AI will kill pharmaceutical advertising (and how the reaction from big pharma might disrupt the whole healthcare ecosystem)

Disclaimer: This is my current personal view, not of any company or institution I am affiliated with.

Summary: In the long term drug differentiation will be minimal but, despite this, marketing will be less relevant as AI makes clear objective comparisons possible. The increased clarity will benefit patients and potentially catalyze a transformation towards personalized care. Vertical integration of pharma and healthcare services is a new frontier that might act as a wild card disrupting the space (and might revive advertising).

The business of discovering and selling medicines is changing fast and while I can’t predict what it will look like at the end of 2024 I see a very clear direction it will go towards based on macro trends that are affecting it globally.

First the ongoing trends that will continue:

Price pressure will continue leading to a value focus:

From Medicare price negotiation, to NICE assessments in the UK, and many other initiatives globally, payers will continue to evaluate more rigorously the value proposition of medicines. This is particularly relevant for smaller companies and the startup/VC world because it is no longer enough to get regulatory approval (or have a promising path to it) to have a good exit. Clinical trials still should produce a strong package of evidence for regulators but, if done well, also a strong package of evidence for payers.

Differentiation is disappearing:

While small molecules still have a big impact on patients most new technologies are moving into much less differentiated modalities. Almost everyone can do an antibody now (the recent Antibody-Drug-Conjugate M&A wave takes it one level further), and the same is true for many other platforms. It’s also becoming easier to go after the same target with several technologies once it is validated so competition does not need to come from the same therapeutic modalities. Current vaccine markets exemplify this, and biosimilars make it self-explanatory. In this world IP offers some protection but incessant competitive pressure should always be expected. Evidently there will be exceptions but the majority of medicines should expect a very packed competitive landscape.

Unencumbered information flow is killing advertising (and likely marketing):

In a world where there is less differentiation across competitors you would expect marketing to be a dominant force. From advertising to generate brand loyalty and demand to distribution to get those customers that don’t care, the market should be very similar to the oligopolies in the Consumer Packaged Goods space.

However, when it comes to medicines I don’t believe this will be the case in the long term. Medicines have a big impact on patients and this drives both patients and health care professionals to seek to be truly informed about treatments. Historically being informed was complex, driving the big pharmaceutical advertising machinery we know today. But the digital transformation has been democratizing knowledge and streamlining information flow. Combined with a continued societal effort to train health care professionals to make the best decisions for patients there is very little, if any, room for brand preference.

Distribution is having a very significant role right now (as an example CVS has an outsized impact when they carry one competitor but not another) but I suspect regulators will erode any of these advantages as the market evolves.

We are not there yet, and it will be a complex transition full of business opportunities in the middle, but the pandemic accelerated the move as evidenced in the current digital focus of pharmaceutical marketing.

How AI combines with the other trends:

AI will impact drug development significantly but I don’t think it will disrupt the space. AI’s impact will be to improve the odds of success during the process of drug development, and to allow better monitoring for value demonstration (both before and after market entry). This impact is not small, but I don’t see it differentiating significantly across companies, just as when computers came into play. AI will also lower the cost of entry into the space so it will have a very positive impact on innovation by empowering startups beyond what they can do today. More innovation is definitively a positive but I don’t think in itself it will change the market structure; perhaps what more innovation will do is accelerate the path towards undifferentiated therapies. AI, specifically Large Language Models, will accelerate democratization of information as it is already much easier for patients to understand complex scientific publications. I see AI having the most disruptive impact on patient care (more on that later).

So where do those trends take us:

Drug development:

Gathering value evidence is key. Value is not new, value based care proponents have been advocating for it for a long time, and it is slowly getting there. Whether it will be a few years or many before it’s there is something I don’t know, but it’s clear that things are going in that direction. Big pharma has a big advantage when it comes to demonstrating value due to scale and balance sheet but, as in any large matrix organization, the interplay between disparate groups is not as streamlined as one would expect. There might be an optimal company size for development around mid/large biotech where the budget, knowledge, and team structure favor this holistic thinking. Startups need to think carefully about payers as their chances of a successful exit depend on obtaining the right data on value. AI will certainly help by lowering the costs of obtaining and processing the information needed to demonstrate value.

Commercialization:

Advertising expenses will eventually disappear as their impact decreases. It’s easy to picture an end state where there is an AI “personalized” formulary based on patient health records, coverage options, personal preferences, and risk tolerance, such that drug makers only need to ensure prescriptions are distributed. It’s a long transition until then so, in the meantime, we will see advertising and distribution evolve fast to adapt to the changing environment. A very strong focus on branding should be evident as differentiation decreases. Another emphasis area of the transition will be on removing friction from the healthcare system: helping patients get a therapy that clearly is the best value proposition for them but that they are not getting because they are getting lost in the process. This “friction removal” is the argument Lilly stated for its recent foray into prescription and delivery of some of its products.

Investment:

Cheaper costs of development due to AI and other technologies will lead to more innovation. While it’s been argued that price pressures reduce innovation, the evidence I’ve seen published does not reflect this. If there is a negative impact it is projected to be small (the US Congressional Budget Office calculated 13 less drugs in 30 years due to Medicare price negotiation), and I believe the increases in productivity driven by AI will far outweigh a potentially diminished level of investment.

The hypercompetitive market will drive a push for scale to obtain a cost advantage (and/or risk pooling), and platform or other large concept plays will be attractive. The Flagship Pioneering model comes to mind as a good example. At the same time, the value of a truly differentiated therapy will be higher in a space where there are less of them.

All of these factors sort of pull in opposite directions so I don’t see big changes, just the trends I described. The only exception is for small companies with single or few undifferentiated assets going into an already crowded market. These companies exist hoping for a “small fraction” of a proven existing large market (the classic 5-10% of market arbitrary projection). I don’t think they’ll get investor attention any more as that “small fraction” just shrinks and shrinks given more crowded spaces and higher development costs to obtain value data (vs just medical data).

Patients win:

Increased competition of similar therapies focused on value will lead to a better outcome for patients over time. However, the big change I see is in AI having a truly transformative role in patient care. Currently healthcare professionals are incredibly busy (with strong personnel challenges in the space due to undercompensation and poor quality of life). It is hard for marketers to get their attention, but the interaction still happens because health care professionals want to know what is best for patients. The moment there is a way for healthcare professionals to have ALL information on therapies available easily, from a trusted source, pharmaceutical advertising is dead. I believe it will be a short time before a solution capitalizing on new AI capabilities to compile and navigate all this information becomes available.

This innovation might come from any of many possible actors but I see a strong incentive for payers to develop it as it will allow them to maximize value by ensuring each patient receives the best therapy. The process efficiencies that AI will bring are also likely to allow deep personalization, as monitoring of outcomes becomes more efficient and integration of health records stops being a challenge. Such a system will also eliminate the complexities of dealing with drug-drug interactions. Clearly there will be a lot of bias and mistakes to be corrected along the way, AI can only perform based on the data available and the instructions given. But I’m optimistic that regulators, researchers, and the press will be enough of a force to remove such mistakes so that we end up with a much better patient experience that is also cheaper.

The wild card of integration:

Finally, this is the one trend where I don’t see a clear outcome. From CVS partnering to manufacture biosimilars to Lilly offering prescriptions and distributing its medicines, we have just started to see a new wave of vertical integration. I don’t think this has happened since the origins of pharma, when doctors combined ingredients to develop their own medicines (I don’t know all the relevant history so would not be surprised to learn it was tried at some other point). I envision most major players joining shortly on this trend, but I don’t know where it will lead.

It might not change much for the status quo, simply being limited to a few instances where it’s purely a logical business move for some high revenue assets. In this case pharma and biotech will end up focused almost exclusively on drug discovery and development. Or it might be truly disruptive of the whole industry. After all, medicines are a small fraction of the overall cost of healthcare. For well capitalized pharma this creates a strong incentive to capture directly some of that profit down the line (that better value therapies generate for current payers and providers). This integration could potentially “revive” marketing; it is easy to imagine a duopoly of obesity treatment clinics by Lilly and Novo Nordisk where the medical offering is similar enough it becomes about brand and auxiliary services. I haven’t seen enough to try to speculate as to how the rest of the ecosystem would react. I also don’t think there’s clear evidence of the side towards which it will trend but this integration is definitively a wildcard worth paying attention to.

I don’t know when we will reach this end-state I’ve described, or if the vertical integration will lead to a very different one. But I know it will definitely not be an easy or linear transformation, and I am excited to be part of the industry at this time.

As I said in the beginning, this is just my current view, it will change as I learn more. I would definitely like to know more about anything I missed so as my friend Dr. Amine ZORGANI says often: happy to hear your thoughts and stay positive!