Redesigning a Platform Experience at Scale
Redesigning a Platform Experience at Scale
This talk focuses on the evolution of ServiceNow’s platform by following VP Troy Azmoon’s journey from early days of an IT ticketing system to a world-class, comprehensive platform for enterprise workflow products. As the platform evolved, so did Troy – overcoming challenges as he and his team resolutely focused on understanding and then solving the complex problem of creating a unified platform experience at scale across an inherently chaotic, technical, and complex enterprise software environment.
Hello everyone, my name is Troy Azmoon and I'm here today to talk to you about redesigning the platform experience, at scale. Now, before I begin this talk, set the context a little bit, I work for a company called ServiceNow. ServiceNow is a large company that works with some of the biggest names in business, including fortune 2000 to digitize their workflows, and digitize their businesses through process automation, workflow automation, our customers include everything from Disney and entertainment to UnitedHealthcare group. Several multinational banks, as well as the Government's Department of the Navy, United States, and several others, quite a wide array of customers using ServiceNow to run their mission-critical business applications. Now, the platform we're going to be talking about redesigning is used by over 70 million people on a daily basis. These folks are often referred to as technologists, developers, knowledge workers, and service agents. But also, ServiceNow is used by hundreds of millions of people, including many of those of you that are watching this presentation, to do things like making a request for IT services, HR requests, customer support, or any number of activities you do as an employee, or even as a consumer, in your lives every day.
As for myself, I joined ServiceNow, in 2013, and came to the company as a product manager, later co-founding the platform product accelerator group, and then subsequently finding the design and research organizations, now today I hold the title of vice president of platform design. Now, for those of you that are curious about what a VP of platform design does, one way I like to think about it is, to imagine a city. When you think of a city, you have to think about the designer, somebody, or some group of people who did a master planning activity to create the city, where should it be placed? Is it close to water, and other resources, as well as the planning of the city?
So, think about things like infrastructure, roads, bridges, utilities, also commerce, so harbors, airports, areas of trade, urban development, downtown section services, like fire, medical, police, and security, and so forth and thinking about all this, as you design your city, of course, is very critical because it has to be architected in such a way that over time, your city can go from being relatively small to much larger and grow in a vibrant and scalable way and so, our story begins when the platform experience at ServiceNow, is a vibrant, well light city, but a classical city and our job is going to be to transform that experience of living in that city from one of classical antiquity to a modern major metropolis, where the existing inhabitants are not disrupted by this, what seems to be an overnight transformation to them and in addition, it also accelerates and attracts more population to this metropolis, and subsequent generations can continue building upon this new city of the future, if you will and it's a long story, even though it will appear to our users as an overnight success. It involves 1000s of people across ServiceNow, a million hours of investment, and a total dedication to realizing the dream of transforming the city from one of antiquity to one of the future. So, let's start this story, which for me, I consider to be the beginning and that's at the end of 2013 when I took a job at ServiceNow as a product manager, assigned to work with a small team of about eight engineers, led by the company founder and chief product officer frontlet.
Now, this team was located in a tiny corner, conference room. It can hold about 12 people total and we got to know each other very well. Hello very quickly and I can still remember that first week I was there, I was looking at the platform UI, which I was Product Manager for and I was a bit surprised because even in 2013, this looks a bit antiquated. To my eyes. It was hard to use a lot of whitespaces for that didn't need to be things were convoluted where they shouldn't be and I was concerned, I was very concerned for the experience as it related to people trying to be productive, working through their workflows and processes on this platform and a few weeks before prior, one of those engineers had been hired, and we got along pretty well and I kind of motion for him to come over to my area and take a look at my screen and we're only about 10 or 15 feet away from the founder of the company, he's happily coding behind us and I remember whispering to that point screen and I said, this is terrible and he looks at the screen kind of nods his head and he goes, I know. So, I look back over a half or so trying to keep this on the download here. Don't want to be upset Fred said how is it worth $10 billion and he shakes his head and he goes, I don't know and it was crazy because here we had this super successful company and there was this antiquated, out of date, hard to use platform interface to work with these automated digitized workflows and it really confusing me. Because from what I can tell, these people seem very bright. In fact, these are some of the smartest people I've ever worked with and that just didn't make sense. How could such smart people let this interface language be digitized in a prior age. So, I came up with three possibilities.
One, they fooled me, they're not very bright. It's some very clever intelligence Ponzi scheme, where everybody is pretending the other people are really bright. And, and I'm just in the middle of this crucible, and I'm being fooled.
Secondly, maybe I'm a super genius and maybe only I can see what's wrong with this platform experience and how it's getting in the way of productivity and it's very obvious how to fix this, and just no one else really gets it. Or see, I don't understand the problem well enough yet and within a few seconds, I came to the conclusion that the most likely answer was going to be C.
Because even though the interface was a bit of a challenge to use for regular people, there was a lot of things to like about the ServiceNow platform, and it was already very successful and well-loved by the companies that adopted it and I would come to learn that the way in which you were able to digitize any workflow, any process and bring it into the platform, and how easy it was with a low level of programmatic skill to bring those experiences into the platform. But those workflows in there, you can declaratively sort of edit and modify some of the experience with very little programming knowledge. We're strong attractors to the platform, despite the clue genius of the UI in some areas and so over the next three years, I would work directly for Fred Luddy, the founder of that company, and took me under his wing and became my mentor and that was an awesome experience. I learned so much about how to build products. Also, the platform itself, the data model, our customers, why things were the way they were? What could be better? What made us so great at what we were doing? And with this awesome team, this core team of eight or so engineers, we started to create a successful store string of products, starting with Visual Task Boards, which is a Kanban based experience on the ServiceNow platform, followed up by ServiceNow Connect, which is a real-time workflow, collaboration experience and then ServiceNow mobile experience, several different iterations of that classic UI, where we did make incremental improvements to it hugely valuable to understanding the way work got done in our platform and the way customers use it and extended it and so forth.
It's a great time, learned a ton, I built a ton of products all been very successful with a customer base and at the same time, the team of eight people grew by in order of magnitude and so, during that process of growing the team, I spent probably half my time coming up with products for the other half the time was hiring, that that core engineering team of eight folks was very serious about bringing good people in and that's just the way it goes when you have a core team of highly talented people that tend to attract more highly talented people and it became a virtuous circle. So now we have a team of about 80 people by the end of this three-year period and one of those individuals we hired. His name is Aaron has a close friend of mine. He was exceptionally talented in the area of concept design and Fred noticed this very quickly and one day approached Aaron and said, Aaron, what would it be like, If the interface to our platform was so simple, that anybody could use it to get their work done, anybody with no trainer? And one of the things that are so hard to do in design, and product design, is making something so simple.
So, your answer, well, I'm going to probably need at least three months to figure that out and we said, okay, that's fine, take all the time you need and come back to us when it's done and about three months later, he comes back with a functional prototype, which is a total, reimagines it, if you will, of the platform experience for ServiceNow and this is in, you know, early mid-2016 and it's a total departure, a lot of the UI is missing. It's not worrying about technological constraints, as we've implemented them, debt around the design of the current platform and existing customer needs, it's really purely focused on the simplest, most intuitive way to work with workflows that a regular person, little or no training, could experience to get their job done and so, it's in the context of this provocation, this catalyst and the string of successful products over the past three years, that I start to think about the big picture and seeing this in building those products and having those experiences with customers, and understanding our architecture and our technology stack. I decided I need to take a step back and really think about everything holistically to decompartmentalize, the knowledge I had gained, and the experiences of working with all these highly talented people.
And then went on vacation, I took a break, I went out into a very remote area of Baja California on a surf trip and I spend my days surfing and there's no electricity, or water or internet or any of those things out here and in the evening, I would just simply stare up into the sky and think about how everything is disconnected, yet connected and what would that mean, for the future of our platform? And how do we bring all these disparate possibilities together, I came back from the vacation and I presented to folks at ServiceNow, a product organization, a vision of the future, and outline 3 basic visions of the future where we would go one would be in 10 years, we would have hundreds of products. Today, we might have a few dozen, but we'd have hundreds. Another was that everything in the world would become connected. Everything from your doorknob to your oven, to your car, your phones, of course, every single thing would be connected like all those stars in the sky. Lastly, with everything becoming connected, there be much more data has been an exponential increase in data, that we'd be creating more data, more data becoming complex to us and it would all relate together to create even more complex than we ever experienced before as a result of all these dependencies and data being generated.
That means that ServiceNow is going to need a scalable platform experience to handle the multitude of products that have not yet been invented that would be inside this platform. It meant we would need a platform first approach to the way we design products so we could scale properly and lastly meant we really needed to rethink the way information interacted with the past was about going out and finding information. But now we had the burden of choice nearly upon us, information that you care about we need to come to you. By the end of this presentation, we were so excited about these thoughts. We were ready to do this, we were ready to reimagine the platform experience under these conditions and we started the project off, we took some of our development engineers or designers as a cross-functional team of product people and we started building out hypotheses testing kind of what we thought would be the future of the experience-based off everything we've learned and everything that we'd sort of dream about. But it wasn't to be because just a few months later, Fred Luddy, Chief Product Officer, and founder of ServiceNow, my mentor, would retire.
But a few months after kicking this project off again the design organization encountered more disruption. The decision was made to bring in a global head of design than a far more experienced than myself in this role to lead the team and as a result, the work that I'd commissioned was put on hold. People were reassigned to other projects and those are hard times. And, you know, I don't think you could follow me for wondering what I was going to do next at this point, because that is a tough pill to swallow. Now three times, first with Fred leaving. Second, with Frank leaving the team being disbanded and third with the new team that I built now being given to someone else to lead. So, there was some soul searching there. But the good news was, it was an opportunity for me to grow as a leader because I still had a lot to learn and this is a time of self-reflection, and developing self-awareness and ultimately, that experience was very valuable to me, and made me a more complete person and a better leader.
So, I'm happy to learn a lot. But it was tough and wasn't an easy experience to go through. So, with the emphasis now being focused on products versus the platform experience for the design group, and my attempts to make it a priority of the platform. Three design number one but so, basically getting pushback and a new leader coming in, we were now going to focus on the design system, which was great news, actually, because we knew we needed a new design system with this new platform experience. So, I was very excited by that. But it was a very challenging problem to solve because the design system needed to support the creation of many products on a platform. But it also needed to support 10s of 1000s of products that hadn't even been created or thought of yet. Right, so we had this moving target that didn't even yet exist in most cases. Furthermore, with an infinite number of use cases, any business could digitize their business and put it on our platform for any type of user.
Any type of use case, infinite uses, with a design system that can actually help you build products quickly, is a very tough problem to solve and the conventional approach is around being product-focused building and jobs to be done, frameworks simply would not scale to solve this problem. I'd spend some time thinking about this and have figured out a solution. But initially, the decision was made to take a more conventional, less risky approach. But of course, the less risky approach wouldn't scale. So, in my mind, it was going to be the riskiest approach for us to take a ball and eventually, things came to a head where the project was behind schedule and I was tasked to come in with a team of folks and lead that team to come up with a new way forward to design a scalable, platform wide design system that could create products quickly and like the Calvary coming over the hill, we got there just in the nick of time, and turn the design system project around, ended up solving for design system outcomes with 400% increase efficiency, meaning one engineer can now do the work of for one designer, the work of four could use the modern UI tech stack and it really helped my team's reputational credibility as well as my own and that was going to be very important, because the new leadership team, you know, I was relatively unknown to them and pulling off something like this at the 11th hour, with a low probability of success and ending up very successful with it gave me the capital that I was going to need to pursue bigger dreams, bigger accomplishments and outcomes at ServiceNow in the future because with this design system, we could do more than build products.
We could do things like build a new platform experience because through all this time, I think now we're at about four years or five years in ServiceNow is a journey with myself, at least, I hadn't given up the dream, I was still dreaming about redesigning the platform UI ever since that time out in the desert and coming back and saying this is what the future is, like, we need to prepare for the future. I was putting together you know, manifesto, sitting with UX architects, you know, in you know, in the time permitted to do so, there were still people working on prototypes, you know, on riskiest off-hours, like this thing had to happen, we knew it had to happen and given the design system done, and doing it in a way where the work was really recognized and appreciated was critical for the future success in the design system outcome is promoted and it was very helpful with my reputation in being able to drive us towards this outcome and finally, after all this work, and in the design, system did, as some of the things that I predicted would happen in the future started to happen, the future was manifesting itself over time.
As it does, we're building lots and lots of products with the new design system. But at the same time, customers are starting to get frustrated with the products and the product design. So, I was going out to see customers talk with customers and it was finding more often than not, it wasn't the products per se that were frustrating them but the platform experience of getting to the products, finding the products, working with the experience to these brand-new products in the context of an old platform UI and it was just driving people nuts, start documenting all this working with research folks on the team, making sure we kept a long list of grievances. People are frustrated with the issues so we can use those later to support our assertion that we needed to improve the platform experience at the same time, I've been working closely with the largest division of ServiceNow technologies, workload efficiency, and they wanted a unified experience for their multitude of products. They had a lot of products in their portfolio and the general manager came to me one day and said, do you have any ideas on how we could just bring all this stuff together I basically said, you know, it's funny, you should ask that. We've been thinking about this for some time and we do have some ideas. If you could support us in spinning up a team to solve your problem, your product unification problem, I think we can come to a solution that your customers will appreciate. So, he said, that sounds great and so just tell me what you need. So then, I went back to the team and I said, Guess what, everybody, we got a new project that I want to call fusion. It's about future-proofing the platform, improving our visual design, and bringing all the experiences together old and new, you know, all customers and so, we call this fusion and it's going to solve a problem that this technology workflows and division needs for the products, but it's also going to solve all the problems across the platform. And, you know, the GM said, yeah, that sounds good. I mean, as long as it solves our problem, we can solve more people's problems. That's even better. So, let's do that. So, I call up Aaron, the guy that had worked on the initial provocation, and I said, hey, I'd really love for you to get involved in this with us, because we're going to push the envelope, but we got to bring it all together, love your point of view, and talk to all the leaders of the teams said, you know, this is our moment. This is our time. We're doing this. This time is the time and so, we all started working together, the visual designers started working on new visual design ideas, and influences from film noir.
You know, lots of cool video games that people have been playing. There's an AWESOME movie. It's a masterclass in jewel tones in the mood for love. I want to check that out. UX architects a lot of whiteboarding burning through magic markers, you know, just humming and hawing, prodding and everything, researchers doing sentiment analysis qualified, and everybody's working in parallel, knowing that, hey, this is a huge possibility here to pull this off. You know, we put a lot of time and thinking into this over the years, let's make it happen. So a very highly motivated and excited team and dreaming about the opportunity to do this. It's a rare tree and so, we have all those streams going in real-time. But there was one problem as we made progress and this problem took us a while to figure out, and it was a navigator system, because our navigation system was this left-hand navigation, which you saw at the beginning of the presentation, and it was pretty clunky. But existing customers love this thing.
We call the left nav filter list. I think as a general rule, any navigation construct that you need to use five words to describe is probably pretty hard to use. But it was very powerful and as people got used to it, sometimes they tended to like it and our existing customers had gotten used to it, they built experiences on top of it. So meanwhile, you can see where this is going to go: new customers didn't like it, it was ugly, and it was hard to use one of the things that we're constantly complaining about. But with the new platform experience that we were imagining, we wanted to redesign, that we needed one solution to both problems, it's seen a lot of other companies create a classic view and modern view and I view that as a failure on our part, it wasn't acceptable to me to go down that road, I found it to be a lazy solution and it was going to cost us as well as our customers a lot of money to manage the change management for that over time and Aaron, and only, in particular, have a lot of discussions about this, he was of the opinion that we should just make a clean break with the past, do a top menu navigation system, I was much more concerned about the existing customers, having launched several products in the past and service down, made a few changes without asking customers, I'd been locked in conference rooms with angry CIOs having to explain myself, and I didn't want to relive that and I didn't want the customers to relive that. So, we're at an impasse, and the conversations got heated. This is a pretty good summary of this meaning here of the kinds of discussions we had on the individual with the handlebar, horseshoe mustache here and so, a lot of discussions, a lot of energy, we knew how important it was to make this thing look and feel awesome. But we also knew how important it was to not disrupt our legacy customers, making a change that impacted them could cost them millions of dollars in training and re-architecture their instances of ServiceNow that they've customized. So, we spent a solid, I want to say, two months, going back and forth on this problem coming up with a lot of different prototypes, trying to figure out a way to combine this left nav thing with non-left nav where it becomes a little less confusing. A lot of different ideas are thrown around. Here's one we're turning Off the top menu system, Aaron's like, I need this in my life and I'm like, well, we split in the left nav. So that's got to stay there. We weren't happy with any of these solutions, though, we weren't really making a lot of progress.
So, it's at about this time that I decided we need to talk to somebody who's a power user of ServiceNow, that really loves the left navigation system, that's a lot of time invested in it, and get their perspective. Figuring that, if we can convince them to move away from the navigation system of old, then maybe we can just forget about the left nav and undo the top menu system that looks so much more modern and cleaner. So, we head over to ServiceNow as the main engineering building, and we sit down with ServiceNow. CTO, Pat Casey and we have a few sessions with him around this tiny table, talking about the navigation system, see if we can convince him that figure might convince Pat Casey to move to this new menu system, we can probably get it by our users, and they will probably be fine with it overall. But the talks don't really go in the direction of okay, yeah, let's go with something new. He brings up a lot of good points that you can very quickly type commands and this navigator, that if you have 1000s of things in the navigation system, which some people do, but you can find them much more easily and after a few sessions with him, where he's extolling the virtues of the classic left nav list, UI filter navigator, Aaron and I are sitting across from one another, and I look over at him and he has a flash of insight and I have a flash of insight as well and we're just looking at each other right there and it's happening.
There's a spark that kept pushing on in, we figured it out and so, the meeting ends and we thank Pat, you know, because he did a great job of representing our current customers and we walked back to the design studio and on our way back. That's Friday afternoon and I turned to him and said, Aaron, I know you've got an idea on how to fix this problem. I've got an idea, too. But why don't you think about it over the weekend and I'll think about it and then on Monday, come to my office, and I'll show you what I've come up with and you show me what you've come up with and we'll just see what happens, you know, we'll have a synergy and just surprise each other and maybe come up with an even better solution. So, he says, okay, that sounds good. So, we head home for the weekend and sort of ruminate on it and then he comes to my office on Monday morning with his laptop, spins it around towards me, and plays me a prototype you put together, and while I'm watching it point on my whiteboard without skipping a beat and he walks over to it sees the exact same design on my whiteboard, we both solve the problem the same way and what it was so, a pinnable navigation system that allowed existing customers and users to keep the UI they've become accustomed to for as long as they like, while also allowing new users and customers to use a much more modern set of affordances. But the real beauty here was that it was dynamic, meaning any user at any time could change the way they work with the menu system bending depending on their need, or depending on the product they're using. So, it was really maximum flexibility, with absolutely no usage experience overheads. So, with this salt, we were super confident that we were ready to present our work not only to the GM of the technology, and workflow division, who was fully supportive but also to our Chief Product Officer CJ inside and staff saying, we have identified the problem and now we've solved and it was time to tell the world about. So, with that, we set the meeting with the Chief Product Officer and staff and I sort of set the conversation around future service. Now users will be able to use all these different utilities and new experiences, but it'll be a unified experience and it's going to be better than ever before.
You're going to fall in love with how nice this thing looks and it's going to be more productive and smarter and that was a bold proclamation and we're obviously making big changes overall, the look feels the experience and so to help everybody on that journey to go from thinking of this as high risk to low risk, because the biggest risk of all would be to stay static in a dynamic world. I gave some examples, Apple, and Microsoft, two great examples of companies that had to evolve their experiences on their platforms over time. We weren't doing anything radical here. We were doing what we needed to do to continue to evolve and innovate, to stay relevant. At the end of the doubt, everybody was on board and super excited about this new project. Nothing could stop us now. Wait, something could stop us now. Because just a month after presenting this demo and getting major buy-in from everybody to make it happen COVID-19 started up and that was disruptive, to say the least. Furthermore, just two months prior to this, CEO of ServiceNow, new one had come in Bill McDermott had joined ServiceNow and I also had a new boss myself, we had a new global head of design, joining the company and so, it's in all this sort of disruption, where one of my most trusted employees approached me one day and said, Troy, do you really think you can pull this off?
And my answer was, absolutely because in a time of uncertainty, certainty pays a premium and we were 100%, certain about what we needed to accomplish with this project, and why and I laid it out to everybody, up and down, left and right in the company, that the new platform experience would create a cohesive, more productive experience for users who would look great using the latest technology for our engineering teams. It will be personalized, and contextual, and will work for all customers, old and new, everybody comes along for the ride and since we knew exactly what we were focused on dealing with, and we knew how important it was to our users, there was no question we would make this happen and of course, after going through so many false starts and holding on in the stream for so long, I was willing to put it on the line to make it happen and so, from there, we mobilized ServiceNow at scale in a new environment in a remote world, that this project would come to be known eventually, internally as Polaris, our North Star, it will be the biggest project ever at ServiceNow. It garnered support across the organization, our best-programmed managers, our best engineers, best designers, best people in Docs, and the very best people, we're all clamoring to be part of this clear vision of what we're going to set out to do and why it was so important. Even people from that original accelerator team. Now, five, six years in the past that I last worked with them, were coming out of the woodwork getting involved in making huge impacts on building this experience into a reality and this was going to be a design-driven initiative and I put that in quotes. Because when we say it's a design-driven initiative, it's not just about the designer and I want to share with you really quick, this project plan I'd put together at the end of 2019, which I think really goes to show that it's not all about design here.
This was worked on with folks and engineering, and folks and product management and I created this not because I knew it'd be the final output, but to establish a bridgehead and talk with people in a way they understand. If you're going to drive, you have to drive in a way that has common currency. So, putting this in here, sequencing the work, assigning it to engineering teams at a high level, getting feedback from engineering and getting feedback from programmed management, product management as a roadmap establishes bridgehead for all the different parts of the organization that would be necessary to make the dream a reality and so, by the end of 2021, as we got into 2022, we relaunched our new platform UI and this marketing video is just kind of mind-blowing to me because here we are showing off a platform as if it's the coolest thing around, which is awesome to me. In addition to getting the platform redesign, we also took on work around creating several new applications over two dozen new product applications that integrate completely with the new platform experience and altogether, released this at the beginning of 2022 and we call it the next experience.
It's a new platform UI experience from ServiceNow. customers really love it. We're very impressed by it. Here are some of the things customers have been saying. They're in awe. They think it's a huge improvement. They want to throw money at us because of the platform update. How often does that happen? And this last one is Australian. I'm told it's a very good job to keep up the good work. So that's the story in a nutshell, as far as the takeaways. You know, I think it comes down to just a few salient points really at a high level. The first one is genius. Ideas, just one little piece, 99% of this is hard work, the willingness to take your hits and learn along the way. And, you know, keep making progress. Platform innovation is a marathon, taking a good product idea, and making a great product is very hard doing the same thing with a platform is harder yet. Great outcomes are a team sport, you can do good things by yourself. But if you want to achieve great things, you've got to work well with other people to deliver, and of course, that requires working with shared currencies, you need to build those relationships, see things from different points of view, if you're going to create outsized, outsized outcomes and lastly, every failure is a step towards success. You know, when we set out to build this modern metropolis, it wasn't built out of aluminum alloys, we use a special material, which we'd found in abundance over the course of this project and that material was our failure and it was through all those failures, all those learnings, that we were able to take that material and transform it into the success that we've realized within next experience UI. So, I hope that this talk was useful to you and your efforts, and I wish you well in creating product and platform experiences that your customers love. Thanks for watching.
Got a Question?
More like this?
Mon, May 23, 1:00 PM UTCA Comprehensive Guide To A Successful Design Delivery While Building A Design System
Product Design Manager, Design Systems, Disney Streaming