studies, medical literature or information it learned during training.

That’s precisely what a branded promotional chatbot shouldn’t be allowed to do. The system should recognize potential off-label questions and prevent promotional generation. Depending on the company’s approved procedures, the visitor could instead be given the approved indication and offered an appropriate pathway to Medical Information.

4our. Detect Potential Adverse Events

One of the most important features could be this. A user may search: “I have been on Drug X for two weeks, and now I am experiencing terrible headaches.” That is no longer just a website query. It could be a potential adverse-event report. So the search engine now needs drug safety (pharmacovigilance) detection and routing capabilities to detect these potential adverse events and route them into the company’s existing safety reporting process. Artificial intelligence may one day actually help pharmaceutical websites become better at spotting adverse events, rather than hindering them.

5ive. Don’t Let the System Become the Patient’s Doctor

There should be a distinction between providing product info and making individualized medical decisions. A chatbot should not say: “Based on your symptoms, Drug X would probably be appropriate for you.” It could list the FDA-approved indication and make it clear that only a healthcare professional can decide if the treatment is right for a particular patient. The bot can answer questions. It should not diagnose, prescribe, or make treatment decisions for an individual.

6ix. Record Everything

Every interaction should create an auditable record. The company should be able to determine:

• What question was asked?
• How was the question classified?
• What answer was provided?
• What approved source supported the answer?
• Which version of that content was active?
• Was a safety signal detected?
• Was an off-label question detected?
• Was the inquiry escalated?
• When did the interaction occur?

That audit trail could become one of the strongest arguments for a controlled AI system. Traditional websites don’t necessarily tell companies what information patients couldn’t find. An AI query system would.

The Data Could Be Almost as Valuable as the Technology

There’s another reason why pharma should embrace this idea. The queries themselves could be enlightening. What if thousands of people searched: “How long before Drug X kicks in?” Or: “Why is Drug X so expensive?” Or: “Can I take Drug X with ____ medication?” Or: “What happens if I stop taking Drug X?” The questions patients are asking can illuminate the gap between what pharma thinks they should know and what they actually want to know.

Combined with other patient feedback channels and stripped of identifying information, search query data could fuel: Patient education, website content, Medical Affairs planning, FAQs, HCP education, market research, patient support programs, and future website content. Instead of tracking only page views and time on site, pharma can track unanswered patient questions. That’s a much more meaningful digital metric.

Start With Retrieval, Not Generative AI

Pharma companies don’t need to begin with an autonomous conversational AI system. In fact, they probably shouldn’t. A first-generation system could be deliberately conservative. The AI’s job would be primarily to understand natural-language questions. The answers would come from a locked, MLR-approved content library.

Over time, companies could evaluate where carefully controlled generative capabilities provide additional value. This creates a sensible progression:

Phase 1: Natural-language search of approved information.

Phase 2: Approved conversational responses assembled from controlled content.

Phase 3: More sophisticated personalization and generative capabilities where regulatory, medical, legal, and privacy controls permit them.

The technology can evolve as the organization gains experience.

This Could Be Bigger Than a Website Feature

Many people in pharma fall into the trap of thinking about AI as just another digital channel for marketing messages. I don’t think that’s nearly far-reaching enough. Imagine a compliant pharmaceutical query engine sitting between the consumer and virtually every source of company-controlled product information. The same engine powering brand websites, patient-support programs, Medical Information portals, and HCP websites. All of those experiences sit on top of a common regulatory infrastructure determining what information can be presented, how it can be presented, and when an interaction needs to be escalated to a human. The competitive advantage will never be the large language model. LLM’s will become commodities…

The real intellectual property would be the regulatory architecture surrounding the AI:

Approved content management + MLR workflow + fair-balance rules + off-label detection + adverse-event detection + escalation + version control + auditability.

Pharma Shouldn’t Ask Whether AI Can Answer Drug Questions

That’s the wrong question. AI can already answer drug questions. The important question is:

Can pharmaceutical companies create an AI environment in which the answers are controlled, substantiated, appropriately balanced, auditable and compliant?

I believe they can, and online health seekers want it. And there’s an important distinction between allowing an AI system to generate pharmaceutical information and using AI to help consumers find approved pharma information. The latter is far more achievable.

For years, pharma companies have spent enormous amounts of money driving patients to websites and then expected those patients to navigate the information architecture the company created. AI offers an opportunity to reverse that relationship.

Don’t make patients search the website.

Let them ask the website.

If pharma can solve the regulatory architecture behind that deceptively simple query box, it could become one of the most useful applications of AI in pharmaceutical digital marketing.

Pharma companies have become so obsessed with meeting the FDA’s “fair balance” requirement that many branded websites seem designed for regulatory reviewers instead of the people actually searching for health information. The outcome could be the opposite of what pharma marketers are striving for: patients exiting before they learn enough about the therapy to have an informed discussion with their physician.

Does FDA Actually Require Fair Balance on the Homepage?

The short answer is yes—but with an important qualification. The FDA regulates Internet-based prescription-drug promotion, and prescription-drug advertisements generally must present a fair balance between information about effectiveness and information about risks. The FDA describes failure to present a fair balance of risk and benefit information as one of the common problems it identifies in prescription-drug promotion.

But the FDA does not appear to have a simple rule saying: “Every branded drug website must display the complete Important Safety Information prominently at the top of its homepage.” That is a very different proposition.

The FDA’s underlying concern is whether the promotional communication as a whole is truthful, non-misleading, and appropriately balanced. If a manufacturer makes efficacy or other benefit claims, it cannot make patients hunt through another section of the website to discover that the product carries significant risks.

The FDA’s Internet/social-media guidance explains the basic principle clearly: when benefit information is presented, corresponding risk information should be comparable in content and prominence. That is fair balance. It isn’t necessarily a mandate for bad website design.

Pharma Has Turned Regulatory Risk Into a UX Problem

Years of risk-averse decisions have blurred that line. Legal and regulatory teams naturally want to reduce risk. Marketing teams want to say what their product does. Digital teams want patients to connect. The result can be throwing every conceivable detail on the page. Websites’ homepages have become regulatory-feeling documents trying to pass as patient-engagement sites. That might guard against one perceived risk, but it introduces another: patients don’t stick around.

Consider the patient who has just received a diagnosis for a chronic condition. They Google the condition after seeing their physician. They have questions. What is this disease? What should I expect now? What treatment options will I have? How well will they work? What questions should I ask my doctor? Then they click to a pharma website and are immediately greeted with language like “serious infections,” “malignancies,” “cardiovascular events,” “suicidal thoughts,” “liver injury or death.” Sure, these are important and absolutely may need to be communicated. But the when and how is key.

FDA’s Own Research Makes This More Interesting

The FDA has even researched risk information location on branded prescription drug websites. FDA’s Office of Prescription Drug Promotion completed a study in which they analyzed various presentations of benefit and risk information on branded drug websites.

The FDA found that the location of risk information affected consumers’ risk knowledge, with better knowledge when risk information was presented on the homepage. That is a valuable insight. But it leaves open another equally valuable question: How does prominent risk information impact engagement?

Patients could gain a better understanding of risk information AND simultaneously become less likely to continue reading on the website. Both of these outcomes are possible, and they are not mutually exclusive. FDA is appropriately focused on the issue from the perspective of truthful and balanced information. Brand marketers have another obligation: real patient behavior.

We Need to Measure Abandonment

Pharma spends enormous amounts of money driving consumers to branded websites through television advertising, paid search, social media, and other channels. Yet one of the most important digital metrics may receive far less attention: Where do patients leave? Companies should be examining whether exposure to extensive safety information early in the digital journey correlates with:

• homepage abandonment;
• reduced time on site;
• lower engagement with educational content;
• fewer visits to patient-support information;
• fewer physician-discussion-guide downloads;
• lower conversion to “find a doctor” or similar tools; and
• reduced return visits.

Regulatory compliance should not prevent this research. It should encourage it. If the objective is informed decision-making, pharma needs to understand whether its websites are actually informing patients—or frightening them away before meaningful education occurs.

Fair Balance Does Not Have to Mean Terrible Communication

The industry’s error is perceiving regulatory compliance and good user experience as mutually exclusive goals. They’re not. Risk information should be visible, understandable, and accessible. But drug companies can and should test better ways to present it. Information architecture is important. Typography is important. Plain language is important.

Progressive disclosure is important. How benefit information and risk information relate to one another is important. Most of all, context is important. There’s a big difference between obfuscating important safety information and presenting it in a way consumers can actually understand. Stuffing a 400-word block of regulatory verbiage somewhere on a homepage may well check the box for someone’s interpretation of regulatory compliance. But that doesn’t mean patients will read it or, more importantly, understand it.

The Homepage Should Answer a Patient’s First Question

When a consumer lands on your pharma website, you should know the question they’re trying to answer. For the newly diagnosed patient, it could be: “What are my treatment options?” For the patient whose current treatment isn’t working: “Will this treatment work for me?” Or for someone else: “What questions should I ask my doctor?” Your website should start by answering that question while seamlessly weaving in prominent risk information. That’s a UX problem worth tackling. Instead, many pharmaceutical websites open with: “Let me tell you every horrible way this treatment could kill you.” No wonder consumers go to Google, Reddit, TikTok or patient forums when they have health questions.

Pharma Is Losing Control of the Patient Conversation

Perhaps that’s the bigger strategic issue. Patients aren’t going to stop seeking information about medications because brand websites are hard to use. They’ll just look elsewhere. And when that happens, pharma loses control of the message. Context, accuracy, and scientific rigor all go out the window. It’s pretty ironic. By trying to minimize regulatory risk and making websites ever more conservative, pharma could be driving patients to outlets that are out of reach of both the company and the FDA. After all, the FDA recognizes that millions of consumers use digital media as a source of health information. Rather than making branded websites less inviting, the industry should be striving to make them more useful. But still balanced and compliant.

A sample of a clean, well presented fair balance

The Question Pharma Should Be Asking

It shouldn’t be: “How little safety information can we put on the homepage?” But it also shouldn’t be: “How much safety information can Regulatory make us put there?” The right question is: How can we communicate benefit and risk in a way that meets FDA requirements while also inviting patients to learn more? That’s not something you can finesse with a paragraph. It requires regulatory expertise. It requires behavioral research. Usability testing. Analytics. And a much deeper understanding of the online patient journey.

Pharma websites have been optimized for years…for compliance. Now it’s time to optimize them for comprehension.

Because a perfectly compliant website that patients immediately leave isn’t doing much for anyone.

1ne. Pfizer: Billions Spent, but Where Is the Growth?

Pfizer may be the clearest example of the problem. CEO Albert Bourla deserves substantial credit for Pfizer’s COVID-era execution. But running a pharma company after a windfall may ultimately prove a more important test of leadership than managing the windfall itself.

Pfizer spent $43 billion acquiring Seagen in 2023, following other multibillion-dollar transactions intended to rebuild its pipeline. Yet in early 2026, Pfizer disclosed $4.4 billion in impairment charges associated with pipeline adjustmentswhile eliminating another six early-stage programs.

The company is also carrying approximately $60.5 billion in outstanding debt, according to its latest reporting cited by the Financial Times. Pfizer’s market capitalization recently stood around $152.5 billion—below Amgen and even Gilead—and it has become the most heavily shorted pharmaceutical company valued above $50 billion, according to S3 data reported by the FT.

Meanwhile, another asset inherited through Seagen failed a Phase 3 trial this year. Pfizer has now concluded there is no viable U.S. pathway to return Oxbryta to the market following its earlier withdrawal. The problem isn’t that every clinical trial must succeed. Drug development doesn’t work that way.

The problem is capital productivity.

When management spends tens of billions acquiring pipelines, and the resulting organization continues pruning programs, recording impairments, and searching externally for the next growth engine, shareholders should ask whether leadership has demonstrated an ability to identify valuable science before writing enormous checks.

Pfizer shares were up roughly 7.5% in 2026 as of early August, but that trailed Amgen’s approximately 25.5% gain. Pfizer doesn’t have a cost problem nearly as much as it has a credibility problem.

Leadership grade: D

2wo. Novo Nordisk: How Do You Lose a Market You Helped Create?

The deterioration at Novo Nordisk is perhaps even more remarkable. Novo helped create the modern obesity-drug market with Wegovy and dominated diabetes with Ozempic. Demand was extraordinary. Management had one of the most valuable pharmaceutical franchises ever created. And then Eli Lilly arrived.

Novo’s shares have fallen approximately 70% from their June 2024 peak, while competitive pressure, pricing pressure and questions about the company’s pipeline have transformed what looked like an almost unassailable growth story.

Novo now expects adjusted 2026 sales growth of between 0% and negative 6% at constant exchange rates, an improvement from its previous expectation of negative 4% to negative 12%. Adjusted operating-profit growth carries the same 0% to negative 6% range. Think about that.

This is happening while obesity remains one of the largest pharmaceutical growth markets in the world. Leadership’s job isn’t merely to invent a blockbuster. It is to defend the franchise once competitors recognize how valuable the market has become.

Novo’s problems demonstrate one of pharma’s oldest management mistakes: confusing first-mover advantage with sustainable competitive advantage. The company still possesses extraordinary assets and could recover. But investors should ask why an organization with Novo’s scientific expertise, market lead, and cash-generating capability allowed its competitive position to deteriorate this dramatically.

Leadership grade: D+

3hree. Sanofi: When the Board Replaces the CEO, the Verdict Is Pretty Clear

The case against Sanofi’s previous leadership doesn’t require much interpretation. The board effectively delivered the verdict itself. In February, Sanofi announced that CEO Paul Hudson’s director mandate would not be renewed and appointed Belén Garijo as his successor. The company explicitly said Garijo would bring greater rigor to execution and strengthen R&D productivity, governance and innovation.

Corporate language rarely gets much clearer. Hudson had attempted to transform Sanofi into a more innovation-driven pharmaceutical company, but the pipeline repeatedly disappointed.

Among the setbacks were failures involving amlitelimab in asthma, balinatunfib in psoriasis and itepekimab in COPD. Sanofi ultimately abandoned amlitelimab in atopic dermatitis despite the drug once being viewed as a potential $5 billion opportunity.

The strategic problem is particularly serious because Sanofi remains heavily dependent on Dupixent.

Hudson himself acknowledged the central execution issue before leaving: the transformation simply had not occurred as quickly as expected. Garijo shouldn’t be placed on a worst-CEO list after only a few months running the company. In fact, her willingness to aggressively review the late-stage pipeline may be exactly what Sanofi needs.

But Sanofi absolutely belongs on a list of 2026’s clearest examples of failed pharmaceutical leadership, because the board decided a leadership reset was necessary.

Former leadership grade: F
New leadership: Too early to grade

4our. Takeda: When Transformation Requires 4,500 Job Cuts, Ask What Went Wrong Before the Transformation

Takeda is embarking on another enormous restructuring. The company expects approximately 4,500 jobs to disappear during fiscal 2026, with total announced cuts pushing the number of affected positions even higher. Takeda’s transformation is intended eventually to generate more than ¥200 billion in annualized gross savings by FY2028.

Cost cutting itself isn’t evidence of bad leadership. Sometimes it is exactly what responsible management should do. But repeated large-scale restructuring raises another question: Why did the organization become so inefficient in the first place?

Takeda reported FY2025 revenue down 1.7%, with core revenue declining 2.6% at constant exchange rates. For FY2026, management expects a low-single-digit decline in core revenue, a 5%-8% decline in core operating profit and a mid-teens percentage decline in core EPS at constant exchange rates.

There is an important counterargument. Takeda has reported encouraging Phase 3 results for oveporexton, rusfertide and zasocitinib. If those medicines become successful launches, the restructuring may eventually look less like desperation and more like disciplined preparation for the company’s next growth cycle.

That’s why Takeda ranks below Pfizer, Novo and Sanofi on my list.

But eliminating thousands of positions while earnings decline isn’t evidence of a healthy pharmaceutical operating model. It shows the model needed major repair.

Leadership grade: D-

5ive. Viatris: Restructuring Cannot Become the Business Strategy

Viatris illustrates another recurring pharma leadership problem: companies that seem perpetually to be restructuring themselves. The company announced a restructuring that could eliminate up to 3,000 jobs—roughly 10% of its workforce—over three years. Yet the operating picture isn’t uniformly bad.

Second-quarter 2026 revenue reached approximately $3.8 billion, up 5% reported and 3.5% operationally. Adjusted EBITDA reached approximately $1.2 billion, up 8% operationally. The company also returned roughly $550 million to shareholders and reduced its gross leverage ratio to 2.9x.

Those aren’t disastrous numbers. The leadership question is more fundamental. What exactly is Viatris going to become?

Selling assets, reducing debt, repurchasing shares and cutting expenses can improve financial metrics. But pharmaceutical companies ultimately create extraordinary shareholder value through differentiated medicines, successful launches and sustainable intellectual property.

Financial engineering can buy management time. It cannot replace innovation.

Leadership grade: C-

6ix. Geron. One Drug Wonder That Keeps Rearranging Chairs On A Sinking Ship

Few companies demonstrate the difference between scientific perseverance and good corporate governance more than Geron Corporation. Geron accomplished something truly historic this year when it won FDA approval for RYTELO (imetelstat) in 2024. It took decades of effort to bring RYTELO to market and convert Geron from a development-stage biotech into a commercial pharmaceutical company. However, regulatory approval was supposed to be the commencement of value creation—not the culmination of the strategy. What transpired commercially has highlighted execution gaps.

Although third quarter 20 25 product revenue from RYTELO totaled $47.2 million, demand for the drug actually decreased 3% from the prior quarter. Management didn’t hide behind bullish platitudes; they admitted there was room to increase brand awareness and improve how they communicate the drug’s clinical value. For a company that overwhelmingly relies on one commercial product, that is a stunning admission. Management has now responded in kind. John Scarlett, the company’s CEO for over two decades, stepped down in 2025.

Geron’s been under interim leadership ever since Peter Merighi became president and CEO in August. Geron has also made changes to other parts of the executive team. Enter the restructuring. In December 2025, Geron’s board announced a workforce reduction that will cut roughly one-third of the company’s employees (~260). Geron will incur approximately $17 million in restructuring charges. Consider the timing. It takes a company decades to get its first major drug approved, and less than two years after launch, it realizes that one-third of its workforce is unnecessary. That says as much about organizational design and commercialization planning prior to RYTELO’s approval as it does about current events.

Leadership grade: D-

The Bigger Problem: Pharma Leadership Has Become Too Financial

The common denominator across these organizations is not failed leadership. These are experienced executives steering exceptionally complex global businesses. The management model is broken.

Time and again, Big Pharma approaches strategic challenges with reflexive corporate consolations: Acquire. Restructure. Lay off employees. Re-org R&D. Declare productivity gains. Rinse. Repeat. Here are some numbers. During the first six months of 2026, BioSpace tracked 14,427 biopharma employees impacted by layoffs. Three organizations – Takeda, Viatris and BioNTech – accounted for roughly 9,603 of those jobs. Layoffs are not an innovation strategy. Neither are acquisitions.

Nor is mindlessly investing more billions in R&D. The focus should be on return on innovation investment. How much sustainable profit and shareholder value does every additional billion dollars of investment in internal R&D and external business development generate?

My 2026 Pharma Leadership Ranking

1. Pfizer — D: Extraordinary capital deployment without a sufficiently convincing post-COVID growth engine.

2. Novo Nordisk — D+: Allowed a commanding obesity-market advantage to deteriorate while competitors accelerated.

3. Sanofi’s former leadership — F: Pipeline execution became so serious that the board replaced the CEO.

4. Takeda — D-: Thousands of layoffs and another transformation program demonstrate how much organizational repair is required, although its late-stage pipeline provides a credible path to redemption.

5. Viatris — C-: Improving financial metrics haven’t yet answered the larger question of how the company creates differentiated long-term pharmaceutical growth.

6. Geron – D-: One drug wonder with constant management changes.

The companies I would be most concerned about are Pfizer and Novo Nordisk, but for very different reasons. Novo’s problem is competitive execution. Pfizer’s problem is capital allocation.

And capital allocation may ultimately be the more damaging failure. A pharma CEO has access to something very few executives ever receive: billions of dollars that can be deployed toward scientific innovation capable of generating decades of economic value.

If management repeatedly spends that capital without producing an increasingly valuable pipeline, boards shouldn’t automatically authorize another acquisition or another restructuring. They should start asking whether they have the right people deciding where the next billion dollars goes.

In Big Pharma, failed drugs are inevitable. Failed strategy isn’t.

Pharma websites have become eye candy for brand teams. You name it. They stuff our sites with marketing copy to fill an entire sales aid. Months of agency effort. Endless brand meetings. Rounds of medical, legal, and regulatory review. Significant development costs. And after all of this effort, we may end up with a website our customer didn’t actually want in the first place.

We recently tested a completely new concept: What if we told you a Pharma product website could just get to the point? What if we built a one-page website focused on delivering key drug information in a clear, concise format that’s easy to understand? The results were surprising. And overwhelmingly positive.

The Experiment Was Almost Embarrassingly Simple

Rather than forcing visitors to hunt through a traditional pharma website, we put the information they were most likely seeking right in front of them. The product name. The intended audience. Key clinical outcomes. Mechanism of action. Route of administration. Key safety information. Where to go for more information. That’s about it. No internet treasure hunt. No clicking through five menus to find a piece of clinical data. No making people understand the internal framework of the brand team’s content plan. Just handing them what they wanted, fast.

The positive response raises an uncomfortable question:

Are pharma companies designing websites for customers—or for themselves?

The $500,000 Website Nobody Asked For

The digital development process at Pharma has a foreseeable problem. Everyone wants something on the website. Brand wants messaging. Medical wants clinical content. Market access wants reimbursement information. Patient services wants support information. Corporate communications has requirements. Legal and regulatory have mandatory content. The agency has recommendations. Oh yeah, someone wants a video. Someone else wants an interactive MOA. And another stakeholder wants a whole new section just because a competitor has one.

Somewhere along the way the website morphs into a monument to organizational compromise. What it doesn’t necessarily become is easier to use. That’s a key distinction. You can meet every internal stakeholder’s needs and completely fail the customer.

We May Be Measuring the Wrong Things

Pharma faces another issue when determining how to measure digital performance. Page views. Sessions. Time on site. Pages per session. Bounce rate. Downloads. These can all be helpful metrics to track, but they can also create false senses of success. Take “time on site,” for example. Six minutes spent on a drug website by a visitor sounds like stellar engagement. But it also could mean: “I couldn’t find what I was looking for.” Five pages per session may mean great engagement. It may also mean it took five pages to provide an answer that should have taken 30 seconds. This is why we approach measurement differently on a one-page pharmaceutical website.

The Metrics Pharma Should Actually Be Watching

If the purpose of a drug website is to communicate information effectively, then we should measure whether that communication actually occurred. I would focus on six areas.

1ne. Time to information

How quickly can visitors find the information they came for? Clinical results shouldn’t require a treasure hunt. Dosing, indication, patient support, and safety information shouldn’t require a treasure hunt either. Measure the time required to reach high-value information. Then try to reduce it.

2wo. Message comprehension

After interacting with the website, can visitors correctly identify the product’s indication, major clinical benefit, dosing, and important safety information? That’s substantially more meaningful than counting page views.

3hree. Message recall: What do visitors remember five minutes later? What do they remember a day later? If a website generates thousands of visits but nobody remembers the core product message, what exactly did the digital investment accomplish?

4our. Task completion

Give visitors specific objectives. Find the dosing information. Locate the primary clinical endpoint. Determine whether the product may be appropriate for a particular patient profile. Find patient support information. Then measure how many people successfully complete those tasks—and how long it takes.

5ive. Scroll and interaction behavior

On a one-page website, scroll depth becomes particularly informative. Where do people stop? Which sections receive the most attention? What information gets skipped? Which calls to action generate engagement? That behavior can help identify what customers actually value rather than what the brand team assumes they value.

6ix. Customer preference

This may be the most revealing metric of all. Show customers the conventional pharma website and the simplified one-page experience. Then ask: Which would you rather use? The answer might make some brand teams uncomfortable. That’s precisely why the experiment is worth conducting.

We did that, and the vast majority preferred a one-page simplified experience. However, please note that some drugs, like cancer drugs, need a lot of information for newly diagnosed patients. However, since trust in pharma is so low, will newly diagnosed cancer patients trust their product websites? In that case, a special section for newly diagnosed is essential.

Complexity Is Not a Strategy

There’s an inclination in pharma marketing that complex products must have complex websites to match. I don’t think so. Complex science should be communicated as clearly as possible. Just because your drug discovery process is strategic doesn’t mean your website should have 14 navigational options. Just because your work is innovative doesn’t mean you should make physicians click through 5 pages just to read trial results. And just because you have the resources to create hundreds of pieces of content doesn’t mean you’re being customer-centric by making your site home to every last bit of it. Sometimes being sophisticated is knowing when less is more.

Think About the Physician on the Other Side of the Screen

This is where pharma needs a reality check. Physicians don’t have all day to discover what’s inside our digital ecosystems. They have patients to see. So if an HCP is looking for information on how your drug performed versus placebo in a trial, make it easy to find. If they want dosing information, don’t make them hunt for it. If they want to know the mechanism of action, speak clearly. If they want the Prescribing Information, bring it to them. Don’t make them dig for it. Every extra click creates friction. And friction creates another chance for someone to bounce.

One Page Doesn’t Mean Less Science

This is important. Bear with me. I’m not saying we shouldn’t have clinical info. I’m saying we need to establish an information hierarchy. Huge difference. We can design a one-page experience that tells the product story while providing links to clinical studies, PIs, publications, and more for those who want to dig deeper. Think of that homepage as the drug’s executive summary. Visitors should find everything they need to know about the product there. If they want more, they can click through to learn more. It’s very different than forcing every visitor to explore the entire site.

Run the Test

Every pharma brand team should consider a relatively simple experiment. Take the existing branded website. Then build a one-page alternative. Give both versions to comparable groups of HCPs or patients.

Measure:

• Time to find key information
• Task completion
• Comprehension
• Message recall
• Scroll depth
• Clinical-data engagement
• Prescribing-information access
• Patient-support engagement
• Overall satisfaction
• Preference between experiences

Then ask yourself one last question: Which site would you actually use again? Let the customer be the judge. Not your agency. Not your brand team. Not your digital center of excellence. The customer.

The Bigger Issue Isn’t Website Design

But wait, there’s more than just websites at stake here. Pharma has become addicted to content. Agencies are churning out content. Brand teams are asking for content. Review committees are approving content. Digital teams are publishing content. Even our analytics teams measure interactions with content. The beast just keeps churning out more.

Maybe before we build the next web page, video, interactive tool or PDF, we should ask ourselves a simpler question: Do customers really need this stuff? Pharma digital marketing needs discipline. Not more content. Not more tech. Not more re-designs. It needs focus. Simplification.

Our little experiment with a one-page drug website has shown that when you cut through the organizational complexity and simply give people what they’re looking for, some very nice things happen: They love it. They understand it. And they can find what they need, much faster. Maybe the future of pharma digital is not some multimillion-dollar digital ecosystem. Maybe it’s just one really good page. Make your point. Your customers will thank you for it. One particularly powerful follow-on article would be “The Pharma Website Scorecard”—a 10-point metric that brand leaders can use to grade their existing product websites on speed, comprehension, usability, accessibility of clinical data, mobile experience, and conversion.

This is a sample of a one-page summary that gives online health seekers essential information