In this episode of the Design Systems Podcast, guest host Andrew Rohman sits down with Jon Warden to explore how GSK leverages design systems to drive consistency, innovation, and efficiency across a highly complex global organization. With multiple brands, multi-product ecosystems, and diverse regional needs, GSK must balance scale and flexibility while embracing automation, AI, and modern workflows—all while keeping the customer experience at the center of their digital strategy.
Key Topics Covered:
- How GSK creates design consistency across multiple brands, products, and global regions
- The role of design systems in managing complexity at scale
- AI and automation in streamlining workflows and driving efficiency
- How GSK balances innovation with regulatory and operational constraints
- The impact of scalable workflows on accelerating digital transformation
- Building digital experiences that meet customer needs while maintaining global consistency
Tune in for a deep dive into how one of the world’s leading pharmaceutical companies is leveraging design systems to power its digital strategy, drive innovation, and create seamless customer experiences at scale.
Guest
Jon Warden is a user experience, design, and product strategist with over 25 years of experience. In the past decade, he has created, led, and managed UX, design, and product teams across various sectors, including news (The Times, London), B2B, B2C, media agencies, telecoms, and charity. Jon focuses on providing clear UX direction throughout the development lifecycle. He and his teams emphasize outcome-based user research, data, and user testing programs. Through UX design practices, they support the creation of user-centric products that engage customers across all channels. Currently, Jon is the Global Director of UX, Design & Research in the Commercial Digital & AI division at GSK.
Transcript
Chris Strahl [00:00:00]:
Hi, welcome to the Design Systems Podcast. This is the place where we explore where design, development and product overlap, hear from experts about their experiences and the lessons they learned along the way, and get insights into the latest trends impacting digital product development and design systems from the people who pioneered the industry. As always, this podcast is brought to you by Knapsack. Check us out at Knapsack Cloud. If you want to get in touch with the show, ask some questions or tell us what you think, send us a message over on LinkedIn. You can find a direct link to our page in the show Notes. We'd love to hear from you.
Andrew Rohman [00:00:25]:
Welcome to the Design System Podcast, brought to you as always by Knapsack. I'm Andrew Roman, standing in for our regular host, Chris Strahl. Before we get started with our guest today, a quick plug. As you know, Knapsack does a ton of events and content, always trying to bring more out to the community and bring the community together. We have a few upcoming Patterns Leadership Summits. We're going to be in Chicago in April and in New York City in May. And of course, we do weekly live demos every Thursday and keep an eye on our blog for more posts coming out every week. You can check out all of our events and blogs at Knapsack.cloud and get in touch with us on LinkedIn.
With me today is Jon Warden, Global Director of UX at Glaxo Smith Kline, one of the largest pharmaceutical companies in the world. Jon, really excited to have you here. Thanks for being here.
Jon Warden [00:01:12]:
Great stuff. Thanks for having me as well.
A little bit of my background actually. So I've been at GSK for about six years now. I've got a team of UX, UI and research people. Fundamentally, part of that group is looking after design system management. I've been in the world of design and user experience for probably 25 years. Got a relatively big orchestration in terms of what we do across the different levels of our design ethic.
I work in a group where we have many more other teams who do design. In gsk, it's not just me and my own group, but we do try to orchestrate commercial and medical marketing and field experiences in our space.
Andrew Rohman [00:01:47]:
Well, there's a lot that I'm excited to get into today with you around design systems, around just the sheer scale at which you're operating, obviously around AI and the future opportunities. But I think, you know, it really all starts from a core common need and theme, especially in your enterprise context, which is how do we make a lot of stuff in A way that reduce costs, creates efficiencies, improves outcomes, you know, not just for the programs in your case, also for patients, for providers. And so, you know, when we think about that idea of scale, maybe you can talk a little bit about just the scale of ecosystem and programs and channels that you and some of the peers in your, in your area are responsible for.
Jon Warden [00:02:24]:
Yeah, sure. We predominantly look after our, in the marketing space, look after our channel deliverables. So that's our kind of enterprise web frameworks, email management, edetail management. We have a series of Personas, we have marketeers, we have sales reps. Obviously we go out further than that to external customer base as well. We look after healthcare professional and also to some degree patients. But what we're trying to do is give our colleagues and the people who work in our company the ability to have good design at scale. And we work, as I said, across multiple different areas of that, but because of the landscape, because of the scale of the company.
So there's 100,000 people who work at GSK. We work across multiple different markets. So we global and then local. My team are in global. So we have to orchestrate. And this is the important part, consistency at scale. So how do we get consistency at scale when there's individual use case and different need? We have it at a global level. We have to disseminate that across to local markets because what you have in France, what you have in Germany, maybe what you have in a different locale is a different set of requirements.
I think when you work in pharma as well, we then have regulatory issues as well. So you can't always do the same cricket cut across everything you do. So when we think about design and how do you think about delivering, maybe take the word design out, just delivering good experiences to customer base, you then got to think of where all those nuances that come into play as well. How do we get the right mandrakes in? How do we get the right accessibility in different environments? How do you get the right design treatment in. When in certain markets they read from right to left rather than left to right, there's things like that that come into play where you have to think about the bigger picture. We don't always get it right, by the way. And I think that's an important factor as well, because trying to deliver something of scale for that amount of different variation is actually quite hard to do. So we create global services that we can then cookie cut to some degree, but we allow the flex to happen in areas where there's a different need or a different requirement and a different regulatory need.
And that's actually when you think about the models, when you think about web, you think about email, you think about all the different channels, social and so on. You do that at scale across the board. Getting that quite correct is actually very difficult to do. Which is why design system management comes into play when we do think about design. But there's also, there's a lot of hand holding and a lot of conversation that happens internally as well.
Andrew Rohman [00:04:52]:
Yeah, no doubt. You said something in one of our prep conversations that, you know, design at scale or just this, you know, ability to create consistently at scale is about hygiene. Right. And you have this interesting challenge where not only do you have the complexity of global and local and cultural considerations and all the things you mentioned, you also have not only that 100,000 person workforce, you have external contributors and collaborators all over the place. Right. Large agency and consultancy, you know, support network. And maybe this is the segue into the systems conversation. But how do you enable that type of operation, that type of hygiene when there really is that much complexity, just at a sheer human operational communication level, even before we get to fancy tools?
Jon Warden [00:05:32]:
Yeah, and I think communication is key in this aspect. So what we're doing at the moment is trying to create integral central repository of design ethic, call it through a design system management. But it's wider than that. It's about the right approaches that come in, call it decision tree. So how can you come in as an external and just context to what you just talked about? We have 70 plus brands in GSK. I don't work with all of them, but for every brand there is an agency somewhere, an external AOR that actually is doing a lot of the campaign management. They may be creating content or they may be creating design ethic, they may be using our design system or in some cases Figma direction. We use Figma as a service to then utilize that and then create the abilities for them, it come back in.
And I think when you try and do that across multiple multiples, when actually when you give something to designers, we will know design is integral in terms of creativity. But you can't then cookie cut that because each designer will interpret it in a different way. So what we started to do and are doing is creating that central destination where they will come in and we have to give them access to our services and tools through various different methods to be able to give the right directions. As I said, I'm a agency designer, I need X. So how do I find what I need to find. I'm an internal designer, or in some cases, I've just joined. I'm an internal designer who's been onboarded. How do I find out the information that's needed? I think in gsk, what we're trying to do is consolidate a lot of that experience.
But what we're trying to do in and I look after commercial and medical, by the way, that's the area of internal. My groups are part of. We're trying to give that right directional approach, we're trying to give that right hygiene, because it does come back to consistency. So we have different nested design system approaches where for our HCP portals, we have one set of criteria. For our email treatments, we have another set of criteria, even from a point of view around data management as well and data services, we have a different design system of treatment for that. And the job we're working on now is consolidating. So there is that singular point of view. And the mantra from back in the day, when I started this journey a few years back, is a button is a button, is a button and a button.
At the end of the day, a button should be a button. You can style it and you can move it and you can change it, you can change the text on it, but it's still a button. So Instead of having 15,000 different buttons, how do you get one or how do you get as close as you possibly can? Because there are different use cases. That is a button that goes into power bi may be very different from a button you might use on the website, but at the end of the day it's still a button. So how do you get that integral kind of approach around consistency and hygiene? So we're doing use and reuse as opposed to reinventing the wheel over and over again. The interesting trick, though, is when you go out to agency and then you've got a different designer, unless they know a button is a button and they actually are driving that approach, they have to be very kind of focused because otherwise they will start to move further and further away to where that creativity lies, as opposed to these are the things that GSK do. So we have to be very quiet and direct and we do a lot of hand holding. As I said, my team are always on with some of that content author type that we work with or talk to agency as we need to, with some of the sort of account directors to give them the right direction about how we work, what we could do for them.
Andrew Rohman [00:08:46]:
Yeah, you're focusing on A lot of the most important pieces, which are, you know, a lot of it's that human element. Right. And you spoke to some of the natural human tendencies of if the right balance isn't struck between sort of structure and boundary and, you know, freedom, they're going to err towards creativity. And that has ripple effects. And you know, there's working directly with content authors. I know in your world that that is a huge part of the creative experience. We talk so much about designing code, but, you know, a lot of what gets produced, especially in a highly regulated environment, you know, the content author is one of those primary stakeholders. So it's awesome to hear how you approach that kind of strategically.
I'm curious also from a bit more of a tactical level, whether this is about tools or workflows, how have you seen this sort of systematic enablement or production evolve over the past, I don't know, 15, 20 years, how's it changed and what's working, what's not these days?
Jon Warden [00:09:33]:
Yeah, I think GSK wise is the biggest place I've worked in terms of like the scale for each different brand that's part of our space is a different branded experience. They've got different colors and different approaches and different ways of working, different font, different tone of voice to some degree. So that scale of that side of it is huge. But to answer your question, I mean, I've been around design systems for a very long time. I started my kind of, my big kind of world of getting involved in web actually is how I kind of started out. And I got my role at the Times, the Times newspaper, London. And I kind of worked my way up through the ranks and I think not necessarily design systems, but systems thinking and how systems connect. Because we had, back at the Times we had a publishing organization, part of the arm, where you had a service called Hermes, which was where the editorial staff used to write into that environment.
But then I worked in the digital space and we had to bring together how does Hermes work alongside a cms and how do we then get the data points together? Back in the day, then we're talking about cope. So create once, publish everywhere was the mantra that went out. How do you do that when you're trying to get something from a singular source of raw data to be able to be pushed across a newspaper, digital product from a web perspective, emails get pumped out. Mobile devices in its infancy, back in the day when we were doing that, to be able to get the same content type, outturn the mass across mediums of channels. So that was My sort of starting point of how do we get into that? And then over time it started to evolve into, well, how does design work when you're trying to do that as well?
So interestingly, I worked at another corporation after I left the Times, where I actually started to craft a. Call it Style guide, slash Design System, baby design system. But we were doing it in WordPress because back in the day when we started cutting story short, you used to do a style guide, right? You do an illustrator and PDF it and then share it with the mass. And then obviously the advent of, like, how the web is kicking in and how did you start to share that documentation in a digital format with wider people than just the people internally from a PDF perspective? And how did that work? So I remember kind of crafting multiple components and putting them into a WordPress pages and kind of like, and this is the article teaser and this is what it does, and how do you then do that for them? Basically that system's thinking again is, how do the sum of the parts come together to create the whole? And that's, I think, when Brad Frost started dropping in Atomic Design and thinking about how that worked and like, you know, Smashing magazine, I went to a few conferences and saw some of the stuff he was talking about.
I was like, yeah, there's a penny drop moment. That's exactly how you start to label it all. So, I mean, rolling forward, this is over quite a few years. Rolling forward. When I got to jsk, I was then like, well, how do we get that going? Because there wasn't anything like that. It was very individual by brand, but that in terms of, like, kind of efficiency and how do you reduce costs? Because for each brand there was a different way of working. And what I noticed was it was starting to hit our engineering teams. There's only one engineering team centrally, but there was lots of different design agencies doing lots of different designs and they were almost coding it from scratch.
So that whole piece was. Then I did a massive roadshow in kind of going out and showing how a design system might reduce costs, increase kind of efficiencies, simplicity in terms of ease of use. And one of the things I learned from doing that as well, good design. At the base level of our design systems, we had some really good agencies who came in and helped and worked with us, who are actually the crafters of the brands themselves, and they pick up our very basic design system at the time and really knock it out of the park. What it said to me was, good design within a framework is really Good design because it's something that kind of comes back to something I used to teach my junior designers. You're not designing for you, you're designing for your customer, which is a really good mantra to have to be able to take you further forwards.
Andrew Rohman [00:13:36]:
What I'm curious about in all that is how do you close the gap between we've decided on standards and guidelines, we've even communicated them well, and the actual uptake and sort of use of them. Right. Because going back to that human behavior, it still comes down to, does someone read that, you know, acknowledge that, do that, you know, follow that behavior? I'm wondering if you figured out that silver bullet just yet for our audience.
Jon Warden [00:14:00]:
I think it's really, really hard because I think also especially when your environment is completely changing all the time, what I mean by that is we don't, you know, we're not an organization that has the same consistent amount of people in it all the time. So for each brand, there's a set of different marketeers who are different thinking processes, and they turn quite a lot. They move to a different brand, they might move to a different company. So it's persistence. There's a statement of like, there's a difference between tenacity and flogging a dead horse. But I think in this one, you need to be persistent in your message continually. And I think it goes back to my thinking of, like, how do messages get shared? How does communication get shared? Back in the old days, in the, you know, medieval times and so on, they used to have people who used to walk from town to town sharing the news, and they used to stand in the middle of the town and the town cryo would be like. And so.
And so it's happened to the. We spend quite a lot of time and it's peaks and troughs of it, but we spend quite a lot of time doing communication methods. So we use various different channels within gsk. Having destinations where people come to. We have in the past on training to onboard agencies and kind of reinvent the wheel when they do come in to have conversation around how that works, it's about continually putting your finger on that pulse. When the infancy of the design systems in GSK was starting to go, I did so many presentations of the same thing over and over and over again just to kind of get it into the mental model of how people can take this information, use it, run with it, do stuff with it, and then consume it and then turn something good with it. It's not an exact science because everyone's got a different interpretation. What I think and what I say is different to what someone else might think and say.
It's about making it very, very clear. And we're on a push at the moment to even do more of that at the moment, to share our comms in a wider way, just to get that consistency of the message. Don't go off and do your own thing, use our stuff. Because I think one of the important parts about that as well is if you use our stuff, it will get done faster as opposed to go off our pathway. It will take more time to do it.
Andrew Rohman [00:16:03]:
This seems to me the segue to talk a little bit about automation. You know, that always has seemed to me in systems, to your point, not necessarily unique to digital production. That's part of the goal of systems. If you can, you know, create repeatable solutions to commonly reoccurring problems, which is how our CTO would define the word pattern in this context, you know, that then leads you to the ability to automate more of those. And obviously we'll get to AI a little bit. But what are you seeing as the opportunities? Have you experimented? Is GSK experimenting with automation to close that gap of how much you still need the person to do what you asked?
Jon Warden [00:16:34]:
Yeah, well, I think we amongst many others are always in the space of experimentation. You know, how do we evolve our business models? How do we simplify to get faster efficiencies in place? Obviously, the advent of AI is really helping turn that corner and speeding up those processes. But I think the thing that we need to do is be able to automate our processes so you can put design in the hands of non designers to be able to get the same, if not better experiences out. And that ultimately comes back down to how do you deliver really good experiences, not for the marketeer or for the sales rep, but for the end customer, which is the healthcare professional or a patient. And how can you seamlessly engage simplicity to get that out the door faster? So that's our journey and that's our pathway to do that through different components. I mean, calling AI out here, you can do this. Now you've got a good design system with an infinite amount of components, although you don't need infinite, you need just the right amount of components to be able to deliver X or Y. And then you've got to be able to put the content that you have created.
And I'd much prefer people come up with really good content as opposed to worry about the design treatment as such, to be able to deliver those experiences. Because I think you got to think about your audience as well. Do healthcare professionals, do doctors really need to see a flying image flowing around the screen and you know, et cetera, et cetera, or do they just want to read concise content? So depending on your audience and depending what you're trying to share and do with a series of building blocks, you can create some sort of vehicle that enables you to put the right content in it and get it out faster. And I think that's where automation comes into play. If you don't have to wait around for engineers to code the experience, if you don't have to wait around for a content author or an authoring team to fulfill your like building blocks with the right content and you can do it yourself, I think that then enables the speed to happen. And in our space we have to wait for kind of like approval methods as well. It's not a case of like, you know, X person can just review that and it's all taken down. We have medical and legal review bodies.
We have to go through certain heaps to go through so where we can reduce frustration points and friction points where we can get it through those processes faster. That's where automation comes into play to get these things out of the door. It's not going to be like instant, but it's going to be faster than it actually was before. That's our aim.
Chris Strahl [00:19:05]:
Hey everyone, I'd like to take a quick break and tell you about Knapsack's leadership summits. Every month we host an exclusive in person event in different cities all around the country, bringing together design, engineering and product leaders. These summits are all about sharing our learning with tailored discussions to address the challenges you're facing. The the best part, it's totally free. Head over to Knapsack Cloud events to find an upcoming summit near you and apply to join us. We'd love to see you there.
Andrew Rohman [00:19:32]:
Yeah, you know, I love that because you're speaking to one. The whole hey, systems are going to kill my creativity piece, which is. No, the goal is should we should help focus the creativity on the things that are most worth our time and effort and creativity. But you know, I also think, you know, when you talk about the regulatory side, it's really interesting to me because I'm always interested in how structure that teams need to work within can aid the sort of system mission. Systems are all about structure in my mind. Right. It's we've defined parameters that if we work within these, we can solve for a lot of those patterns. A lot of those commonly reoccurring problems really easily.
And, you know, then we can decide where we don't have a solution and also maybe where we want to modify a solution. But in that way, product planning experience, design planning becomes more of a triaging exercise where we say, hey, we know how we can use a bunch of stuff to solve these problems. Let's go focus our effort over here. To me, though, one of the challenges we faced, other than the human challenges, is how do we work within that structure with some rigor. We've got components that have properties, we have themes. But if the design is happening in an infinite canvas and you can go change the components and change the theme structure with no consequence, that's very loose structure. The reason I bring this up in your world is I think content is a really interesting angle for structure. Right? You have certain things that you have to have certain fields basically, that need to be there for regulatory approval.
And, you know, in a way, I'm a believer, and it's going to get us into AI as well. I'm a believer that we need to have a little more rigidity in that creative process in order to get that outcome that you're talking about in order to focus the creativity. Because as long as it's easy to break the structure, structure, we're going to have that ripple effect and challenges and all the things we've struggled with in shipping products the last few decades. But it's why at Knapsack, we've started building some of those prototyping tools that say, hey, you can't change the structure of your component, you can't change the structure of your theme, you can't change the structure of the web. Right. The way a web browser works for us, those are the three parameters you have to work within. Content represents a fourth one. So I find that fascinating anytime that we have the ability to say, hey, there's stuff you can't argue with.
You can argue there should be a different prop on the card component, right? But like, we got to have this black box warning, we got to have this medical whatever it is, and how do we use that as a way to kind of embrace the structure more than lean towards the creativity, as you talked about earlier.
Jon Warden [00:21:50]:
So on that, I'm going to pivot slightly out of our world. Let's do it. I wanted to be an architect at one stage in my life building houses. And actually, if you think about it, and I used to use in some of the decks I had that an architectural drawing is the same as, like, UX like architecture actually you're going to build a house. To build a basic house you need four walls, you need a roof, you need a front door, you need some windows. And how you design that house to make it your own experience is very much about how you, where you put the walls and how long one wall is compared to another wall. And where is the front door going to go? Is it going to go on one side of the building or another side of the building, etc, etc. And then when you brand it, you come in and you have an interior designer do some bits or you start it in a different way.
You might, you know, you get my point. When we're talking about structure and rigidity, I think there are some basic constructs that you have when you build out these kind of systems. Integral the systems which you are non negotiable. You have to have them to be able to do that. I mean you talked earlier on about in a browser you have a certain way of a format for it to play. If you break down a web page or an email to its some of the parts into individual elements, there's only a certain amount. That's why I said earlier on. But it's not infinite in terms of how many parts you need.
You need a certain amount and then you start hitting a gray area. We probably don't need them as much anymore because you've done, you've got 15, 20, 30, 40 components. And then after that do you really need those extra ones? Because it's about, if you go back to the architectural thought process, it's about how you move those parts around and what you do with them. And that's where the art of really good design comes in. I think this is where. And everyone goes, well design system by everyone. But people go, oh, design systems, you know, they're there for a certain view. And I want to be really creative.
My point is it's where good design comes into play. You can take those parts and it's about how you assemble them. And if we allow the flex in the assembly and then the visual theming that goes over the top of it, regardless of what the channel is, you can get really good experience. And as one of my good friends used to say, content is king. Because it's not necessarily about the vehicle, it's about the imagery, it's about the visual interpretation, it's about the content you put in it to make that thing sing. If you can get that right, doesn't really matter, does it? In terms of like, you know, oh, that's over here and that's over here and that pixel is slightly out. As long as your system does what it does and that's where the rigor comes in, then you can get creativity at scale, faster efficiency and if you've done it properly, cost reduction because you don't have to reinvent the wheel over and over again. And the beauty as well, I know you guys are looking at extra additional seat management where you are allowing others to come in to play, to play around in playgrounds so that they can then see how those layouts work.
If I move a component up, if I move a component down, if I want to turn the button off and put a hyperlink in instead, I can do that in an interface, which means that that puts the power into the hands of not necessarily non creative, but maybe non designer or non engineering type people. Because the more breadth and depth that you've got of people who can do that, the faster you can get stuff out the door.
Andrew Rohman [00:25:04]:
Absolutely. And I think, you know, structure looks different for every company. Right. We know a Fortune 500 company whose goal, whether or not they think they'll ever get there. Their goal, kind of North Star, is no components. They want to enable teams to do what they need to do in a way where the structure is mostly focused on tokenization and, you know, they've got a path that they believe works for them and that's fine. Right. The structure should be appropriate for, you know, what creates a base of stuff that we can reuse so that I think a lot of what you spoke to is we're going to have the creative process, but can we move it further down the production cycle? Can we let a non technical person make, you know, again, our prototyping you spoke about is I'm going to go make a thing and I don't expect most teams, at least today are going to ship the exact thing they made.
Maybe. Yes. The marketing landing page, that's 100% components.
Jon Warden [00:25:48]:
Great.
Andrew Rohman [00:25:48]:
You literally made the whole thing and you cut off 80% of the design and production cycle. That's amazing. The more likely thing is we hit the limit. We do that triaging and we hit the other two buckets of, well, I need to make these things and I don't like these two, so I got to modify that. Awesome. Right now we have some iteration cycles or maybe you do ship the thing you can make in an afternoon with the intention of measuring and experimenting with some other creatives, some other options and then you've got this focus. Then you know, it's Going to be a lot of behavior change that comes with this where there's still a lot of teams beholden to some executive is expecting to approve a PDF before they hear about any lick of code being written. And like that's kind of backwards to this.
The cultural change management, as is usually the case, probably be the hardest thing getting humans to change behavior. But I think that's the real opportunity of this idea of working within structure is not to say you just got to use it and if it's, you know, don't make anything new. It's let's just do everything with intention. The structure and these systems enable you to do the creative process with a ton of intention and cross functional clarity. We all know what we're doing and why. Wow, wouldn't that be nice at the start of a design to dev workflow.
Jon Warden [00:26:51]:
There's another part to that as well which is the low hanging fruit. So if you can do the base levels well, if you can enable teams to do that low hanging fruit, I need a 5 page microsite to do a campaign on whatever drug, maybe patient side that's going out there at the moment. I've just been very loose around that time. It needs to be done quite quickly and it shouldn't really be a hardship to do it. And if you provide that kind of a service to quite a large percentage of the populace, then great. What then that does is if you got that ability in motion, it then frees up your own teams to think about the bigger, harder bets and the things that need to be done. They have a little bit more thought and a bit more process around them ideation that's needed. I think maybe, you know, there's always the thought process about going out inward, outward.
But what about our own internal groups and how do we then organize those to be, you know, freer to ideate and also as well to add more complexity to what you've already got. So you know, you've got design systems, you've got systems thinking how can we add to that, how can we make them more efficient? That is definitely a bonus of doing this. You know, you can get out a whole campaign suite at a basic level. I keep on saying to my groups, you need that kind of mailchimp slash squarespace experience so that anyone can come in and create a website, keep an email or do a social post, whatever it is they need to do. And then it frees my teams up to get on with some of the bigger bets we're working on because the stuff we're talking about here is one of those areas. We have lots of other things that we do as well.
Andrew Rohman [00:28:25]:
Yeah. And this is interesting double edged sword here too of the efficiency concept where it's often this, hey, we want to ship more product or ship more faster and like kind of. Right. That hopefully that's not the goal as stated. Right, ship more. We also see the other side of teams saying we've wrestled with this as some of our ROI as a platform is make your teams generally more efficient. And we're like, that doesn't mean in these. Sarah, go lay off 15, 30% of your team, please.
We feel bad if you do that. If teams think about that, my questions are always, do you not have a backlog and have you never had an awesome idea that you think would 10x something in the product that you just didn't have time to get to? Right. What's your cutting room floor look like and what's your backlog look like? And that usually is a good, you know, different way of thinking about. Like you said, whether it's we're adding complexity, we're finally doing that conversion optimization project or whatever it is, or we're finally going to tackle some of that design debt, some of that technical debt that we keep sweeping under the rug. There's so many good benefits that come from a little bit of resource efficiency.
Jon Warden [00:29:21]:
Even going back into the world of like, you know, understanding the customer base themselves. I mean, we're trying to set it up so we get better feedback loops in our own design system. So we're continually learning. I'm in a massive fan of Teresa Torres in terms of like continual delivery, continual discovery approaches so that you're always having your finger on the pulse. And I think there is definitely a use case in terms of user research of all the services and all the things you do to be able to then feed it back into the systems you've got. And if you're always having to create, create, create and you don't get that pause, then you never learn about what you've actually just created. So always having that understanding in place to be able to then feed it back into the loop or feed it back into your backlog or reprioritize whatever it may be, because you learned something new in a phase where you release a new feature and it did X and then how do you then bring that back in to course correct what you're doing or pivot your product? So yeah, with you on that one.
Andrew Rohman [00:30:14]:
So this to me is where I think workflow wise, the AI tools we're seeing can really potentially start to help. I think there's still an interesting gap though, especially in these highly regulated worlds. Maybe I'll get your thoughts on this first. You know, you've started to see some posts recently. I'm starting to see them on X, on LinkedIn, around. Hey, how can I get one of these tools like replit or whatever, v0 how do I get it to use my stuff? Right? Because it's amazing. Robots are starting to make production quality code. I can take Figma and I can make a shippable website.
But you don't want a bunch of fresh code because that's a bunch of fresh regulatory review and all that stuff. Just like you don't want, you know, bespoke content for everything that's never been approved before. Right. So are you thinking about that problem? Or maybe this is just how are we thinking about AI? Because I think if we solve that problem in your world, some of what Knapsack's trying to do, you know, that AI opportunity starts to get exactly what we're talking about. How do we just fast forward to that part where we do that creative iteration cycle on the stuff that matters? What are you seeing? How are we going to get there? What's it look like? What are you dreaming about?
Jon Warden [00:31:15]:
Purely from my own point of view, I think automated systems driven by AI to help engage speed factors, create content in a certain way and deliver out to a customer in a very fit for purpose, exciting viewpoint is where we need to be. I think what's happening in the AI world at the moment is enabling us to do that. As you said earlier on, there's many tools, many services that are now starting to fly and they're coming out regularly. It's almost, it's very difficult to keep up with them all. You know, things like Gumloop, looking at agentic AI and how that's working and how do you enable what you already have to be hooked up to be able to create better ways of working, to deliver better experiences? How do you harness the current platform structures and then bring that together and mold it and use the power of AI to actually push that out in that faster, more effective way. So, I mean, I won't go any further than that, but I think we're doing quite a lot of work in that space. We're experimenting, we're looking from a design system perspective. If you've got a fit for purpose design system with, you know, design token management or component management or atomic design principles, you have a destination of choice where people can look at and do and build in the background.
But also if you can create an AI that does build for you and you can feed your vision, your visual identity practices, your theme management, your content approaches, we have lots of content from all the different areas and then you can then form that into some sort of vehicle that get out into front of customers and you can do that with speed and efficiency and simplicity. Then I think the opportunities are there. I think the thing for me is how do we build it? How do we get it to that point? There's a bit between where those platforms or the agentic AI or your gen platform management is sitting and then where the service management of comic marketing or field experience is sitting, there's a juxtaposition of like joining it up. And that's the sweet spot. And I think that's the bit that is needed to kind of bring it together. It's not just our company, I think that's every company that's out there. And so how do you utilize and harness it to be able to create and craft something that's fit for purpose for the use cases in front of you? There's a little bit of a journey to go on there. But like everybody else, we're all on that journey.
We're all going to learn and work together and see how that plays.
Andrew Rohman [00:33:43]:
Yeah, absolutely. I mean, that's a lot of what we're seeing is, you know, the opportunity to really create more of a control point for AI. And you know, I've seen a lot of other folks recently talking about how do you enable a better agent experience, you know, and thinking about, you know, you got a generative AI backend, you've got content stores, basically large headless CMSs somewhere, you've got deployment architecture where things need to live and you've got some input, even data coming from figma, whether that's, you know, definition of components, definition of theme structure. Obviously you've got real bespoke, high quality, artfully crafted, coded components that are accessible and all that. And how do you create this ability to say, not just we have these source elements, but you know, a home for that structure. We're talking about, hey, these are the allowable combinations. This is what you could put in a card. This is the theme structure.
And then when you start to have something that can, you know, and we're trying to head there from structured data and APIs to say, all right, let me hit something to understand what components are available, what can I use? What kind of data do they expect. How do I feed that data in from something like a content store, you know, and ultimately come out the other end with something that I know I could ship quickly or we can go iterate on some stuff. So I, you know, there's, there's certainly a little more infrastructure and glue to be done there, I think, in industry in general. But, you know, it's both the opportunity and the lack we're seeing now is, you know, how do we create more control points so large, massive companies like GSK can take more advantage of those opportunities and not just create code, but create something with your code.
Jon Warden [00:35:11]:
Good experiences. I think the true part of that is just being ready. So, you know, it's evolving all the time. And I think services like yours, we're working with you to help support our end games, but we're trying to get ourselves ready so we can, you know, do I want to rephrase it and say the AI piece is something that's helping to join the dots up? That's the bit that I'm talking about with the umbrella that sits across with it all is that the connector to bring the different stacks together to be able to give that harmony out there. And actually, I think going back to being ready, we're talking about our current channel methods and who knows where those channels will be in the next five years. Will people even be looking at websites, et cetera, et cetera. So I think it comes back to, if I go back to what I was talking about at the times and having that repository of content that we used to create, the editors that used to write, how do you put that into any environment with a good visual experience? Because obviously my background is the design space, so I'm going to talk to that. But how do you then harness that to be able to put it in whatever medium is needed? And that's where the power of AI comes in for the right customer
Because then you are giving the people the more they want when they need it. That's the most powerful piece around all this stuff. If you can wrap it in a nice experience, which is why design systems and other things and centralized approaches are really important, then you might win more than others.
Andrew Rohman [00:36:39]:
Absolutely. I mean, it's consistent. One of the top trends we just see in general with CTO organizations is a lot of replatforming, a lot of core infrastructure work that's probably been put off for a long time because I don't think these things have all been figured out yet. And honestly, while their resources are there, there's also the highly regulated, the large industries, there's a lot to wade through to get there. And so there's both a lot of innovation and also probably a longish path. But you know, I think there's acknowledgement that if we don't standardize, even within the time that the web will exist, if we don't standardize our front end of, you know, how we deploy websites after we've acquired 15 companies in the last 10 years, you know, we're not going to really be able to leverage these generative production tools to the extent we want. So I agree 100% with what you're saying. We're seeing a lot of that foundational work, not just on let's all migrate everything to web components, but you know, on a lot of the infrastructure side as well.
Right. And that's we're playing a lot is how do we get the right connections, APIs, et cetera, in place to be able to facilitate, you know, those more automated workflows. But certainly an exciting time to be alive.
Jon Warden [00:37:39]:
It is. I think there's a fine line between, you know, good design and good experiences and functional requirements and functional specs. And I think what we're talking about is how do you harmonize that kind of side of the creativity and give the ability to be as creative as possible. Because no one wants dull and boring, right? You have to have some sort of level, obviously it's could be baked towards your Persona, etc, but as long as we can provide that type of experience in whatever method or medium or channel it goes out in, and you can benchmark that against clean code, clean ability, you know, AI driven functionality and so on and so on, then that's where we're trying to get our sweet spot of mixing these kind of tech approaches and creative experiences together. And if we can get that, that's kind of like half the battle. That's why. And we're always reinventing the wheel around that kind of stuff because we're always thinking about how do you evangelize this, how do we go on the journey? One of the things in a big corporation that's really important to bring to mind around this is we are thinking at the forefront of some of these things. But there's people in the organizations and there's many people, probably some of your listeners who maybe work at bigger organizations who aren't in that space goes back to my town crier piece.
You've got to communicate clearly, you've got to communicate often and you've got to have a vision and a mission and people have to be aware of what that means. It's really important that you get that into context because you've got to bring the mass with you on this journey. So at least they have insider context. Depending how farther down the line they go, they were aware of the proposition about how you're trying to engage this experience. That's a storytelling, Mario. You've got to be a storyteller in the world. I live as much as you are as a creator and a deliver over experiences.
Andrew Rohman [00:39:20]:
Absolutely. Again, it comes down to the human element. It's not just about solving what we're talking about and finally getting your tech stack in order. You, you got all that change to deal with. Just like if we want to prototype and code before we go to design. Right. All of these represent change. You've got agency engagements, you've got legal engagements that define how you work with your agencies.
Jon Warden [00:39:36]:
Right.
Andrew Rohman [00:39:36]:
So there's certainly a lot, a lot that comes with that. But again, I think as folks that have been in this space for a couple of decades or more, it's exciting because it feels like we're finally taking advantage of the medium that we've been working in for these decades. Right. And moving past a world where we design a poster like we're in Mad Men and then we go make it in code and then feel frustrated when it doesn't look the same. And somehow that's surprising. So I'm excited to be at the knee of this curve finally.
Jon Warden [00:39:59]:
And a funny story on that one is I was doing Dex for probably five years ago at GSK on design system management. And then I met you guys and went, oh, that vision I had of how design systems work is what knapsack are building. So we've obviously Maggie for a few years now, but that vision of like what I had in my head, it was great to see that there's another organization out there was trying to do, if not build, the same thing. So we're on the same vector, which I think is really nice in terms of how that story is going to evolve. So let's see where we go with it and see what happens. I mean, it's going to be an experience with all the AI stuff as well. On top of that, it's going to be a definite experience.
Absolutely. I mean, it's week over week at this point. It's fascinating to watch. Well, Jon, awesome to connect with you here. I love hearing your vision as always sharing it with our audience. I really appreciate you taking some time out and joining us on the podcast today.
Jon Warden [00:40:50]:
Yeah, thanks for the time. Brilliant.
Chris Strahl [00:40:52]:
Hey everyone, thanks for listening to another episode of the Design Systems podcast. If you have any questions, topic suggestions, or want to share feedback, go ahead and reach out to us on LinkedIn. Our profile is linked in the show Notes. As always, the podcast is brought to you by Knapsack. Check us out at Knapsack.cloud.
Andrew Rohman [00:41:05]:
Have a great day everyone.