Modernising A Legacy Platform

Alexandre Ramos at UXDX EMEA. Video: https://youtu.be/4f9OXS44qrc

Readable transcript: edited from the recording's captions for readability (fillers and false starts removed, punctuation and section headings added). Wording is the speaker's own. Timestamps are positions in the video. Names marked [?] could not be verified against the audio.

Why Liberty Europe set out to modernize

[00:00:00] The topic today is how do we modernize. I know that we stopped saying words like digital transformation, so I won't use it, and thank God the topic is modernizing a legacy platform. I'll start first by showing what Liberty is.

[00:01:21] Okay. I wanted to start this little mini story saying that what I showed you is what Liberty Europe is, and what we embarked on. Just for contextual perspective, we are four countries, and we have technology everywhere. We came together in what we call the market and embarked on how do we modernize our legacy, and that's the story I'm going to tell you.

[00:02:06] Obviously in here there are several things that have been achieved while we embarked on this amazing journey, and I'll just say, because it's going to connect to the journey, the top left corner: being nominated in the four countries a Great Place to Work in 2020 and 2021 is for us a very proud moment, because it's about our employees.

[00:02:38] Starting with the why. The why is what grounds us most: we need to know why we want to modernize. There were several asks: being fully digital, simplify, become easy to interact with, which is a dream for every company. Enable innovation that we don't know where it's coming from, how it's coming, when it's coming, so that was a powerful ask. Ensure scalability and modularity, fast time to market, fast time to value, being very consumption based. And make sure we don't modernize to then further down the line have those moments where we need to refresh. We had this dream of what if we were always up to date. That was the why.

Building a new digital ecosystem on the side

[00:03:45] How did we do this? We decided to build a new digital ecosystem, completely new, on the side, where we took some critical decisions. We have to have product and pricing capabilities that are state of the art, because that's where we differentiate: that's our products and services. We have to have modularity: pick and choose whatever you want. Completely cloud-based, completely public cloud-based; there are no data centers in Europe. It had to be 100% API driven.

[00:04:30] No connections to the current. I'll reinforce this one: no connections, not even the smallest one, and you can now imagine the amount of pain that brings. But that was, and is, a principle. A greenfield approach, thinking new.

[00:04:58] Let's not chase the state-of-the-art solutions for everything. Let's go for the next level, for something that will allow us to do the business but can grow with us. We don't need all those bells and whistles. We think we do, but then time tells us that sometimes we don't. Out of the box: if we decide to buy a solution, to consume a solution, we have to leverage it, and if that is not adequate we'll change it. It's modular. Business processes are the ones that came with this out of the box.

[00:05:46] We have to have data supporting all this ecosystem, which we do. And we decided to give it to someone else to manage the existing, because we need to focus on this more than ever.

The challenges of the journey

[00:06:06] What about the journey? I really like this quote, because I can tell you, and many of you will know: if you believe it, it's halfway. But the journey brings challenges. For a company that is operating under this great Liberty Mutual, but specifically in Europe in four countries, there are cultural challenges. There is a complete change of processes and technology. Imagine anything end to end for any insurance process: it will change, so that comes with pain.

[00:06:52] People's mindset. It's painful for a human being to change, and we see this in this journey. It's very painful, not because people don't want this, just because it's difficult. We go from expert on something to very basic on something, and that is very challenging to overcome.

[00:07:14] The MVP approach: how many of you feel this? MVP is wonderful, but the business sometimes, and even the technology, get frustrated, because when are we going to have that thing that we really want? It's not easy to capture. It will evolve, and we'll stop at a moment that is different from the one we thought before, because the world is completely changing every day. This iterative prototyping: let's put it in practice and then let's iterate.

[00:07:51] There will always be business pressure, either from the markets, from the competition, or internally. The duration is a challenge. There's lots of frustration for A, B, C, D; it doesn't matter, we have to build this resilience. There are lots of opinions. Everyone has an opinion: "Oh, I think we should," "I think, I think, I think." We need to be able to manage this. We will find issues, and it's a tremendous test for an organization, a tremendous stress test. If we go through this, we'll go through anything.

Foundational principles

[00:08:36] What were the principles? The principles were foundational. We will do this incrementally and we will use an agile mindset. We clearly acknowledge that different people value different things. That doesn't mean that they disagree with whatever. It's that some value the customer more, others value efficiency more, others value society more. There are so many different value perspectives, so we need to bring them in, but let's not try to make them all value the same, because that won't work.

[00:09:18] We have to be, more than ever, an active listener. It's not only listening to managers; I would say it's more listening to the people that do the work every day, that are picking up a call from a customer, that are managing a claim. Everyone has a good view of what is really needed, so active listening and collaboration is key.

[00:09:47] We need to make sure that we bring our people, our employees, on this journey. Not "in addition we'll bring them"; no, they're the ones. They are the ones that do this, no one else. It doesn't mean that we don't want to have support, but it's not the other way around.

[00:10:07] Massive reskilling and upskilling, because people have been doing something for 10, 15, 20 years, and then they have to change to do something different, in a different way, in a different technology, in a different process. They need to learn, and we need to acknowledge that yes, there is a ramp-up period.

[00:10:28] Very light but effective risk management, because this transformation brings risk acceptance, but we need to know what risks we're taking and accept them, monitor them, evaluate them and move on, not as a blocker but as an enabler.

[00:10:50] This doesn't mean that prior experiences are not good. All experience is useful, but we cannot apply the same medicine that we applied ten years ago or five years ago and expect the same result. Use experience, but have more focus on initiative, on willingness, on bravery, on risk-taking, on "I don't know how this is going to play out, but it seems a good idea, let's go for it." Be brave. We have to be very brave.

[00:11:29] And be supportive, and this is for managers, leaders, senior leaders. We have to be super supportive, and we need to give the accountability to teams to make it happen. We'll fix what doesn't go right, we'll fix what can be fixed, we'll change what needs to be changed, but we delegate, because this cannot be achieved by an umbrella of a few folks. It is achieved by the organization moving to the same place.

Leadership, architecture and quality

[00:12:04] This takes me to leadership. You must have complete alignment, dedication and support from the senior leadership. There's no other way, because this is so painful, as I mentioned before, that it only works with an immovable, supportive team: "Yes, it's a pain. What can we do to change it, to improve it? Let's go for it." "Oh, this didn't go well." "That's fine, test and learn."

[00:12:40] Common goal: regardless of who you talk to, everyone knows the goal, everyone knows what it is. It's very clear. More than ever we need to lead by example. We need to be super transparent: "Yeah, we have problems, and this is the problem. I don't know the solution, but I'll figure it out; in one day I hope I have it." This is the level of transparency. Let's not go old school: "Oh, I'll figure it out, and then I'll say we had a problem but we solved it." No.

[00:13:13] Architects and security have to be fully dedicated, fully engaged, like anyone, but these roles are critical. I'm talking about business architects, I'm talking about technical architects, I'm talking about security that goes beyond what level of defense we're applying now, looking at it in a holistic way. Because this is cloud consumption based, we have to have architects and cloud engineers that are constantly seeing what else we need to do.

[00:13:55] Releasing. Everyone knows the stories of big companies out there that have done an amazing job on releasing things every day, every hour, but it's not easy, so let's put on the agenda that we need to have that under control. Forward-thinking, innovative. And we have to deliver with quality. Quality is non-negotiable. That doesn't mean that MVP1 is poorer than MVP2, but it means that we're always aiming for good quality.

Imagining time, experiments and real retrospectives

[00:14:32] For me, this is one of the most important aspects. We rush; we rush for everything. We know that we have to have thinking time, but then: "What, are you going to spend two days thinking about that? Come on, can't we go directly to execution?" No. We need to have imagining time, because if we want to fulfill the ask of being innovation ready, being ready for the future, you have to imagine what the future will look like. You need to find that time, engage with the right people, and allow them to think. Allow them to think.

[00:15:19] Not forgetting that it's not spending weeks or months thinking about it. No: let's take a couple of days to forward think, get out of the transactions, then go off and do it. Put that framework in place.

[00:15:35] Experiment. We cannot be ready if we don't experiment. And an experiment, like the word says, means I don't know the outcome. I guess the outcome, but I don't know the outcome. So let's not make experiments so solid that they take so long because they need to be the real thing. Put the experiment mindset in place.

[00:16:03] Collaboration I spoke about. Let's do imagining not only with a few people, the architects, the product owners, the managers, the product managers. No: bring people from all levels, from all teams. They have the right way of thinking, the way of imagining. Bring them in and enlarge the vision. Always be consequential, because it's about the results as well, so we cannot be wondering and not delivering.

[00:16:52] And the last one: avoid following a methodology just because it tells you to. If you are thinking agile and you're applying the program increments and you're doing sprints, "Oh yes, we have a ceremony of retrospective," and then they become just another check in the box. No. Practice real retrospectives. Bring the actions to improve the next day. It doesn't mean that you're going to improve everything, but build that mechanic, because if we want to get better we need to learn from the things that we don't do as well as we would like.

What's next: pluggable, data and cloud

[00:17:40] What's next? The next is very ambitious: having all kinds of ability to plug, because everything is API driven, it's pluggable, whatever is used and consumed. I can give you that: in this modernization we've replaced core components that were unimaginable to replace in the past, and we replaced them in a couple of sprints. It is challenging, but it's possible. You just have to have the right architecture.

[00:18:20] Data. Everyone says data is the new world. Yes, data has been around for many years; it was oil before. Now the thing is we have technologies that allow us to leverage that, but you have to have a sane architecture that actually enables that, and we think that's in place.

[00:18:43] And cloud. Cloud is the only way for us not to be limited. Yes, there's a sense of security, but I want to say that if we as any organization think that we have enough resources, enough knowledge, enough time to prevent, or to increase the level of security, we can only be mistaken, because the cloud providers do this every day, hundreds of people. We need to have the right frameworks, methodologies, controls, all of that, but for us cloud is the only way to be elastic and scalable.

[00:19:39] And somehow, I know this is contradictory for many people, it's predictable. We can be predictable, because we know how much it increases based on whatever we grow or reduce, so the level of predictability is very high. And last, there are a number of technologies that can leverage data, but you have to have the right architecture.

It's all about the people

[00:20:09] I'm almost finishing, to say that this is what our customers, providers and partners want, because in insurance the value chain is very big. They want better, simpler, faster, and they cannot be precise: "Oh, I want it this way, this time, this manner." No. We need to build this future without having those boundaries.

[00:20:42] And to finalize: it's all about the people. It's all about the teams, the engagement, the willingness to leave the business in a better shape than we found it. It's about interacting, collaborating, but it's all about people. Let's not go the easy way, well, the perceived easy way, which is to put some technology in place, and by the way do some RFP to pick the right one and commit that the right one is going to be there for 20 years. There's no such thing anymore. We commit for two years, three years, then the world will change and things will have to change.

[00:21:34] If we manage to keep this constant up-to-date mentality, we will always be there as things change. But in the end, you can only achieve a modernization with people. Take them on the journey. Tell them that we don't know all the answers, but we'll figure it out together. And with this, I don't know if there's any room for questions.

Q&A

[00:22:06] Host: A question from John: could we get some practical examples of product risk management from you?

[00:22:13] Alexandre: Thank you for the question. One of the things that we've done is we iterated; I think we are in version three of this matrix that assesses risk. The normal matrixes are based on risk exposure and then impacts. What we've done is made it in a way that risks are pushed to the foundation of the organization, to the squads. They have to manage the risk. They cannot only share the risk and say, "Oh, we're exposing this risk, therefore someone, God knows where, is going to accept it." What we've done is made it so that risks are the day to day.

[00:23:09] People identify, for instance, "We're going to do MVP2 of a certain capability, and that capability is short. It's not automated, it has some manual workarounds, so when the customer calls, I need to do something in the screen, I need to send an email to someone, because it's MVP2." That is a big operational risk, because what about the controls and all of that? Yes. So we accept those risks, because we know that as we do the next program increments we're going to prioritize what's missing there, because it's really more important.

[00:23:54] As you build the risk management into the framework of prioritization and execution, you're bringing risk to the day to day, and not just something you have to fill in and show every month and say, "Oh, this risk has to be extended," blah blah. That's the way we do it.

[00:24:12] Host: So it's also an iterative approach to risk management as well?

[00:24:16] Alexandre: Yes.

[00:24:18] Host: If it answers the question for me, I hope John, you're happy as well. The next one: what are the biggest challenges in the insurance industry going forward?

[00:24:29] Alexandre: Well, there are so many, but I would say first the pandemic is not gone. Even if I'm in a lucky country, I'm in Portugal, so I'm in a lucky country that is getting better, the pandemic will bring challenges to the business. There are constraints in vehicle sales, there are constraints on manufacturing cars, and if you want to buy something electronic, the likelihood of that being delayed is high. All of that affects the economy where we operate, so that's one of the challenges we see, and it's not going to go away, in our perspective, in the coming months or year. It's going to take time regardless of what the governments say. That's what we feel.

[00:25:23] The second challenge is that customers, with this pandemic, realized that there's a world out there digitally that they can use and they were not leveraging it. So how do we, as a society, and therefore us as an insurance company, blend? It's not about having places where people go online and buy a policy. Products and services have to be blended into the day to day of the customer. How do we do that with partnership? That's why the partnerships, the open concept that we're trying to create or enable, which is this idea that we all need each other and we should leverage each other in terms of "I do this well, so I'll be in that moment of the journey adding value," and the customer perceives that as value.

[00:26:22] That's one side. The other side, which for an insurance company is still a reality, it's a fact and we embrace that, is we have partners, what we call intermediaries, that do the business for us. We need to bring them on a journey of digitalization, of being digital savvy, to actually build that ecosystem with us. Those are the biggest challenges that the pandemic has shown us. We've been saying that for so many years; now it happened, and by the way we all survived. Well, not all survived, that's not a good word, but most of us survived.

[00:27:05] Those are the main ones. Then there are others, which are the trends: the fact that we have to be sensible about environmental challenges, the consumption-based usage. Kids don't want to buy a car, a house, so how do we prepare for that? And I have to say that this ecosystem, this modernization, was also to enable whatever that future is going to look like.