Good morning, good afternoon, good evening, good whatever time it is wherever you are in this amazing planet of ours. I am Malcolm Hawker, and I’m the host of the CDO Matters podcast. Today, I’m joined by Rebecca O’Kill. She is the chief data and analytics officer of Axis Capital. Rebecca, happy Friday. Welcome.

Thank you very much. Happy Friday to you too.

Brilliant. Well, today, we are going to talk about navigating a data and analytics function to this amazing future state. Maybe it’s AI, maybe it’s adaptive, maybe it is who knows what the future looks like, but what we’re gonna talk today about today is how do we adapt our operating models to where we need to be. Things are changing in our world.

We just did the prophecy data summit yesterday. We were talking about all the changes. We’re talking about the noise. How do you navigate that noise?

We’re gonna continue those conversations today on the CTO Matters podcast. We’re gonna talk things about, like, governance, talk about the team.

We’re also gonna take some time to celebrate successes. I don’t think we do that nearly enough in our world. So if these things interest you, if you are a data and analytics leader wondering about how to adapt your operating model for this brave new world, stick with us. It’s gonna be a great conversation. So let’s start with the fun stuff first. Let’s let’s start with some successes. Rebecca, what can you share that you’re particularly proud of that you’ve done over the last year that that that you’d love to share with some of the listeners and viewers?

I mean, it’s it’s a great question. And I think it’s like you said, I think it’s really important to take the time to reflect on the things that have gone well because I think quite often in this role, in data roles in general, you can end up feeling disheartened a lot of the time about things that haven’t gone well. For me, a lot of the successes that have really come up in the last year have been around kind of building building relationships and changing people’s perspectives about data. And I think we’ve we’ve gone through a bit of a process whereby we’ve had skeptics, and we’ve managed to bring those people to become to become allies.

And for me, those have been some of the more successful things that I think it’s great to look back on and and reflect and feel very very proud of because I think that those are the sorts of successes that then have knock on effect and enable us to to then have more successes in the future and to gather more men momentum and generally to be much more successful as a DNA organization.

Okay. So that necessarily begs the question. How how were you able to turn the skeptics into believers? Is it just charm? Is is it more than that? What do you think were the the secrets of success in getting people on board?

Again, great great question.

And I like to think charm is part of it, but definitely doesn’t doesn’t solve the full problem.

Really, our focus has been and our focus continues to be as a kind of data analytics team is on delivering value. It’s about understanding what the business needs and what our business stakeholders needs and then working with them to deliver it. So I would say the places where we’ve had a lot of traction and where we’ve made real progress in that area is where we’ve spent time, first of all, I guess, spending time with our business stakeholders, understanding what it is that they want to use data for, where they can get more value from data, where they have kind of analytical or AI questions that that well, analytical AI solutions that that could be used to solve some of their problems, and then partnering with them to then work out, okay, what does that mean we as a team need to be focused on, and how do we build all of the things that are required to get them the thing that they need?

And so in places where we’ve really kind of managed to win people over, it’s where we’ve listened, we’ve understood, and then we have created either the beginnings of a solution or a small initial solution, and then they’ve started to see the light and think, oh, hang on a second. There is value here. These people understand me, and if I work with them, then we can get we can get to a to a good place.

Okay. So who are the people doing the listening? I I’m I’m really interested in this because a lot of us are investing in this idea of data products. Right?

With the whole idea that the product is kind of the output of that listening process. But I’m I’m interested in understanding who is doing the listening. Is there or do you have analysts? Do you have product managers?

Do you have you assigned people like data product owners? Who are the folks that are directly inter interfacing with people on the business side, gathering requirements, understanding business processes? What is what does that role look like in your organization?

So this is something that I would say has been evolving and is still very much evolving.

So in terms of who’s done listening today, I spend a lot of time out there with the business listening and translating to my team, and I think that’s an important part of my role to kind of build that to build those relationships and spend time with the the business leaders.

And alongside that, in terms of the people that are on the ground actually understanding the more specific requirements, I’ve got data insights lead at the moment who who are kind of hybrid people, I guess, in that they are they understand data, but they also have some kind of business experience, so they can kind of do that translation piece.

And, yeah, we’ve got we’ve got a few of those. I wish I have more because I think they are quite difficult people to find. But a lot of those people actually a number of the a number of the people that I have that are very successful in those in those roles are people that I’ve nicked from the nicked from the business. So people that have been within business functions and have a interest in data and therefore have ended up moving into my into my team.

Really? I love it. And there’s absolutely nothing wrong with that. You know, developing people from the business side, I mean, that’s something is when I was an analyst, I talked about all the time as a best practice because ultimately, those folks should be domain experts. That’s the word we’re using a lot these days is, you know, these should be domain experts. What about what about prioritization? What are you what are you focusing on and and how do you make decisions related to that?

Prioritization, again, is, I think, one of the biggest challenges of all because everybody wants stuff. Everybody wants stuff now, and everybody’s desires are the most important things as far as they’re concerned. I think we’ve come from a place where it was all about whoever’s you know, who knows who or who who shouts the loudest.

I would say that we’re still very much on the journey in trying to work out how we can do a lot better at prioritizing.

But for me, there are a number of things that have to be factored in when you’re thinking about how how to prioritize. The first thing is really understanding the business value and being able to kind of estimate what it’s going to deliver in terms of value for the organization, which can be difficult because a lot of the time, you know, it’s we don’t have necessarily the right frameworks in place to be able to say, right. Well, it’s going to enable ten million dollars of value or top line, bottom line, or whatever. But we have to get a bit better at that, and we have to start sizing things from a value perspective. We’re on a bit of a journey there, and I think we’ve got a lot a lot better than we we were.

The other thing is so we’re I I guess, just to step back for a moment and give a little bit of background about the journey that that we’re on.

So, yeah, Access Capital insurance organization, we sell lots of different products in lots of different geographies. And, historically, the way that the organization has kind of come together has been almost like lots of little separate businesses. And so our data estate reflects that. We’ve got lots and lots of different repositories of of data, and things are defined in very, very different ways.

So what I’m kind of I’ve been asked to do, I guess, over the next few years is really to try and consolidate that and to create a wonderful data platform that serves all of our needs and enables us to do all the wonderful things we can’t do today.

Prioritization is incredibly important because that is a huge task, it’s not something you can just say, Right. Okay. Let’s take all that data, put it over here, and our problems are solved.

Sequencing is actually really important because what I want to do is align kind of the things that the business needs with with kind of then bring doing all of the foundational stuff alongside it. So we understand the needs of the business. We build out solutions for the business, and and alongside that, we’re bringing all the data into a new platform. We’re applying governance. We’re doing the definitions. We’re doing all of the good stuff that you need to do to have a a really fantastic data estate that’s ready for AI and all the other stuff.

So so, yeah, there is there is definitely also as well as kind of understanding the value of the the business requirements. There’s also a logical sequence to doing that so that you get you you build it up in a in a way that that gets you to the right place. So particularly when you think about insurance, you know, claim premium policy data, data around policies is one building block. Data around claims is another building block, but you can’t have data around claims until you’ve got the data around policies because claims are attached to policies.

And so there’s a logical order there as well. So we try and take into account the the the value as well as the impact it has on our wider data strategy.

So from a priority perspective, those same folks I assume that are doing the listening, right, that are understanding these, that are understanding claims processes, you know, building and writing new policies. They’re trying to understand all of that stuff. They’re also trying to understand the value equation, it seems like to me. In putting all that together. Do you have, like, a separate PMO type organization internally, or is it is it you that are managing the road maps?

Yeah. And, again, that’s the thing that that we’ve been evolving over the course of the last year and a half.

We do have a a PMO organization, but really we’re shifting to do much more agile delivery. So therefore, we’re we’re doing that ourselves, which essentially means that those people that are doing the the designing piece then have they’re kind of almost leading delivery squads, and we have all the right kind of skill sets working within that squad so that we can do the building or the architecting, which is very, very important. The building, the the governance activities as well, kind of liaising with the business to to explain to them what their what their roles are, gathering definitions, and then kind of creating the solutions that’s on top of all of that.

Got it. Got it. Excuse me. Still my first cup of coffee here on the East Coast in the US, so clear in my throat here. My apologies.

So when it comes to data science, do you have a separate data science team? If yes, is that under you? Is that under another group? What does that what does that look like at Axis?

So, yes, we do. We have I have actually, in the last month, had a new head of data science and AI delivery is the job title. And and so I have within that team a number of data scientists and AI engineers, and they are building out data science and AI solutions support for business needs.

Okay. And that’s still entirely under your your remit. So Yeah. Okay. So when the business is asking for cool, new, fun AI stuff, that is squarely looking at you.

It is squarely looking at me, which has become quite a challenge because everybody wants cool new AI stuff.

Oh, yeah.

Not necessarily the understanding of them of exactly what it is that they really want and exactly what it is that it will deliver. So we have recently kind of evolved our governance around AI. We’re implementing what we’re calling AI champions, so every area of the business has somebody within their business area who we’re training up and will be like the conduit for all AI stuff. Their their it’s their mandate to upscale their organization around what AI is, what’s available to them, what we already have internally, how we’re thinking about AI strategically. But they’re also then that kind of, initial filtration point when it comes to ideas and things that they might want building. Rather than us getting all of the millions of thoughts that people have had around potential AI solutions that might solve a problem, there’s somebody who can bring those together and help shape them a little bit before they then come into the AI delivery pipeline.

That’s interesting. A couple of years ago, threw out the idea of this role called an AI ambassador, where I saw this need to go to go out, right, and and to and to work proactively and to, like, educate and help the business understand, hey, this is a lot of noise, but this is where you potentially could be focused, the right use cases, the right AI for the right use cases, for the right outcomes. It sounds like exactly what you’re trying to get done.

Yes. Yeah. That is exactly what we’re trying to do. Okay. Because I think, when I think about AI and the future of AI, it’s not going to be it’s just not I mean, it’s already getting to the point where it doesn’t work to have one central team who’s the, you know, the people that know and understand it.

It’s going to proliferate, sorry, across the whole of the organization, and we need to change the culture of the organization so that they’re ready to embrace that. I think having that more federated model where you’ve got somebody within each of the different business functions who can then take what we’re learning and developing centrally and then help us to get the value from that across the organization will be hopefully a big step forward. And I’m hoping that those AI champions will then foster kind of these little groups of mini AI champions that that come together and discuss these things.

So it becomes it it then helps that cultural shift.

Well, it’s not unlike governance as well. Right? I mean, that’s what many of us are trying to do just broadly from a governance perspective. But I there’s something interesting here.

I completely and totally agree. And if and if and if you’re not helping the business along from an AI perspective, they’re just gonna go do it themselves. Yes. So so either we can we can be invited to the party or we can’t be invited to the party, they’re just gonna go do their own thing either way.

We’ve we’ve seen this over and over and over again. Arguably, the democratization of IT was very much a function of that very ethos.

Yeah. However, I see I don’t know. I’d love your opinion on this. I do see so there’s certainly a desire from to to federate, and I think that there’s a necessity to federate when it comes to AI governance as well.

But at the same time, a technology and infrastructure perspective, I do see attention there in towards more centralized approaches. Right? So so I do see a little bit of attention there from a technology perspective and a process and operating model perspective. Are you seeing the same?

Yes. Definitely. And and that, we we want to keep as central as possible for the for the time being. So, actually, when it comes to the building, that sits that sits centrally.

And the there’s still very much kind of approval process. It’s not like they can they’re just free rein to go off and, like, license a lot of tools and do things their own. Right. They don’t have the the the authority to do that or the but there’s no budget that sits with them to go and do that.

So it’s more about fostering the ideas and helping to shape those ideas and then bring them into a central team for delivery. And then we have kind of our strategy around, you know, our general purpose tools that people can go and use, and we can train them up on those. We then have kind of our okay. This thing, you know, we we can build individual solutions for.

We then have kind of our strategic layer where we’re trying to build out AI capabilities that we can deploy into lots of different processes. Then there’s that kind of the buy layer as well where we say, right, well, actually, the thing that you want, we don’t want to build that because it’s not central to our business strategy. It’s not going to be worth the time or the effort to do so, and the tool that you’ve brought forward to us sounds good.

It fulfills the criteria, so go go for it if you can pay for it. Or we might say when somebody thinks that they wanna buy something, we might say, actually, you don’t wanna buy that because we’ve got this thing that we license over here that you could use instead. So it’s kind of it it helps to shape the ideas and to get the business thinking about the value of AI and helps to educate them about strategy. But actually, when it comes to delivery and decision making, that that still happens very much in the central hub.

Yeah. I’m I’m seeing the same. I’m seeing a lot of teams trying to not only kind of integrate and federate the domains in into you just everything you just said. Right?

Let let’s let’s have a holistic review of does this make sense? Does it not make sense? Does this tool integrate to our overall IT strategy? Is this the right thing to do, the wrong thing to do?

I’m seeing that same thing happen within internal data science teams, where a lot of a lot of the the the CDOs that I’m talking to are trying to find ways so that data science teams stop kind of approaching everything from, you know, a Snowflake perspective. And I and I don’t mean that like Snowflake the the infrastructure. I mean, like, net new. Like, everything needs to start from scratch.

Everything is this net new initiative.

I need to go directly to the source to get all the data, and I’m gonna do everything from scratch. And I’m gonna ignore the fact that we’ve got all these other data management tools over here because I’m gonna do Python for matching, and I’m gonna do my own ETL, my own stuff.

I’m seeing a lot of CEOs focus on that. Are you are you is is that something is that’s kind of on your radar as well as kind of having more of this unified, you know, from an infrastructure perspective, certainly, but also from a management and operating perspective, more more of a holistic view across both the traditional BI worlds and data science and AI?

So, I mean, I’ve got quite a small data science team at the moment. So we’ve we’ve the way that we’ve grown that is is, you know, we’ve got our frameworks and our approaches, and so everybody follows those. We haven’t got any ongoing rogue yet, which is handy.

But I guess as that as that grows, and I’m assuming it will grow as more as as there’s more and more demand for what we’re doing, then we need to be very careful about that because I can see that becoming a challenge, and I understand that in larger organizations that probably is is definitely a challenge. But, yeah, I think we’ve spent a lot of time over the last few months just trying to make sure that we have consistency in the way that we approach things and kind of thinking, yeah, thinking about capabilities in a very modular fashion.

Making sure that we don’t just go and recreate lots of the same thing in different ways, but saying, right, we’re going to create a modular capability and then we’re going to use that over and over again in problems that look similar and, you know, feel similar to the problem that we’re trying to solve.

Again, I think because we’re very early on in the journey, it’s easy to have that approach. And I’m sure that as demand increases and, you know, the the kind of the the whole process around AI and the way that we set ourselves up increases, it will be more challenging challenging to implement that. But I think by starting with that approach and those principles in place, then hopefully that stands us in good stead to be successful.

Brilliant. Okay. So we’ve talked about this. I don’t know if if tension is the right word, but this need to find the balance between kind of pure domain centricity and federation and more centralization from maybe an infrastructure and technology perspective.

We also hinted at this tension that is existing between I want it all now.

Right? I want the shiny thing. I want some AI and traditional foundations. Right? Which I know that you’re working on.

Right? I I you know, I’ve I’ve heard your talks with Juan and Tim, which is fantastic. I know you’re working on governance, metadata management. I know you’re certainly working on master data management.

So how do you find the balance between those foundational things and I want some AI now?

Yeah. This is a, I think, a really interesting question because I think it is a very big challenge, and I think we’ve been on a bit of a journey with it as well. So when I first came into this role and and started Access Capital, I started thinking about I I kinda came in with a very traditional mindset about data governance, which is we need to have a data governance team that just gradually works through the churn of putting governance in place, And that’s just important foundations that we just have to get cracking with. And, you know, a couple maybe a year into into the role, I kind of took a step back and reflected across the whole of my team, And I thought, right.

Well, most of it’s going quite well, but governance is just failing. And it always seems to fail in in kind of you see it failing in in lots of organizations. And and I was thinking, I don’t understand that we’ve got somebody who’s really bubbly, who can engage with the business, who can kind of do that and and be a conduit to data governance, but it just wasn’t working. So at that point, I had a bit of a rethink and thought, right.

No. Governance shouldn’t be a separate kind of function that’s sitting over here and working on just kind of building those foundations. Let’s just embed it in delivery. Let’s not bother doing governance for things unless there is a and unless it’s contributing to something that we’re building, and unless we have a incentive for the business to get involved in doing governance with us.

So from that day forward, we removed the the set standalone data governance tower, and I embedded it within my my delivery functions. And so now what happens is if somebody wants an AI thing and it requires data, they can’t have it unless they help us to bring that data into governance. So if they they they’re willing to help us to kind of assign the ownership of that data to help us with the data defining the data, ensuring master data has the necessary governance surrounding the story of beautifully and prophecy, and all of those pieces. So so it’s it then becomes part of the deal, and it also means that they understand why they’re doing it rather than it just seeming like a kind of thankless task that they’re given and they have no incentive to to to be involved in.

So it’s really for for me, it’s about doing only doing the stuff that’s actually going to be necessary and doing it in a way that makes it it makes it understandable and where people are incentivized to do it.

I think that’s it. Right? I I think I I honestly think that’s it.

And I don’t mean to be this, like, like, overly flowery and and but when I’ve looked at governance, I mean, I’ve I’ve literally talked to thousands of CDOs about this topic over over being an analyst for years and talking about what’s working and not working.

And I’ve been thinking about this literally for years. And the more that I think about it, the more I’m forced to the conclusion that governance in and of itself struggles because we’ve separated it from business operations. Right?

Because we’ve made it, like, ironically. Right? But because we’ve made it this own thing that needs to be justified on its own, needs to be net positive ROI, needs to be all of these things optimally, and it just can’t be. Right?

It just it just it can’t be a p and l. Right? Governance just cannot be a a a p and l. It’s it’s kind of like the electric grid.

Right?

It needs some form of support because it suffers from the problem of of of the commons. Right? Because we all kind of benefit from it, but we don’t pay equally. Right? And we don’t benefit equally, but we all need it.

So when we separated it, it created a thing, and what you’re doing is to say your exact words were to make it part of the delivery function, which is It is it is like an ingredient to a cake. You want a cake. You want the cake. You told me you wanted cake.

It’s we believe there’s ROI here. We want to build it. Okay. What’s the governance needed in order to bake the cake?

Sounds like that’s the the basically the approach. Sounds right?

That’s that’s exactly it. Yeah. So we have like a, yeah, like a playbook for delivery, and governance is just interwoven within that. So if somebody wants something, we get our, like, business facing people, our data insights leads, to work with them on on what it is that they want to kind of co create that with them.

We do the architectural pictures because I like everything to be nicely designed. I think part of the reason that we bought into the people that we have is because people just built stuff about designing it. So we we we go through kind of the data architecture step. And and within that, we’re kind of saying, right.

Well, what are our requirements? What do we need in terms of data? And what are the governance activities that need to be carried out to ensure that that data is robust and is reliable and all of good stuff that we want from our data? And so it’s yeah, it’s just a step within our delivery playbook.

I I think I think that’s key. Because when I look at companies as well that are really good, like, really, really good at governance, the governance is actually interwoven into their business operations, not just the delivery model of a data and analytics function, but, like, deeply woven into their business functions. Like, I look at somebody like a Netflix or an Uber.

Governance isn’t an afterthought. It’s just kinda how they operate, but it’s baked into literally how they operate. So you are taking a a a big step towards that. Obviously, it’s part of your delivery, data and analytics, not necessarily business delivery, but absolutely love it. I think you’re onto something. I think more CDOs need to be thinking about this from an operating model perspective. There’s a lot of us that get kind of I I think we get concerned about taking that approach because when we engage consultants or maybe we look at, you know, blue books that are that are the best practices in our field that that talk about, you know, these frameworks, which I’ll say, hey, when you do governance well, you do all of these things.

And that kind of stands in conflict to a certain degree with your approach. Your approach is very pragmatic. Let’s do what’s needed to to order to to support an outcome. But these framework driven approaches are, okay, here’s what you need in order to be considered enterprise class governance. And I think that’s where a a big tension exists, particularly and I’m not I’m not trying to be negative about consultants, but, you know, if you hire a consultant to help you with a governance problem, they’re probably gonna throw a framework at you. Right? They’re probably gonna do a math assessment, some sort of maturity assessment and say, well, here’s what good looks like.

Right?

From a kind of a generic, here’s what good looks like. You’re taking a different approach, which is what is good for this outcome. Absolutely love it. Absolutely love it.

I think it’s the right approach. And I think more and more of us need to be taking your exact approach. Alright. Let’s let’s kind of keep tying the the thread here when it comes to AI.

Where do you see things going?

What what is what does twenty twenty six look like? Are you are you guys focused on kind of agentic stuff? Are you focused on chatbots? Are you focused on maybe thinking about customized smaller language models? What what do you what’s what’s what’s what’s the future look like?

So I I mean, there are huge opportunities for us to enhance what we do with AI.

And a lot of it so I think we’re not we’re not trying to change the world in twenty twenty six, but we want to build out some core capabilities that will enable us to enhance what we do. So some of the big areas that we’re focused on when I think about kind of the the business that we write, there is a huge amount of unstructured data that currently is processed in an incredibly inefficient way. It drives processes, but it drives processes with lots and lots of people picking things out and, you know, doing things that are very, very manual in nature. There is a massive raft of opportunity there, that’s where we’re kind of furthest ahead on our AI journey where we’re looking to kind of extract data from unstructured documents and then use that to power processes.

Then once we’ve kind of built that, then we can start to get more intelligent with that as well. So once we’ve got that unstructured data, then we can start to utilize it and and analyze it and, you know, build build our analytics layer on on top of that. So that’s that’s very much our focus at the moment. I mean, along side that, the other area where I think there is a lot of potential is in the kind of the chatbot space whereby, again, we’ve got lots of unstructured data that we can open up to people to read in a much more efficient way, to kind of take data out of it in a much more efficient way. That just equips people to be a bit more efficient in the way that they do things. Then once we’ve done that as well, then we can start to incentivize people, again, on the governance side of things to start to help us with ensuring that we’ve got the necessary governance around our structured data estate so that those chatbots become a lot more powerful as well.

Yeah. I’d say those are the two streams that we’re thinking about at the moment. We have started to think through the opportunities around agentic AI, I think that will be something that we definitely try to move forward with, but not immediately. That will probably come that kind of deeper into twenty twenty six. I think it’s a space that’s moving really quickly as well, so I don’t want us to necessarily jump in feet first to try and be first to the party early to the party when, you know, in a few months’ time, we might get to a place where it’ll be a lot easier for us to kind of build a lot of things ourselves.

Yeah. We don’t all need to be early adopters, especially if we’re in finance insurance. You know?

There’s we don’t all need to be early adopters. But what you said around unstructured data, I find really interesting because I think more and more of us are starting to focus here.

What is the what is the how kind of look like with with that? Is this is is a knowledge graph a piece of this equation? I know you’re leveraging data dot world dot now service now, you know, to help classify and catalog some of this data, what does the governance of that unstructured data? What do you think that starts to look like? How how are you actually gonna start doing that?

So, again, great question, and it’s it’s something that we’re kind of working through at the moment.

But, I mean, the first step is really just extracting kinda key terms from from unstructured documents. So one of the big areas we’re focused on is is submissions. So if somebody wants insurance, we get sent a submission, which has PDFs. It has number of different file types on there.

And we we know that there are consistent data items we need to extract from that. So it’s about extracting those data items and then pushing those into processes. What we also are very aware of is that that data is not just useful to push into processes, but also then to kinda create or to enhance and augment our data estate for us to then to then utilize further. And I think that’s where we can utilize some of the tools that we have available to us, much like the the kind of the knowledge graph underpinning data.

World, so we can start to utilise that to populate populate our kind of data catalogue and and the kind of ontology around our our data estate as well. So so that’s something that we’re working through. At the moment, we’ve, again, just focused on getting value quickly. And so that is the let’s just extract the items we need, push them into a process.

And then alongside that, we’ve got that kind of thought process around, okay, and how do we then use this to enrich our data to then be able to do more and leverage it further.

So it sounds it sounds a little bit like you’re focused on potentially kind of, is it new policy onboarding, policy processing, evaluation, doing some of this sort of because one of the challenges here, I think, around unstructured data for many of us is that there’s so much of it. Right? And where would where would we even begin to start to focus to get you know, because one of the things I’m recommending is, hey. We need to start governing unstructured data.

It’s what the LLMs want. It is literally the, you know, the nutritional source for these language models. What I think I hear you say is, hey. Let’s focus on optimizing this business process, And we can look specifically at this one business process, start plucking key terms out of PDFs, start understanding what’s sitting out in those PDFs with the goal of optimizing that process.

Is that am I getting that about right?

That that’s it. That’s been the kind of the narrow initial focus for us, which is let’s get some value quickly. So let’s just extract the key terms, and let’s optimize the process.

Okay. Brilliant. So again, a repeating theme here, my friends, if you’re listening to this, which is let’s focus on a goal, right, that we know is important because we’re talking to the business. We have ambassadors that are out there talking every day with the business. We’re understanding potential ROI. We are limiting our scope to things that we know will deliver value, and we’re working backwards from there.

Absolutely, positively love it. These are these are things I’ve been talking about for years, I think, Rebecca, more CDOs should be doing. This has been a wonderful conversation on a Friday afternoon. A great way or Friday afternoon for you, morning for me, a great way to end the week.

Really wanna thank you for the conversation. Also wanna thank you for participating in Prophecy’s Data Hero Summit yesterday. I think it was wonderful.

Thank you so much. Any last words that you wanna add? Things that we didn’t kind of touch on about maybe the future, things that you’re that you’re super proud of? I mean, what I’m hearing, I’m just I’m just loving this conversation. I think you’re doing a lot of things right.

Thank you. No. I mean, all I would say is that we’re really just learning as we go, so I think it’s really important to try things, see whether they work, if they don’t, then be willing to pivot.

And so I definitely don’t think that we’ve definitely got it all all right yet. But but I think the fact I think being nimble and, you know, spending lots of time reflecting and looking back and then working our way to go next has helped us to keep a bit of momentum and to have more successes than if we kind of stay down one track. I that’s an important part of what we’re trying to do here.

Critical. To paraphrase you, what you just said is is we you know, you’re embracing a growth mindset, right, which is you’re you’re trying, you’re throwing a lot of pieces of spaghetti to the fridge, you’re talking closely with customers, you’re setting and managing expectations, you’re not afraid to fail, and you’re learning from everything you do. Doing retrospectives. We need to do more of that, I would argue.

More retrospectives. What worked, what didn’t work. You know, we often in software, we call them postmortems. I’m not crazy about that word.

Yeah. Exactly.

But retrospective. Right? Like, what’s working, what’s not working. That is a growth mindset. That is a trait of the organizations that I’ve talked to over the years that tend to be far more successful than companies that don’t embrace a growth mindset.

So, again, love what I’m hearing. To all of your listeners, if you’ve made it this far, hey. Go ahead and hit like. Maybe subscribe.

It would be great if you joined our growing community of CDOs and people that want to be CDOs here on CDO Matters. I send out a monthly newsletter.

Every third Friday of every month, we do a live event on LinkedIn that you can do. It’s completely and totally free. All this content is free. All I ask is maybe a like.

Hey. Engage with me on LinkedIn. If we’re not connected on LinkedIn, please connect with me on LinkedIn. Rebecca, thank you again so much.

Thank you very much for having me.

Alright. That will do it for this episode of the CDO matters podcast. Tune in to another episode sometime very soon. Thanks, and bye for now.