Though a relative newcomer to the life sciences industry, modular content is transforming the pharma marketing landscape, and few understand its potential better than Joseph Tew.
With over 10 years of experience as a content manager and strategist at firms including Merck & Co and Bristol Myers Squibb (BMS), Tew is an expert in creating standards for modular content and platform integrations, training content managers, and spear-heading omnichannel projects.
Tew spoke to Shaman about the benefits of modular content, typical challenges and pitfalls for pharma marketing teams looking to adopt it, and how he sees the future of modular, omnichannel content going forward.
Could you tell us about your background and connection to modular content?
JT: I come from the nonprofit content creation and content library world. I wound up going into pharma in 2016 as part of a modular content project. Merck & Co was going to modularise and automate all of its content creation, which involved digitising a content library. Basically, the main premise was to break the core content down to its bare essence, its smallest claim possible, and have two processes: the US and ex-US process. Because there were different regulatory standards for both.
We were very successful at Merck; our process saved $19 million over two years in Medical, Legal, and Regulatory (MLR) review and agency costs.
When I moved on to BMS, they wanted to expand upon that by automating the full content lifecycle from ideation to deployment. My role was to determine what modular content was in order to create model contract libraries and educate brands and agencies of record on how to approach content from a modular perspective.
Modular content is associated with various benefits. In the real world, which benefit is clearest to you?
JT: One benefit would be the removal of specialisation for content creation. You don’t need to be a graphic designer or a copywriter to use the tools to create approved content. It comes down to cost saving too. Let's say in the US it would cost $30,000 to create one home office email, including agency and production costs. This means your content steward would be making $60,000 a year creating two emails!
Normally, content creation [in pharma] is linear. You create your core content, you create your first derivative piece, and you move down the line to other tactics. Modular content allows the user to create content once. You can sit down with your toolset, home office email, rep-triggered email, third party tactics, by simply dragging and dropping content into its appropriate spaces within a template.
At Merck, [modular content] took the MLR review time down from three months to a week. We saved a lot of money and a lot of tactics which allowed for personalization to target the HCP and consumer.
Modular content took the MLR review time down from three months to a week.
Another issue that modular content solved is that it removed deviations. With modular content, the references are always attached to the claim, which stops you having to pull back content because it’s referenced incorrectly. There's no human error.
What are the typical misconceptions for teams using modular content for the first time?
JT: Part of the problem of modular content is understanding what it is and what it can or can't do. When we first introduced modular content at BMS, we found there were five or six different interpretations of it. Everyone has an opinion. That was the first pitfall—getting everybody to agree. I found that most of the time, modular content was seen by marketing teams as variable content. I said yes, but it's still created new; it's not part of a content library. You can pick and choose content to reduce your costs.
Business analytics [teams] saw modular content as a way to determine how many prescriptions are written by interaction. I said, no, not even close yet! Because we're not deploying modules to a channel—that's the full email! They saw modules as being deployed to a channel where you can measure its efficacy among the HCPs. But this comes later. The first part is building the process, the library, and the tools. That’s what people didn’t want to hear. They want to be hands off as much as possible and go to their agency. But your agency is charging you three times the cost!
Modular content can work. You just need to be focused and be on the same page. Define upfront what modular content is, define what you want to do to find your processes and have strong institutional backing to push people along.
How do you deal with friction between creative and MLR teams?
JT: This is a true story; it's still funny to me. One brand wanted me to build an email with a splash of yellow in the background. I told them they weren’t going to get that because the tools don’t allow you to create a splash of yellow unless you use inDesign or Photoshop, or something. I said I could give them a yellow box. And at first, they said no. But then I came back and did it really quickly and they went along with it. It's a trade-off: you get speed, you get personalisation, you get content, and you save money, but you're not going to get the fancy schmancy design elements.
The design is not going to overtake the message or the claim. We're not high fashion, we're not Prada, we're not Louis Vuitton. We're not selling something that’s sexy and beautiful—we're selling a drug that saves people's lives.
We're not selling something that’s sexy and beautiful – we're selling a drug that saves people's lives.
HCPs are not going to be wowed by the look of an email. They want to see charts with clear information about what a drug does. They don't want to hear about all the stuff around it. It's not a TV commercial. It's about safety.
For which use cases have you seen modular content add value? How about its limitations?
JT: Modular content works best in Europe because you don't need to have Important Safety Information (ISI) embedded in the content.
In Ireland, for example, this is just a box of content with a link out to the ISI. It’s beautiful. Everything is embedded. Here's your claim, here are your warnings and cautions, here's your black box if you have it.
There's always the risk in the US that if you build an email and simply decide to change a header into a subheader, it can screw up the whole email and MLR will send it back. This is why, at Merck, there were people who seriously didn't want to try modular content in the US, because they weren't getting the 100% efficiency we were seeing in Europe.
Shaman and every [modular content] tool like it is important. Without one, I don't think you'll be successful.
However, I think content authoring tools are central to success. You want someone who can simply go in and know how to drag and drop content, right?
Shaman and every [modular content] tool like it is important. Without one, I don't think you'll be successful.
How do you define personalised content?
JT: I had an interesting conversation with someone at a pretty big company. They said the problem with personalisation is that you’re targeting HCPs and physicians. You're in your office, and your computer is open. You don’t know who’s looking. Could a patient see? Could a nurse see? Could the wrong person see it? That was one of the concerns about personalisation. I think it went too far. For me, content personalisation is about channel preference. I don't want to see an email on my phone. I’d rather get a text message that can link out to the website.
Will pharma ever keep pace with other industries when it comes to modular content, or is it always going to be playing catch-up because of regulations?
JT: Right now there are mass layoffs in the US and pharma is losing money. We need to be more efficient, lean on these new processes, and basically wipe the slate clean and reconfigure how we work. Pharma is behind only because of its own inability to break away from old processes.
Pharma is behind only because of its own inability to break away from old processes.
One company, for instance, wanted to use the Netflix model with modular content. They wanted to abbreviate descriptions on your phone. But you can’t do that. Once you abbreviate something, you change the claim, and that’s when it gets dinged by the FDA. You need to approach it from a pharma perspective; it's not Netflix.
Where do you see the future of modular content heading in pharma, considering advances in AI?
JT: AI is not there yet. I think it has its value, but there's so much risk. If I'm an adult and AI sends me an email for a kids toy—who cares? But if I get a [pharma] email and the brand name is spelled wrong, well…
We need to be realistic. You can push the envelope, but not where you put people's lives at risk.
It’s hard to talk about because people are on board with AI, like it’s the cure-all. Yes, it can help you. One company wants to fully automate MLR review, it’s called MLR scoring, AI would scan the tactic, and say by percentage wise, will it be deployed or go to MLR. But I do that with my eyes! I know whether it’s derivative. So why do I need a machine to look at it?
We need to be realistic. You can push the envelope, but not where you put people's lives at risk.
Do you see alternatives for modular content, workflows, or methods that can accomplish the same benefits but are perhaps more flexible and easier to implement?
JT: I saw a tool that was developed at Merck built and run by EPAM Corporation. It was called an Assembler. But the way it worked, it was almost AI.
It was a JSON package that gave an either/or option. So, let's say you had a headline piece of content tag. It would know: “I'm an email. I must be 18 pixels. I'm an Interactive Visual Aid (IVA).
I must be 24 pixels bold. I'm gonna be in a mobile app. I must be nine pixels.”
That's how the code was written. So when dragged and dropped the content, it would snap into its appropriate place.
It was fake modular because there were no modules at all. Basically everything was a single piece of content packages as a “module”. I liked it because you could do whatever you wanted with it. It was a desktop app that became a web app so you could go into the website and just do it from there.
It’s incredible. But it probably costs a million dollars to develop it. Nobody wants to do that. People want out of the box [functionality]. But whoever can figure that out at scale and sell that, it would be something that's really valuable to Pharma—if they can show them that it’s possible without having to develop it in house.
