PegaWorld iNspire 2024: Finding the Right Operating Model to Drive Innovation and Faster Time to Value

(upbeat pop music) - Okay, I think we're getting a couple, a little bit of a late start, but thank you all for coming. I want to, we're excited to share the stories of our panelists and how they're working through their COE journey on finding the right operating model to innovate and bring faster time to value. So my name is Brenda Looby. I'm with Pega and I'm the head of advisory services for North America. And I'm also gonna introduce or have my colleague introduce himself. - And I'm Jeremy Sirour, responsible for advisory services in EMEA. So, we like cryptic names. So what do we mean by advisory services? It's basically, we are helping our clients set up and run the COE and basically scale Pega footprint and what are the best strategies to build what you would call an innovation factory.

But enough about us. Why don't we introduce our panelists? - Oh me, okay. (Archana laughs) Archana Tikoo. I'm head of digital technology. I work with QBE. I've been with QBE for the last seven years, and Pega C4E has been in QBE for the last six years now. That's me. - [Sanjoy] I'll go next.

- Yeah, go ahead. - Sanjoy here. I lead the robotics and digital automation platforms in First Citizens. It's been five years plus on the COA journey. And you know, we are experiencing a rapid growth. It's been seven plus years, you know, seven BUs that we have been expanded and currently we are in expansion phase. - Hello everyone, Marta Tlalka. I lead the automation platforms and the automation COE at Edward Jones. We've been on our Pega journey for over three years now and we stood up the automation COE last year.

- Okay. So let's dig into a little bit of each of your stories. Or Archana, I know that you have been putting a lot of controls and mechanisms into place around your DevOps processes, and can you tell us a little bit about how you're doing that across different countries and you know, just what you have in place for that? - Sure. Let me give you a bit of a context of actually what we look after. We have 950 plus customers that we look after in European operations. We have 400,000 cases per year that we generate. And then we have eight environments that we look after. We have an uptime of 99.9% availability of our platform all the time.

Guardrail score this morning when I checked was 97, which is really impressive. So what we are doing is like, actually we are doing an upgrade in every year in less than four to six weeks. We do nearly 40 enhancements, 10 bug fixes every month that we do. And this is not possible if we do not have the right operating model in place. So we do have to have the right strong foundation for C4E. So it's all possible if you have the right governance in place, you have the right skills in the team, you have the right, when I say right skills, it has to be with the right roles and responsibilities. You commit to a success plan at the start of the year. And obviously you set up the right forums. We have user groups where we prioritize with the business, what's important, what's priority for the next releases that we should work on.

We have Pega championship calls where we listen to customers and ask them actually what are their pain areas. On top of that, what we also have is like actually we look at our quality assurance quite a lot and we have a 85% automation testing suit. We have automated our unit test cases and our deployments are all automated. So obviously that gives us a much more quality and efficiency in our processes. - So if I hear you correctly, you not only accelerated the pace at which you deliver software, but you also increased the quality. - Yes. - At the same time. - Yes. - Great.

- Thank you. And, and Sanjoy, I know that one of the focus areas for your COE is reuse. Can you talk to us about how you've enabled that and how you're tracking each of the modules or that are available for your teams? - Absolutely. So as I take care of Pega, and Pega is part of one of the major part of my portfolio. To set up the COE, it is very important that how quickly we can deliver the product, right? What is, you know, the minimum time that the product can be meet life? To do that, so let's take an example, right? If we go with any frameworks of the Pega, by default, 50 to 60%, you get directly outta the box from Pega.

Now, on top of that, you have common rules across the entire business, you know, and what whatever the bank that you are working for or any organization that you have working for, some of the things are very common, like the integrations, the branding, the security, the compliance, right? The connections within the internal bank, various systems and various APIs, which is external to the market, to the outside as well, right? So we have created individual modules for all of this. So once we get the framework right, we all add all of these different components, if you will. If you were to call it, you know, eight dot X or, you know, infinity version components, we plug those in and once we plug those in, the out of box framework, give us 65%, adding this 15% or 20%, like gives you 80 plus. And then basically you add a layer, right? Or a flavor to the various lines of business, which gets you the end product, right? But this is not true in case of, right, anything which you are creating new. Like say you get a new framework, then you have to kind of create a component for that new framework, which might take a bit of time, but once you create it right, then it is pretty straightforward.

Okay? Your answers has helped us, right, a lot to create such kind of components and framework. Thanks for that, right? And when we create such framework and go to, you know, our market, right, My main business stakeholders are all here, right, and we- (Brenda laughs) When we go to the market and you know, we go for a new BU, right, we don't say we will go live in a quarter or in six months. We give a date. Trust me, we give a date, right? Like 27th of July, something like that. And we ensure that date is kind of met because, as we are so confident, the structure that we have made out of our CA component, it took time, but we are at the stage where we can give a date and we can ensure we kind of match that date. Thank you.

- So you're saying basically you are able to reuse for every new project, 80 plus percent of what you already have, so you only have to develop the 20 remaining percent? - Exactly right. The example of this is, I can't take the name of the business unit, but we are delivering a fresh new product for a new business unit where we have created the framework in five months. We started in February, 27th of July, we are going live. - That's amazing. - Thank you. (people clapping) - Yeah, I have to say that's very impressive. We're very focused on reuse as well. You know, that was one of the main selling points for us for Pega and one of the main reasons we stood up the COE.

we actually, for the first time this year, we have OKR specific to reuse across all of our digital teams. So not just for platforms, but it also includes our platform teams. And we are very confident that we'll be able to achieve that with Pega and it should be pretty easy for us to show. Just last year we had over a million dollars of impact from reuse. So we're looking to continue to compound on that as we build more workflows. - That's an amazing stat as well, a million dollars in savings. And I think you're also have a big focus on citizen development at Edward Jones. Can you, I know there's lots of interest in, you know, the booth and other COE conversations, but we'd love to hear about that. - Sure, let's talk about citizen development.

Our journey might be a little bit unique because just to be completely honest, we've seen some things go wrong in citizen develop, with citizen development in the past on other platforms. We had instances where, you know, applications started to pop up that maybe our technology folks were not aware of. We had instances where somebody built an application and then they moved to another team and there was nobody to support it, right? So we said, "Okay, with Pega, we wanna do this differently." So what we're currently experimenting with and piloting is fusion teams. And what I mean by fusion teams we're pairing up professional developers with citizen developers to work together as one team to deliver an effort. So just a couple things to highlight a couple things that we've learned so far. You know, setting expectations is probably the most important thing. Again, nothing, you know, crazy there, but we have to make sure that we have all the folks on the right, on the same page as far as what are the objectives for the projects, what are the timelines, what are the outcomes that we want to achieve? Setting up those citizen developers, you know, setting them up for success.

So, you know, some things that we learn, they have to have the right level of training. So again, you know, sounds super simple, but we found out that they need Pega training. Yes, of course. But they also need training in other areas like just general awareness of our development cycle and agile training and things like that to set them up for success that we incorporated into our program. Again, just to make sure they have the right tools to be successful. They, you know, it works really well when those citizen developers are highly engaged. So we actually set a commitment. We realize they have a another job, right? This is not their full-time job, but they have to spend minimum of 20 hours.

They have to commit minimum of 20 hours a week to this program to be a part of a fusion team. And that gives them the space to focus on this, attend the sessions, do the trainings, all of the things. But when they are engaged, we found a lot of great benefits that come out of this, right? When they feel the ownership and they feel part of the projects, they're gonna be the first ones to go out there and, you know, tell everybody how great this workflow is that they're building and how great Pega is. And that just makes change management and adoption so much easier for us. We also have expectations for them to do light level support when the application goes live, right? So they've been under the hood, they know enough about this application, they've been part of the build. So when it goes live, we do expect them to be that frontline and provide support to a certain extent, right? That first level support.

So those are some of the things. Picking the right use case for citizen development is obviously huge and that's where the COE comes in. So we look at every intake to make sure and kind of decide whether it is a good use case for fusion team. And then just like with anything else, having the right level of senior leader support, right? Making sure that your technology leadership and your business leadership are aligned and on the same page. That's been huge for us. So just a couple things, but yeah, experimenting and piloting the fusion teams is where we're at. - Best of luck, Marta. (Marta laughs) Would like to learn more from you.

- Thanks. - That's awesome. And, like Brenda mentioned yesterday, I had many, many discussions with the Innovation Hub about, you know, we want to do citizen development, but how do we get there? So, you've given a number of advices, but what would be your key advice to, you know, make it happen and really make that transition? Or, what was your experience and the key advice you would share with the audience? - Yeah, I think the alignment is key. And then just getting started, you know, we didn't get it perfectly on the first try. We had to kind of level set and backtrack some of the expectations and making sure that we had the, again, the right training in place. But you just kind of have to get started.

- [Jeremy] So start small and- - Right. - Thanks. - Yeah. So we've talked about some of the technical aspects of a COE and Sanjoy, I know that you, it sounds like you have an amazing partnership with your business team that's here. What other things should people be considering as they're developing the COE that's outside of some of the technical pieces? - Thanks for asking this question, Brenda. So first I'll say like, you know, it's better to understand before we go to the technical from where we come from, right? So for FCB, anybody right, in FCB for us, the main primary factor is people, right? And culture.

So it doesn't matter technology, right? Or anything else. The first priority that you ask any FCB here with people and the culture, right? So we follow the people culture, then process, and then technology. Right? So as part of course the technology plays a part, right? But what we have done is as part of the COE module, we started slow, right? So we created technical experts. We then created SMEs or architects.

We created EY expert and then we created a COE leader, basically who kinds of manages that COE. Right? And there can be a fifth component too, where we are not yet, right? So if you go to the, you know, definition of like pure Pega terms, there needs to be five, right? But we started small. We are at the four and that's how we control. But the main key important thing that we have found out as the main product owners, right? So business, right? Fortunately I do have a great business, right?

Key stakeholders and everybody is so much into the product, right? And we do have product owners, not by definition, but they're playing a part in the game, right? And before even I go in or my team go in, they come back with a suggestion, let's say example of the knowledge ready today, the morning itself, like, you know? The team came up like how we can integrate and do all this stuff. So like, I need not to do like, even, right, explain, they are so smart. So they're coming up with the solutions and what we need to go for, right? So as I told, setting up a COE on the technical level, right? You need to have architects, SMEs, UI expert, cable expert, right, another COE manager, as well as relevant product owners from business who have skin in the game. - And that's also another discussion that I had at the lodge yesterday with people starting a COE and then maybe people in the audience who are about to start or just starting a COE.

Would you say all the roles you just listed are needed on day one or is that, or your experience was more gradually you introduced new roles or what is your experience on that? - Again, a great question. So it is a sign curve, right? So, you need to be patient, right? When you start something you might fail, right? I fail too, right? It is not that if somebody says, you know, they didn't see the failure, that's wrong, right? But the point is it take time but once you have a proper approach, you have a structure in place, start small. Maybe, you know, have an architect initially then followed by, you know, technical SMEs, products owner are the important then add UI expert.

Pega out of the box gives quite good UI, not like you need not to be like the cutting edge, right? Then you can add the UI expert and then followed by a manager on the top as a COE manager with respect to that COE is slowly and steadily. So you can start small, grow big, but very important thing is you need to have a proper vendor in place. That is very important. You need to have proper business stakeholders in place and your team needs to be strong, right? And should know what they're doing. Like, you know, make sure your processes are there. You have key stakeholders from business, you have key stakeholders from the various vendors that you have, right? And then basically, you know what basically you are doing and that should be the start, but it will take time and it's not like, you know, you cannot spin up a COE within a year or so.

For us, it took five years, as I told. It takes time but you should know the way, right? You should carve out the path how you need to do it. - So the next question is, and I don't know if anybody in the audience has heard any talk about Blueprint over the last. (people chuckling) I mean, are we doing this? (people laughing) So, Marta I know that at Edward Jones you've gotten, jumped in on Blueprint. Can you talk about just a little bit about how you're leveraging that? - Yeah, we have started to use Blueprint. We are excited about it.

My business analysts are actually right there in the audience. So they've tried it and they're really excited about it. So basically, I'll just share just kind of an example of what we're trying to do. We have a big effort coming our way that we're starting on to reimagine the experience of how our branches submit requests into our home office operations team. So we have 800 different types of requests that they can currently submit, right? So we're looking for a better way, a better branch experience to submit those requests, more automation as they submit them. So there's automated parts of the process currently, but we're looking obviously to automate as much as possible and we're looking for a better way to share status of those requests once they submit them, right? So they always know what's what stage they're in. So as we're doing all that, obviously we're trying to modernize all of the underlying technology.

So as we started thinking about it, obviously a huge effort, all kinds of different request types. And we said, "Okay, well how can we kind of get started and make that first step?" And it was about the same time that Blueprint became available. So we're like, "Okay, let's try this." Right? So we picked one specific request type and we said, "Okay, let's see what we can do with this." Pega led us through this. So they set up a workshop. We basically took an afternoon and said, "Okay, let's get Pega folks in the room, our COE folks as well as the business owners." And we did the workshop. It wasn't a ton of preparation ahead of time, we just basically got everybody, I know Rob did some, so I have to give him credit, but we get everybody in the room and we kind of just dove right into it. So Pega helped us draft the prompt. So that's kind of how we got started.

And you know, I've seen some prompts that are like a line, two lines. There's some that they showed us that were like a page long. Ours was about a paragraph just letting the blueprint know what we're looking for as far as the process. And then we pushed the button and it did the magic, just like you saw on the big stage. And there it was. So when we, actually when we first pushed the button, it came back with six different case types and our first reaction was kinda like, oh, interesting. Because we thought this was a simple process and the AI kind of broke it into six different case types. So we did look through those case types and it kind of sparked that conversation of, maybe we, you know, are we thinking about everything? Is this like why is it thinking there's six different variations of this?

So that was an interesting conversation, but ultimately we decided we can probably consolidate that into maybe one or two different case types. So we back, adjusted the prompt, which basically we told AI this, you know, do what you just did except consolidate that into one or two case types. And it came back and it did just that. I think we did maybe one more round of adjusting the prompt and then we had it in a place where we felt like that was a good start. And from there it just sparked a really good conversation with the business. We got a lot of good engagement and I think it was really a combination of, you know, a lot of excitement. It was like, this is awesome, let's see what it says. But also I think some freedom maybe. I don't know if that's the right word, but it allowed people to have an opinion and maybe look at the process and correct it because it was just generated by AI, right?

It kind of took that human element. Nobody had to put themselves out there and propose this, it was just auto generated. So they had more freedom maybe, you know, adjusting that. So we had really good conversation. From there, we had a couple follow up conversations with the business and then I think we're gonna bring the group back together, do another workshop and at that point we're hoping to have a pretty good workflow mapped out, a pretty good blueprint, so. - It's a great story. It actually ties back very nicely with what we heard this morning in the keynote, the user collaboration. They speak the same language. And actually there's also an element that you described that I liked is really this mindset shift moving to gap analyzation instead of starting from a blank page.

And, really Blueprint would allow you to have 80% correct and then think about your process instead of potentially shifting and lifting a process from one system to the other. You start from industry best practices. So, that's great to hear. - Right. And again, our business analysts point out that they don't necessarily have to be an expert, right, in that process, but they can have something to start with right away. So, that's a huge benefit. - Yeah so, I'm just gonna ask one more question before we open it up to questions if anybody might have some. We've heard a lot of great information. So Archana, I know we've talked about reuse, we've talked about having a roadmap for your COE.

you at QBE are actually starting locally, but thinking globally about how you're establishing your COE. Can you share some of that journey with the audience? - Sure. I mean, what we have done, I mean, it's in partnership with Pega what we are doing and thanks to Pega for that. So what we are doing is like actually we are setting up our customer success plan at the start of the year every year. And this year, for example, what we have done is like actually clearly articulated what should be our focus and the work streams that we should be working on. So one of them is like actually focusing on increasing our release cadence and then, you know, time to deliver faster. Then we have also looked at, you know, what's new, what we can, you know, deliver new, do a POC, do skunk work. Third one is actually leveraging the new hot topics emerging hot topics like process AI, process mining and as well as lunch and learns.

But obviously now as we are thinking much more, we have set it as a focus for European operations. But what we are at the same time doing is like actually sharing this customer success plan with our global Pega C4Es so that we can leverage it globally and share our lessons learned, I mean, and share our training plan. So it won't need to be more locally trainings. These will be much more at the global level and everyone gets, you know, a part of it, being a part of it. So, that's what we are doing. - Awesome. Okay. So, if you do have a question I would ask that you would please come up to the mic and stand close enough to hear. It's a huge room.

Are there any questions? - Hi. We're sort of early in the Pega, you know, life cycle, but I'm planning on starting a project next year that I'd like to start using citizen development for, maybe standing up a COE to help enable that. I'm trying to figure out how much our business analysts need to learn like App Studio nowadays versus just use Blueprint and then hand it off to developers. Is that lifecycle, that phase of the develop development different now with using Blueprint? - So I think it will speed up. I would say we're still early in the process. So if you're starting out, I would definitely recommend your business analyst and all the folks involved going through all the Pega Academy trainings and having all the right, you know, preparation to get started. The same with your citizen developers, right?

I wouldn't just kind of, I think Blueprint is a great tool and it will become more and more part of what we do going forward, but, you know, it's still gonna require the right knowledge from your folks. So, I would definitely not skip the training part. - Other questions? - [Audience Member] This is a little off, but can you guys talk about how you fund the COE like in projects in the COE? 'Cause we have, where I'm at, you know, if I wanna do any sort of project where I require another group like, let's say development from a COE, then I gotta go through this whole cap X process and it's just super painful and it needs to be like way out in the future and that just doesn't work with how software projects are developed these days or should be developed these days. So it's like, have you guys figured out how do you get money from the individual business lines to fund these things but when you're not actually in that business line? Have you guys run into any of that challenge or is that unique to me? - I can take some piece of it. I cannot comment on the monetary piece, but I can let you know that sometimes the COE is nothing but the technical team who is delivering in the initial phases.

Like, if you have two projects running, right, it is the technical team. If you have the two projects, then basically you create a smaller team, right? Maybe BU resources which will create the fundamentals for you, right? Initial phases of the COE are like, in my case, it was not funded, right, to be honest, right? Depends on various banks and how it is, but we wanted to do it, right? So we took out some BU resources, created some fundamentals like, you know, the branding piece, like, you know, it's straightforward, right? Compliance, speed, audit piece, those are standard across the bank. So you can start creating those as you get time. But you know, very valid question.

I would like to pass others for the monetary piece, but you know how I did it is by my main own interest and my team had the interest to create that framework. So we started very low, right? Very slow from the BU efforts. - I would concur with Sanjoy. So it's not, for my organization, we do not have a separate funded team as such. C4E in our organization is the same team with the same developers, LSAs, CLSAs, CSSAs. So I think what we do is like actually form a cadence, form a governance and have the right, share the principles that actually we have to leverage across all the projects that we are running in Pega. So, it's getting that time and it's very important that actually we set up us right. Actually, if you have the right principle set up, you don't have to give a lot of time in C4E.

What you have to do is like, you have to really set up a very clear guardrails around that actually what does our C4E do? So define those principles very clearly and get the right roles and responsibilities set in C4E. Like, what will be my CLS, LSA's role. They have to ensure that the licenses is intact, the guardrails for my platform is up and running, and then obviously what will be my LBS, you know, contribution to that. And, I think if you clearly define those roles and responsibilities, you don't have to really get a different team as such set up. So it has to be the same team work on as developers or architects in your team. I think that would be my suggestion. - Yeah, I would say that's probably a question that a lot of folks have. For us, we said from the beginning that our strategy is that the COE itself is not a source of funding.

We have other structures already in place of how we fund projects through our product teams. So we just kept that structure and then they come to us for help with a specific project. - [Audience Member] Marta, can you expand on what roadmap you use to get citizen development accepted as part of your SDLC process? - What roadmap? - Or strategy. - So I would say we're still figuring this out. (Marta laughs) But yeah, I think the biggest kind of lessons learned, and it shouldn't be a surprise, but it was like our technology folks kinda breathe this every day, right? They work in Jira, they have all the tools and then the citizen developers, they had to be on the same page to make sure that they could work together as a team effectively. So where we started is just education and training with them and just making sure that they even speak the same language, right, and have the same tools.

- [Brenda] And I think that you mentioned the intake process that you make sure that you're regulating what goes to a citizen developer versus what go, should maybe go to a more professional developer. Is that? - Correct. We are very centralized in that regard and again, we've had experiences where projects were just popping up on other platforms that we don't want to repeat, right? So everything has to go through the COE before any Pega projects get started. - So, you know, I was actually reading the title of this, you know, discussion again and I kind of got a question that is more relevant to me and my team right now. So if you're trying to deliver value and you know, if innovation is the key, if you take innovation as a key for driving value, innovation definitely means something new, right? Not the traditional things that we have been, you know, usually doing. So now technology, the way technology looks at innovation, the way business looks at innovation could be totally different because they both are like two different organizations in their own silos.

Business has their own lens of looking at things and techno. So what's the way, you know, I'm a Pega lead architect and you know, the easiest way that I could, you know, showcase an innovation, you know, to my business is to build a POC and then show it, right? And now my POCs might make sense to business or no because I don't understand a portion of business. The same thing goes right with them that they might not understand what technology strength they deliver. What's the easiest and best way to bridge that gap? You know, any techniques that you followed to bring both onto the same, you know, platform to accept an innovation and adopt a new idea? - I'll just say that actually what we have set up in our organization is the Pega championship calls that we have every month. So what we do is like we listen to our Pega, you know, who use our Pega applications and then we look at their pain areas. So we've got a list of actually their pain areas and then obviously any innovation that we are working on and actually whatever maps up to that, we go back during our Pega championship call and show them actually what we have come up with and tthat Pega championship call runs every month.

So it's not like as one off. Those are always in the calendar. And what we have done is like, actually we've got the right business stakeholders in there because obviously there are lots and everyone will have their own pain areas. So we have got the nominated team leads from each kind of business or team leads from their side. And I think when they come to us with those pain areas, we give them those ideas and obviously those some pain areas are some good initiatives or kickoffs for a good innovation, to be honest. So I think that's what we have set up but if Sanjoy or- - I will definitely add some points, right? And I'll give you example how we did it. Anything which is related to technology as innovation might not add returns of, you know, ROI for business. So how you do it is you clump the things that, we are starting a new project.

We just started a new project, a new framework, right, which is supposed to go live early next year, right? And as part of the project we did the upgrade piece also included as a project, right? Because just doing a Pega upgrade doesn't help business much, right? It doesn't give you literally anything, right? So going and asking for a funding just for upgrade doesn't, may or may not, right? But when you clump that with a project, right, and just say it is going to take, you know, one month just extra and you get five things on top of that by default, then that adds value. Right? So I do that piece. Second piece is you need to be very early adopter, right?

I found that out from my account executive, you know, Brent and Chris as well, that we were the first, one of the first, if not the first client to adopt Pega Cloud. Right? So you need not to be like, you need to be upfront, right? And ensure you are early adopter of the technology, but try to have the technical depth, I would say, you know, upgrade, you know, adding a chat bot, doing some POC, getting anything new, right? You need to add that as part of some program, right? So that it doesn't look like you're asking for a funding from business. Right? Right, so I do it in that way, right? And that have been kind of working for me.

The, you know, lowest Pega version that I have in FCV is 8.7.5. That is the lowest, right? We are upgrading it. Like, most of our instances are infinitely. We are just going ahead at 24.1. So, that's how we do it, right? We don't make anything as a technical depth. - [Audience Member] Yeah, so I think that's a smart way of doing it, right? - Yeah.

- You're giving what business wants and you're also adding what they didn't ask for. If they see value, then they buy it. So- - Yeah. That is like we need, right? So I cannot sell that, right? So, in that way. - Yeah, yeah. That's very informative. Thank you.

- Thank you. - Okay, well I think that we are about out of time. We, again, really appreciate you coming. If you've got other questions, I think we'll be available to respond. Otherwise, reach out at any time and we'll get you in touch with the people that you need to get your questions answered. Thank you very much. - Thank you. (audience clapping) (upbeat pop music)