[Music] Hi everybody. Can you all hear me okay? All right. My name is Duke Marr and I run strategy for a little company named Corra. Proud partner of Adobe for 20 plus years that was actually acquired by Publicis Sapient this year, so we're very happy to be here today underneath the Sapient banner, in their booth. So it's great to be here. I'd like to introduce my client and friend, Nathan Voss. Please say hi. [Nathan Voss] Hello everyone. As Duke said, my name is Nathan Voss. I work for a company called Boehringer Ingelheim. Yeah, that's easy to say.

It's a Germain company. I've heard it pronounced many different ways, Bear-ringer, Bo-ringer, but it's Boehringer. If you take nothing else away from today's discussion, which is a bit like a fireside chat, if you know how to say Boehringer, we will all be very happy. Today we're going to talk about a few lessons from the work we've been doing together over the last few years, as Boehringer has been on a journey from a monolithic type of structure with their B2B commerce systems to a more composable, or microservices based approach.

And so we're going to cover that history, the different phases, starting with just a couple of markets and then really going global, and then try and focus in a little bit on some of the lessons that they've learned in the business and technical benefits they've gotten from that approach together. -Does that sound good? -Sounds great. All right. Let's talk a little bit first about who Boehringer is if anyone is unfamiliar. Can you just talk to us a little bit, Nathan, about what you guys do, how big you are. Sure. Boehringer is one of the leading pharmaceutical companies in the world. There are three different groups or organizations within Boehringer. There's the human pharma side, there's the animal health side, and then there's a biopharmaceutical division. These numbers that we're showing up here from 2022, they haven't released the 2023 numbers yet, though it should be here any day now, probably tomorrow they'll be released would be my guess. The animal health side, as you can see, is 4.6 billion euros for 22. I think the human pharma side was somewhere around 18 1/2 billion euros. Just a small little family. That's a great point. It is family owned, so it's completely private. No shareholders other than the family.

Talk to us about some of the brands that you guys have that the audience may know. Sure. As I said, we're both on the human pharma side as well as the animal health side. Up across the top here we have some of the products that we make on the animal health side. Frontline, NexGard, Heartgard. And then on the large animal side you'll see Ivomec. That's a very popular cattle product. And then across the bottom we have a number of human pharma products that some of you might have seen or even used. Awesome. Let's go back to when we first got started together which was, I believe, January or so of 2020. It may be even closer to March. Why did Boehringer choose Corra to be your technology partner back at that point in time? Before I get to that question, let me back up just a little bit. Sure. In the late 20-teens, we had two systems, two disparate systems. We had a loyalty platform and we had an ecommerce platform. And we used all of the duct tape and all of the band-aids we could find to put those two things together. And it was just as much as we could possibly do. It just wasn't able to support our business strategy. So we knew we needed to go to something more modern, something with flexible, something scalable. We ended up choosing Adobe Commerce as our product. But, since we didn't know anything about Adobe Commerce, we needed somebody to help us with that. That's when we reached out to Corra and used their services, and they've been a great partner ever since. One of the things that we liked about you guys is that you are a partner with Adobe, you had several certified developers as well. That was something that we looked for because we didn't have any expertise. We needed somebody that had that expertise.

And just for kicks and giggles, can you remind me what the loyalty and e-comm platforms were previously built on top of? Do you remember? Yes. The loyalty platform was built on SharePoint and the ecommerce platform was built on a different tool called Insight. Those are both gone now from our repertoire of systems. Successfully deprecated. That's correct. Yes. All right, as we dive into this, just real quick, show of hands, I'm curious, is anyone in this room at all confused by terms like composable, MACH, monolithic? Any of these terms confusing or is it all completely clear to everyone? I see a few hands. All right. Well, I'm not going to tell you it will all be clear by the end of this session, I'm not going to promise that, but I do hope we will have a sense of the decision Boehringer has made. So just a quick refresher for those like me who kind of struggle to keep these things straight. A monolithic platform, your front end and your back end is completely coupled. It is one system. The system that your customers experience when they're placing orders is completely tied to all of your back end commercial systems. So, product data, customer data, order data, they all flow through one thing. At the other extreme of this spectrum is a fully composable solution where all of those individual components are itemized into individual APIs. The solution that's kind of in the middle is when you build a composable approach, or what I would maybe think of as a microservices based approach on top of a monolithic platform, and that is the road we eventually got Boehringer Ingelheim to take with us together. -Did I get that right, Nathan? -That is correct.

Let's talk about phase one. We started with just two markets, B2B transactional website for markets in the US and France. Can you just speak to us a little bit about what this site was about and who the customers were. Sure. So as I mentioned before, we had these two legacy systems, two disparate systems that weren't working anymore. So we needed to overhaul those and we were bringing in Adobe Commerce. What we wanted to do is have a unified loyalty and commerce platform for our customers. We are a B2B company. Our target is primarily, on the animal health side, it's the veterinarians, it's the clinic's staff, and to some extent the distributers. We'll get to that more later. And so what we needed to do is build something that would work for those vets and their staff. So, the people, just to put a scenario on it, you're basically dealing with an office assistant or an office manager in a veterinary office who is putting in next month's supply of Nexguard for example. Exactly.

The animal health industry, there's all kinds of veterinarians. We've got the small mom and pop, where it's the veterinarian who does the marketing, who does the books, who treats the animals, to these large corporate accounts where they've got a division of purchasers and they've got their own sales folks, and they've got an army of veterinarians across the entire US. So, we had to span that bridge of customer types and segments. Okay. You spoke a little bit before about why you chose Adobe Commerce. Can you just elaborate a little bit on some of the feature sets. Whether it was the built-in CMS or other corrects that came with the platform. Adobe Commerce came with a number of built-in tools and functions and features that we were looking for. We knew we were going to take advantage of that because we didn't have any of that before in the legacy systems. And, you mentioned the CMS. That was something that was very important to us because we didn't want to have a separate CMS that we'd also have to integrate into this Commerce platform. So, having that built-in CMS was important for us. The scalability, the flexibility, the adaptability, the ability to work with APIs, that was all critical for us.

So, let's talk about some of the challenges. Like I said, we got started on this project, the pandemic was just hitting. What were some of the biggest things that we faced as we got started? Well, the pandemic was huge. We had intended to all come together and have a number of workshops to work through the requirements. That didn't happen. Everything was done remotely. But we still made it to work. Some of the challenges, on top of that, were that as you start to expand globally, you start to run into different kinds of rules, different regulations, different laws, different requirements. So, for this approach where we were targeting the US and France, we had to make sure we were able to cover both sets of requirements for both markets. Multi-currency, we were going to have different languages, different tax rules, different product selections that the customer could see, and just the pack shots and how everything was going to work together. It was incredibly difficult. Am I remembering correctly, there's also different regulations in France about how vets log in and use the systems and be credentialed? -Am I right about that? -You're right. In France, only the veterinarian is allowed to use the site. In order for us to verify it's truly the veterinarian, when they login we have to check against the national registry of veterinarians in the country. If they're registered, they have the vet license, then they're able to get into the site. Now that is lessining a little bit.

It's loosening up a little bit. Now, some of the nurses are able to register for access to the site in France.

So, let's talk about this architecture question for phase one US and France. know our solutions strategist and our engineers and tech leads at Corra are very forward thinking and I know we pushed you guys to think about a headless approach back in 2020. Can you take me into the room when those pitches were made and how the decision was made not to do that in 2020? Yeah. As I mentioned, we were brand new to Adobe Commerce.

We were looking to Corra to help us. They came to us with this idea of a PWA and a headless.

We had no idea what it really was. We were clueless. We didn't feel, Boehringer didn't feel that we had the technical skill set in-house to be able to support that. And for us, we felt like we would be more comfortable that we could better support a monolithic build. After several discussions back and forth, we decided to go with the monolithic build for the US and France. Got you. All right, so we get that launched in roughly 2021? Yes, 2021, and then we had a relaunch in 2022 because, to make things more fun we had an ERP migration during that timeframe. -Always fun to do the ERP migrations. -Yes. So let's talk about phase two. We partner together, we successfully get these two large markets launched. -Yeah. -What happens next? We decide to go global. And what does that mean, business-wise for Boehringer? So what it meant is that we had to modify our approach a little bit. When we targeted the US and France, we worked very closely with those markets, we gathered all of the requirements line by line by line, and we knew exactly what they wanted. We knew if we were going to go global, we wouldn't be able to do that with all the markets. We didn't have all of the resources to take the time to do all that. So what we decided to do was come up with a more of a templated approach. And as we looked at this templated approach, we knew that we could roll out faster and we could give these markets more of that core commerce functionality that they were looking for. The US and France had sophisticated loyalty programs and loyalty features. We didn't see that in many of the other markets, so we didn't feel like we needed to bring that monolithic approach to these other markets. We wanted something that was a little bit more flexible and we could focus on the core commerce components. What were some of the other local variations we were dealing with in this next set of markets? So, when we go to the other markets, we've got various animal species. You know, pets, equine, cattle, swine, poultry. In the US and France it was primarily pets, some equine. Now we're looking at all the other species. And in some of these markets, we were also targeting the human pharma customers as well. So there's a different product set. Like before, it's a multi-currency, multi-language sometimes within the same storefront. We had different taxes, different regulations, different ordering flows, different payment flows. It was really quite complex to roll out to these other markets. So, would it be fair to say you basically had unique, or not unique, but very different front end requirements per market? Absolutely. 100% yes. So, knowing that, we charge the architecture again. Yeah. Suggested a headless approach a second time and how did it go? We were very much open to that idea. That is the way we went. No surprises here, right? This is what we're here to talk about. It was the right thing to do and I'm glad we went down that approach so I appreciate Corra with your advice and your consulting on that and helping us with that development effort. Let's look at kind of how it's structured here. Can you walk us through this diagram a little bit with the PWA front ends representing each of the different phase two markets? So, across the top we have the various PWA front ends. So those are going to be the different markets, market one, two, three four plus however many others. And they are all linked with the core Adobe Commerce back end. We've got the various components, various modules.

On top of that, we have a third party seller in some markets that we work with that also needed access to some of these components. And then, through those and through the orchestration layer in the middle layer, we connect to our various systems. We've got ERP, CRM, territory management. Any number of other systems that we've got going, various payment gateways that we connect to. And speaking of that orchestration layer, I think that's really where the magic happens, right? Exactly, yeah. And I believe we're already actively discussing shifting that custom built layer to API Mesh and App builder. -Is that correct? -Exactly. You're kind of stealing the thunder for what's coming next, but that's exactly right. We've got a POC in the works to see how we can move to API mesh and App builder.

All right, so just to kind of recap again some of the concepts here with the technology. Phase two, all of our global markets, it's a headless application built on top of the Adobe Commerce infrastructure. The front ends are also PWAs. These are kind of old news these days, but a progressive web app which really just gives the benefit of faster load times and faster speed. So that was a big part of the benefits here for our clients.

And we talked about composable a little bit at the beginning. I know it's a lot of buzz out there. Quotes like this one about time to market from Gartner is a really key one and I think was part of the discussion as well. But fundamentally, the way I'm starting to think about it, Nathan, and tell me what you think. What we really have done here is develop a microservices based approach. -Right? -Exactly. And we're continuing to migrate along that path to more and more composable as we move along this journey. Let's talk about some of the business benefits of this microservices approach built on top of Adobe Commerce because going back to what you said at the beginning, why you chose Adobe, the vast feature set that's built into the platform. By applying a microservices based approach on top of that, are you a little bit trying to have your cake and eat it too? We are. Not only do we have the technical benefits but also we get some of the business benefits. It gives them flexibility. They can-- Each market can do their own thing. We provide templates for them and then they can create their own pages, do their own promotions, do their own programs. That's flexibility. It also, you know, it gives us-- They're only working with one vendor, right. They're working with the global solution, us, and each individual market doesn't have to go out and shop around and try to find their own vendor to work with. So it also helps them knowing that there's only one team that's going to be supporting them as well. It also gives you more flexibility to meet your needs of your individual markets, right? So let's say Brazil says, "Hey, we have a whole new payment system we need to integrate," you can do that just for Brazil much easier, right? Exactly. With this PWA deployment, with the PWA front end, excuse me, we can do deployments just for a particular market and not have to worry about the back end and disrupting everybody else. Because, as a global company, we've got markets all over in all the time zones, inevitably you're going to interrupt somebody if you do a mass deployment. So we really tried to reduce that as much as possible. Got you. One more feature on top of that is, as you were mentioning, perhaps Brazil asks for something. Well, when Brazil asks for it, and we create that, it now becomes available for the other markets. So the other markets can take advantage of this feature. So, as we continue to build up the base, each feature that we add adds to the basic footprint of the application. Whereas, if you had built it like the US and France in phase one, as a full monolith, you would have had to prioritize those requests against each other rather than independently, is that right? Correct, but we still have to prioritize the requests that are coming in. One of the things that we wanted to do is develop what's going to be the most value for the business. And whether that's for a single market or if it's going to benefit multiple markets, we need to weigh those priorities of what those demands coming in.

All right, let's talk about some of our technical benefits.

Right. Obviously it gives us a lot of flexibility, we've got scalability, we've got the microservices so we can focus teams to work on certain areas. We've got an order to cash team, we could have an accounts to reporting team, so our A to R team. We also have a logistics team, or our 3PL team. So we're able to split up and really focus in on these microservices. And do you find that that organization change, dividing the squads up in that manner, makes for more efficient teams, happier teams, happier developers? I think so. It allows them to specialize, it allows them to focus. It's also nice though to have some folks who have that broader view across the platform and across the various systems, because we need to make sure that we're all communicating and that if we have a demand that's going to go across the order to cash team, but it also has a comp onent on the logistics side, that those two teams are talking to each other and that their backlogs are working through at about the same rate so when that feature is developed it will be ready at one time and we can take advantage of it, rather than having part of it wait for a couple of weeks or a couple of sprints while the other team is building up their portion of it. Understood. All right, let's talk a little bit about the business impacts of digital transformation. Because, fundamentally, this story together, this journey together is one of digital transformation. How have your markets that have adopted this headless microservices based approach done in revenue versus those who have not? When we launched this platform, sometimes we would launch it into brand new markets, other times we had legacy platforms that we were replacing. And when we launched in those markets where they had a legacy system, we saw a two-fold increase in incremental revenue. So, in growth in those markets. That was quite impressive.

I'm sure your business stakeholders were happy with that. Absolutely. That's what allows us to continue to seek and ask for more investment and say, "If you invest this much in us, this is what we can do. If you invest more, obviously we'll be able to do more." That's one of the messages that we do. The other thing that we've been able to do is utilize that up-sale, cross-sale functionality within Commerce more effectively, more efficiently, and that has also added to the incremental revenue. Which again, that's a base feature of Adobe Commerce, baked into the platform that you didn't have to go rebuild through another vendor. Yes, exactly. Let's talk about time to market. We had that slide a minute ago with the Gartner quote. 80% faster et cetera, et cetera. Has that proven to be true for Boehringer? I do believe so. We've done some measurement over the last few months and we've looked at the time to market. We've measured two different things. We've measured time to launch a market and we've also measured for those life markets that have a new demand or a new request for a new feature, how long is it taking us to deliver on those. When we look at the launch of a new market, we see that it's gone from, you know, the US is a very large market and it took a year or so to launch that. We did have the ERP migration in the middle of that though.

And we look at some of the newer markets and within six weeks or so we can launch a new market now.

If we look at the time to market for some of these demands, we see that we can get a demand, we can process it, prioritize it and have it developed usually within two to three weeks. That's much faster than what we were seeing before where it was, it might be three or four months for us to get through that demand. So, definitely we're seeing quite a bit of progress in time to market. Fantastic. From months to weeks. That is the headline. Let's talk about data efficiencies. How has this microservices based approach helped you guys manage your data a little bit better? For us, one of the things is-- We're a very large organization and we've got data all over in various platforms and systems.

What Commerce has allowed us to do through these microservices is to select certain data and share it with a customer in ways that we haven't been able to do before. A great example of this is for our loyalty programs. Our customers had different loyalty tiers, like mini loyalty programs. What we can do now is when that customer logs on, we can show them where they are in their loyalty tier, and what is remaining to get to the next tier. One of the things that we're going to do now we've got a POC ongoing, is we're going to do an AI product recommendation based on their purchase history, as well as which program they're in and the tier and what they need to get to recommend the next set of products they should buy, or create a suggested order for them. And are those loyalty rewards, are they targeted more at the office managers in these veterinary offices or at the vets themselves? How does that work? It would be targeted more at the veterinarians and purchasing products so they can then support and sell to their customers. Got you. Let's talk a little bit about an organizational change. You mentioned before, reorganizing your technical teams, kind of in parallel with Corra's technical teams to be centered around individual services. Can you tell us a little bit more about that? So, not only did we have our technical teams reorg, but we also did a reorg around how the business will support these platforms. Previously in our older system they were supported mainly by customer service. Once the product was launched in a market it usually just sat there. It was very passive.

Now when we launch these products or these platforms in our markets, we're very much hands on and we're engaging with the markets and asking them to make sure that they're reaching out to their leaders, to their stakeholders to make sure that they're engaged in sales and marketing strategies and that these platforms become an integral part of the channels for engaging customers in sales and then selling in these markets. So it sounds like your local market leadership is just getting a lot more specialized attention and focus than they previously would have gotten from their technology teams. Exactly. We're trying to imbed the capability of digital commerce rather than just a platform. There's an analogy that I was once told that I can give you a laptop but that doesn't mean you're going to make money. You actually need to do something with that laptop in order to make money. And that's kind of what we're saying with these markets. We will give you the tool, but you can't just sit there. You actually need to do something with that tool in order to make money with it. Understood, understood. All right, let's talk a little bit about what's next on the road map. What's coming? Well, we've already had a couple of sneaks on that, right. We're looking into App builder and API Mesh. Super excited about that. I think that's going to be a game changer for us. As I mentioned, we're looking at AI product recommendations. But we've got a couple of big things coming up.

So, the US and France we talked about are on that monolithic build. We're actually going to move them over to our PWA global build. We're going to start that later this year and that will probably carry into next year. But I see that we're going to gain a lot of efficiencies out of that and a lot of savings. We're looking forward to that. The other thing that we're looking at doing is we're going to completely revamp our loyalty portal or portions of our site.

It's just going to be a complete transformation overhaul. Hopefully we'll be able to take this then and take aspects of it and apply it globally because most of those global markets as I mentioned before, are focused on commerce right now and we want to be able to give them some of those loyalty features. And that loyalty initiative is also not just specific to animal health, right? That's also human health in certain markets, right, who have different programs I would assume. Right, right. Exactly. So, yes it is primarily animal health, but yes, if there's a market that has human pharma that we're in and we have the platform there and they're selling their human pharma products on our platform, then absolutely they can take advantage of the loyalty components as well. Got you. All right, let's talk about some tips for this kind of transformation and what a successful transition into a headless or microservices based approach on top of Adobe may look like. What advice, Nathan, do you have for anyone else in the room who may be contemplating a project like this with their own organization? Obviously, I think you need to have an agile approach. There's going to be some ups and downs as well as you need to be able to deliver quickly. I think having this iterative approach will allow you to do that. I think one of the big things that you really need to do is get that stakeholder alignment. One of the things that we continuously do is we want to be top of mind with our stakeholders. Especially our executive stakeholders on both the business and IT side. We are constantly providing them updates on the actions that we're taking, as well as the value and the impact that we're delivering to our markets. And what we do is we have our markets actually report back to us instead of just us saying to our executives, "Hey, this is what we've done," it's actually from the market saying, "Hey, this is what was delivered to us and this is what we've gotten out of it." That's really powerful when the executives can see first-hand what it's doing for those markets. It sounds like you've just developed a much tighter partnership with that individual market leadership and your technology teams. -Exactly. -Got you.

Okay, so that's basically what we have today.