Data Analytics Chat

The Future of Data Scientists and Data Engineers: How Data Teams Must Change

• Ben Parker • Episode 73

Share episode

Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.

0:00 | 37:57

In this episode of Data Analytics Chat, we’re joined by Phoenix Pei, SVP, Analytics Manager at Truist, to explore how the roles of Data Scientists and Data Engineers are changing in the age of AI and automation.

 

Phoenix explains why trust, leadership alignment, and business understanding are becoming more critical than technical skills alone, and how generative AI is reshaping how data teams work. We discuss which skills will matter most in the future, why so many data initiatives struggle to deliver impact, and how organisations should rethink team structure and collaboration to stay relevant.

 

A practical conversation for leaders and practitioners who want to build data teams that actually deliver value in an AI-driven world.


00:00 Introduction to Scaling Analytics and AI
00:50 The Role of Leadership in Data Science
01:24 Exploring the Future of Data Roles
01:46 Guest Introduction: Phoenix Pei
02:18 Phoenix's Career Journey and Insights
09:34 The Evolution of Data Roles with AI
11:08 The Importance of Business Understanding in Data Science
24:54 Challenges in Data Science Implementation
25:07 Leadership Alignment and Data-Driven Decisions
33:53 Preparing Data Teams for the Future
37:17 Conclusion and Final Thoughts

Thank you for listening!

phoenix:

Because you cannot scale analytics, right? Or AI without trust. I do believe that automation and AI reduce a lot of manual work. We traditionally associate with data science and engineering. So I think the gen AI doesn't re, doesn't replace the judgment. Actually. It's, I think it's it amplifies it a little bit. People may actually be surprised that in a lot of the big organizations, the fi, the CFO group they're still using Excel to do a lot of the financial reporting to do their financial planning analysis. So the differentiator. Is not mastering a specific technology, right? It's learning how to learn and also being comfortable working through ambiguous problems where the answer isn't obvious. In my experience, leadership alignment is really the biggest blocker. It's not a technology or TA talent. The reason is that I want to make sure the executives, the business leaders, see the data science team as. The driver, as the collaborator, not just the supporter.

ben parker:

The roles of data scientists and data engineers are changing with Gen AI automation and product led data platforms. Many teams are starting to import, ask some important questions, which roles will still matter in the future? Which skills actually gonna differentiate people and how should data teams really be structured over the next five to 10 years? In today's episode of Data Analytics Chat, I'm joined by Phoenix Pay, SVP of Analytics Manager at truist, and we're exploring the future of data roles. So if you're building, leading, or working in a data team today, this conversation is about how the profession is evolving and hopefully you'll be of interest to yourself. Phoenix, welcome to the show.

phoenix:

Thank you, Ben. It's a pleasure to be here.

ben parker:

Good, and I'm looking forward to today's topic.

phoenix:

Same here.

ben parker:

So I guess before we dive in, Phoenix, do you wanna just give listeners a quick introduction to who you are and the work you are leading today?

phoenix:

Absolutely. My name is Phoenix Pay and I am an electrical engineer by training. Today I lead data science and analytics teams in the financial service industry. So I have been focusing on building and scale scaling analytics, AI solutions that drive business outcomes, mainly focusing on revenue growth, operational efficiency, and the risk reduction. I spent about eight years in consulting earlier in my career, which I often describe as the greatest professional bootcamp. It forces me to solve hard problems quickly. Work across industries and communicate clearly with the senior leaders, often with the imperfect data and the tied them tight timelines. That experience really shaped how I think about the analytics in practice today. today my work sits as the intersection of engineering, data science, and business strategy. I not only care about the technical solution, but also make sure. Solutions are production ready, trusted, and well adopted by business user. So what really motivates me in helping teams what really motivates me is helping the teams to turn advanced analytics solution into something that really changes how a business operates.

ben parker:

Brilliant. And obviously I like hat where you mentioned about consulting, giving you, I guess that working in different fast paced environments.'cause it gives you a lot, you get more exposure to a lot of different projects and also helps your learning, doesn't it?

phoenix:

That's right. That's absolutely right. I really. Enjoy the looking back, right? I really enjoyed the consulting life in my early twenties, late twenties. It really forced me to pick up new tech technologies new business quickly across industries. And it gives me the exposure to communicate with senior leaders, which really don't happen in a more, pyramid hierarchy in the banks.

ben parker:

Yeah, and I've recruited across different, like technical areas. So ai, I mean done SAP in the past and software development like that. A lot of businesses do really value that experience as well.'cause you're coming in with different experiences. And obviously with that you're coming across different challenges, so it's, yeah, definitely beneficial.

phoenix:

That's right.

ben parker:

so obviously that's obviously probably a, yeah. Your great experience has shaped your approach to work today. Is there anything else that's really benefited your career, would you say?

phoenix:

Yeah, I am thinking that in the consulting life I worked on a lot of the analytics initiatives, large analytics initiatives where the solutions are. Technically strong are very technically strong and well designed. It takes months, hours hundreds of hours for us to develop those solutions. But it's harder to make it to production or actually change any decisions. And I also, spend a significant part of my time building. Sound risk and analyst platform in what we call the second line of defense in the risk management space where the priorities, control, governance and getting things right for the regulators, right? Even it takes time. So that re that experience really helped me think about what it means to have a sound, risk management system data quality lineage, documentation, risk discipline, whatnot. Because you cannot scale analytics, right? Or AI without trust. Right now as I'm leaving the consulting, I'm sitting much closer to the business in the first line of defense where the reality is a little bit different. The decisions need to happen very fast. The business just cannot wait for months for perfect models or frameworks. So the challenge that I'm focusing now is to deliver both how to build an analytical. Analytically sound well governed programs while meeting business need bus business speed. That means designing the governance not to block the process, but to enable it. And embedding some of the controls into the platform instead of. Laying them late on later and just very be very clear about which decisions need precision versus which need direction for the ones that need the direction, right? We can deliver the solution the work faster. And for the ones that which need more precision or for more regulatory scrutiny will definitely take more time to build out the controlled, governed solution. So the balance between rigor and velocity is what I think shapes how I lead the AI initiative initiatives today.

ben parker:

Yeah, and I think it's, that's one of the. Big challenges. Now, obviously we're blessed with generative AI and all these new solutions, but also it gives you more headaches. Every business function will be coming to you with a use case you gotta be able to manage. So what's the one that's going to, move the needle now?'cause you can't do it. You can't do in one go, can you?'cause there's so much like the biggest. A lot of people know where they want to go to, like the cha, the obviously the hurdles they wanna get over, the projects they want to deliver. But when you change culture, change is the hardest thing, isn't it? Technology change. You've gotta embed that. It's not, you can't keep, you can't get in a mess, can you really, if you're doing too much at once, you're gonna get in a situation where it's undeliverable.

phoenix:

That's right. Yeah. And sometimes I think you probably also have this experience the business executives have a, 10,000 view right? On how this solution should work. But coming down to really implementation execution. They sometimes underestimate how long it would actually take to implement it, execute it. So a lot of times it started with a big blueprint, but then the detailed roadmap is not carefully designed. So it leaves as the data engineers, data scientist very little room to to make it designed to make it well controlled, governed, so we were, yeah. Implicitly or explicitly pressured to deliver something just to, for the sake of delivery.

ben parker:

Yeah, and especially the industry you're working in as well, you've got more, there's other gaps. You've got the regulatory stuff to go through as well, so it is, it takes time, doesn't it?

phoenix:

That's right. It does.

ben parker:

Okay, brilliant. Let's move on to the data topic. So I also mentioned prior to recording this obviously when I started hiring it did a lot of Oracle DBAs, obviously Yes. You don't do that anymore. Obviously they're ingrained in data engineers. So things are moving really quick. So how, how do you see the roles of data scientists, data engineers, evolving over the next sort of five to 10 years, especially I guess with the rise of gen AI and now and automation?

phoenix:

Yeah, I think both roles will. Still be there and they'll be actually become more impactful, but also more clearly differentiated in different functions, different purpose. I do believe that automation and AI reduce a lot of manual work. We traditionally associate with data science and engineering. For example, in my own work, I regularly leverage tools like copilot to help with coding. So that efficiency really frees up the, my cognitive space. So I spend less time on syntax and the more time thinking strategically about what we are building, why we are building it, and how it should be implemented so that we can scale freely in a real business environment. So looking ahead, I. Think data engineerings will still need to focus on building reliable carbon to scale scalable data products, right? Because data is after all the foundation but maybe less, more. Pipeline plumbing more platform and the architecture thinking, and then the data scientists will spend less time on feature wrangling, right? Finding the best feature, engineerings all of that, but more time on problem framing and decisioning science and translating analytics into action. I do see a still I do still see a big gap between the decision. That data scientists made versus the decisions that the business made, right? So how to enable the data scientist to actually bridge this gap to be able to tell the story to the business. I think that's very important. So I think the common thread is really this. Both roles will need a much stronger understanding of their work. And how that connect to the business outcome, how that connect to the financial outcome. So I think the gen AI doesn't re, doesn't replace the judgment. Actually. It's, I think it's it amplifies it a little bit.

ben parker:

Yeah, no. Even doing hiring three years ago for data scientists, it used to just be focused on tech skills. But like now with a different era, aren't we now? I mean we, like you said, you've got co-pilot can do the heavy lifting for you and it's definitely getting more strategic, like it's getting more creative. If

phoenix:

Yes,

ben parker:

the business domain and add value in that regard to adding to that solution you're gonna under, you're gonna know better where to just gonna get the better results and how you're gonna implement to get better solutions, aren't you?

phoenix:

that's right. That's right. Yeah. And I think we also leverage co-pilot to do a lot of, cloud-based transformation, migration type of work, right? Instead of manually search, replace a lot of the target source tables, targeted tables, we leverage a copilot pilot to do that. That definitely. Saves a lot of time. And then it's really a efficiency gain in our work. And with that time saved, we can definitely spend more time to think about how to actually curate the data, right? Instead of just to do lift shift, how to cur it so that it can be designed better. And that can be, used by the data consumers better.

ben parker:

Yeah, and I guess we've already seen, it's already like data product scientists, that's a bit more domain, but I think over years to come. Do you think it'll get even more like specialized like in each department within a business, say like one that's obviously you. Yeah. I've got marketing data scientists. But would it go deeper and deeper into the business? Do you feel?

phoenix:

I do think so. I do think that it will, the, in the future, the, whether it's data engineers or data scientists that they need, we need to have a stronger. Business, a stronger business connection in terms of knowing how, what product are we building, right? And why we're building that and what is the, what type of outcomes? What about decisions that we're driving? So I think if anything, the team would sit close to the business, but operate more like a product and engineering teams. And honestly, that's exactly why I enjoy being in the first line of defense, which is the line of business because I think we are close to the business and to understand the real problems, to use data, to influence decisions in real time and see the impact directly. What's the revenue that's gained? What's the efficiency? What's the cost saved and what are the clients. Out outcomes. What are the SAT client satisfaction scores, things like that. So yeah, I do think that it does require the data engineers, data scientists to know the business better. It may not necessarily be all banking related business, right? It can be any type of industry consumer goods energies technologies, right? But close to the customers, close to the business development side.

ben parker:

Yeah, no, definitely. And I think it's gonna be important for businesses, maybe if they can embed, if you could work with say the finance function or whatever function it is to actually get an understanding their, the business problems as opposed to the tech problems.'cause. Once you become more rounded, you are, you've got a more rounded view of what's happening.

phoenix:

That's right. And you mentioned about the finance component, right? People may actually be surprised that in a lot of the big organizations, the fi, the CFO group they're still using Excel to do a lot of the financial reporting to do their financial planning analysis. And then it's just so hard to change this Excel mindset, even though there's so many, advanced FinTech solutions. Technologies using Excel to build out the report income statement, financial statement is still a big part of their job. So even how to modernize that type of work. I think how to, in how to leverage the data scientists to enable some of the financial analysts to. With their work. I think that's even something to explore.

ben parker:

Okay. And do you believe there would be more of a sort of full stack data professional? Will that become the norm or will enterprise just like deeply always need deeply specialized talent in, so like engineering analytics, ml governance.

phoenix:

Yeah, that's a good question. And honestly, that's something that I have been, discovering in my different roles. So I think the idea of a single person doing everything end to end works very well in a specific in certain, project moments, but it doesn't scale. As a permanent operating model in a larger group in practice, I do this myself today. When a solution is high profile time sensitive and the needs rapid delivery, I often build it end-to-end data sourcing, modeling logic, even parts of the implementation, and I would present the solution to the business. So that speed matters early on, especially when the business. Need as a result, right? Need the confidence quickly. But once the solution is about 70 to 80% established once the architecture, the patents values are clear, that's really when I pass on the project to specialized teammates. So the engineers, data engineers were hard in the pipeline. The data scientists were refined, the model, and then the, governance partners will ensure, it's scaled safely. So I don't think that the full stack execution will become a norm, but it's more a full stack awareness. So the best teams should understand the entire life cycle, right? And if they spec, if they specialize deeply in one part of it, that's great. But the, this full stack awareness is definitely very important.

ben parker:

And I think also if you specialize someone so much. Surely they're gonna get bored of their role. Like you, I guess if you are doing multiple different jobs as a data scientist, data engineer, you're gonna that blender skillset, aren't you?

phoenix:

That's right. Yeah, exactly. Yeah.

ben parker:

they're say doing one specific requirement, yeah, it's gonna be, it'd be good for certain bit, but I think your mindset, your brain, you want to be tested doing different things and overcoming different challenges. Don't, won't you.

phoenix:

Absolutely. Absolutely. I think you're right that I want to write as a data scientist or engineer. I may not have to be the business analyst to collect the requirements in the beginning to document it, but I definitely want to know why the business require, why the business require make the request like this, right? Why they want the requirements like that. I don't want to just take it as this. I want to ask questions on why they want that, because I may be able to, as a data engineer and data scientist, I may be able to support them, advise them to come up with a better. Requirement so that we can build out this solution better. So yeah, so I think I do hope all the data engineers, data scientists, and even the future AI developers right, can have this full stack mindset that's how the pro product is developed end to end.

ben parker:

So obviously with the changes that's happening, what do you feel? What skills are gonna separate the best talent from the average performance in the future?

phoenix:

Yeah, I also ask this question to myself all the time. I have recruited many people in the past few years both in consulting and at Truist, and I think, sound technical skill. The fundamental technical skill is very important. I definitely want to hire somebody that's are fluent with SQL or Python, right? The fundamental programming tools. But I don't think this is a differentiator. If there are two talents, they're both very strong with sql, Python, R BI, tools like Tableau, click View, power bi. What are the differentiator? I think it is the ability to connect the dots. It is ability to keep learning new skills. Even looking at my own career, I realized that I learned the core programming skills in the earlier career of my life and then. That foundation made everything possible, right? So once I understand how data moves, how metrics are defined and how the system are behave behaving picking up new tools become just much easier. So that foundation allowed me to adapt as a technology to evolve. I later on learned the platforms like Alteryx, MongoDB, GraphQL, AWS Snowflake as the needs of the business changed. So I think the reality is that the cycle never stops, right? In our area in this world. You mentioned that we're just always learning. We always have to learn something. So the differentiator. Is not mastering a specific technology, right? It's learning how to learn and also being comfortable working through ambiguous problems where the answer isn't obvious. I think that for the talents that I hired, and even for me, I think one of the important differentiator is how we solve, how we solve a problem. That's probably not clearly defined and how we walk, work through the ambiguous problems. And how we develop a trade off with the business stakeholders to say, Hey, I understand you are asking for this, but this is how we are actually, building it. This is how it yearly happens, and how we can meet half halfway, how we can find the balance. So I think. Understanding the business context and the framing the problem before jumping to the solution. Explain the trade offs clearly and the really influence the decisions without hiding behind the technical complexities. I think that's very important. I know we are all very smart people to be able to work with data, but not to. But don't hide it behind it to say, oh, this is very technical. It's hard, it's too technical for me to explain to you. I don't think that is the answer. So I think the ability to really be able to explain all of this to explain to the business stakeholder the technical approach to a business audience, right? And I think that's really important.

ben parker:

Yeah, definitely. Like this the story again, it's storytelling, isn't it? The, yeah. Able to articulate the tech terms into business context and. Do it in basic layman terms'cause and they leaders are so busy and they're not interested. They just wanna know the outcome really that how you're gonna get there. Not the ins and outs of everything. They just wanna know the process.

phoenix:

Yeah.

ben parker:

So then, yeah, I guess if you are advising someone to enter the field today, what I mean, where should they invest learning.

phoenix:

I still think that the fundamental, the foundation the statistical statistics, sql and the, and a data modeling Python are, this level of information is still very important. We need to build a strong foundation first. But, the other is as important as the first one is really just to understand how business actually makes decisions. We often think that, okay, our data is enabling the business to make the decision, but in reality the ugly truth is not the business make decisions based on a high level picture, right? They sometimes use the data to support, to justify the decisions. So I think be able to understand how business makes the decision. Sit on the meeting and learn how the metrics are used or misused, right? I think this really helps and even make friends with business stakeholders to educate them why the data important, why the analytics is important, and then be able to find more cha, data campaigns, right? To work with the as I think that's very important.

ben parker:

Yeah, no, definitely agree with you, especially on the statistics side. Like even when we are recruiting for a lot of our clients, that's the key things they want. Yeah, you still need the foundations and of statistics, like that's the key component, key ingredient for a lot of the hires. They want that. Can you spot the gaps that is gonna be needed, apply the right models. I think a lot of, I know there's a lot of excitement now with generative AI and moving into that, but you still need to have the foundations as the key.

phoenix:

That's right. Yep. Yep.

ben parker:

So I guess like most organizations are talking about becoming data driven, but I guess most many still struggle to get value from their data. Where do you see the biggest blockers today for organizations?

phoenix:

Yeah. In my experience, leadership alignment is really the biggest blocker. It's not a technology or TA talent. I worked with many top banks over the past decades and I realized that many organizations invest heavily in tools and teams, but they don't really clearly define what decisions data is meant to improve. When that happens, data team, we're often pressured to shape the analysis to fit the existing narratives or broken processes. We're not really using the data to enable the business outcomes, so I think that's where things really break down. The data shouldn't be used to justify the decisions after the fact, but rather it's if the underlying sales or operational disciplines isn't the sound. Then no amount of analytics will fix it. So I, and I also see successful organizations, groups that are are doing this better where they are willing to fix the core processes. Setting clear accountabilities and the then using strong government data governed data to measure the reality, right? Whether the results are comfortable or not. So I think culture and the governance matter, but the leadership really sets a tone. If the leaders are more open to change how the business operates based on what the data actually says, that's where that's when the data actually starts creating the real value.

ben parker:

Yeah. And obviously it's, yeah it's obviously a massive challenge for companies, isn't it? If you are, because you talk, when you're talking about leadership, you're digging into culture. There's a lot of change that, and it's hard to do, isn't it, for businesses?

phoenix:

Yeah, that's right. Yeah. And I think I'm not in the executive role yet, but I definitely. Want to be able to educate those executive leaders more. And I think the more people are doing this to just bring awareness of the data, the solution technology to the executive leadership the more opportunities they get right. To learn, to be trained yeah. And to build that awareness.

ben parker:

Yeah, and I do think that awareness is getting stronger day by day, especially since generative ar. I think it's how important like data is to a business. I do think things are getting more closer together. So it's interesting to see how the future plays out.

phoenix:

Yeah, absolutely.

ben parker:

Then I guess we, like many data science projects fail to make it into production. Why do you think like initiatives stall and what must change in organizations to, improve the success rate? I.

phoenix:

I think the common theme for. feel like our topic and a lot of the discussion around the future of data scientists, the data engineers, I think it's still the disconnect between the business and the data science. I realize that many bus business leaders don't have a strong data science background, which means that they sometimes struggle to give clear direction what problems analytics should solve. Are how success should be defined. And at the same time and I think the roles are the same. The scientists are often not very business savvy enough to to, to pitch their idea, right? Or they may not be given an opportunity or space to really pitch their work in the business term. So I think, the gap is that as a business can't clearly articulate what they need, and the data teams focus on the model performance rather than the decision impact. So when the projects move forward, but without a strong sponsorship or ownership or path forward, then it's never reached to the production, right? Because the business, you don't get business buyin. And then the, if you don't get the business buyin, then the solutions. You build, no matter how strong technical they are, it's of little value. So I think what needs to change is awareness, we talked about it, and also just a shared accountability. I think that for any data science project to make it to production, it shouldn't just be data scientist job, right? The data scientists need to be working closely with the business to understand the. Business context and be empowered to tell the story why their work matters. And on the other hand, the business stakeholders need to be the champion, right to to engage earlier, asking the better questions providing better requirements, shaping the priorities, and then really gave a clear direction on how the solution should be. Should it be developed? And how should it be implemented? So I think when that dialogue is missing, the solution, the data science product is struggle to survive, right? So when it's there, it happened earlier, I think it become, the production becomes a natural step, right? It's not a hurdle it's what meant to happen.

ben parker:

So do you think then future data teams should sit closer to the business then?

phoenix:

I do. Yeah, I think we, yeah, I do. I think

ben parker:

just.

phoenix:

Yeah I do, I think we touched business in earlier as well. I do think they should sit close to the business and, but should operate like a strong product and engineering team so that they can be embedded with a clear ownership of the product and share the roadmaps and measurable outcomes while holding some self to high engineering and analytical standard. However, I do think that there are companies or functions or industries that they may have a reason why the data team should be separate share service, right? Instead of sitting close to the business. But I do think for the banking industry where we are now, I enjoy being the first line of business, first line of defense because it's close to the business. And I. Understand exactly how my analysis, how my solution is used by the frontline teammates

ben parker:

Yeah, not

phoenix:

and yeah. And also I get the feedback immediately, right? I think the feedback loop is also immediate, so I. I'm not only just seeing, my results are used, but I also know what, what works, what doesn't work, and it enables me to refine my approach refine my models better.

ben parker:

Yeah, no, I definitely feel interesting to see how businesses deal with it.'cause, I've seen other people, other companies that do like hackathons where they get the business and tech data teams altogether. So I guess you can have like collaboration it, you don't necessarily need like sitting on like desks close to each other. But if you're doing this like collaboration work. It's gonna benefit your business. Everyone's gonna know everyone's problems, and then it's gonna be more knowledge to share, isn't it? To communicate. And you're gonna be able to create more impact. And then guess what you want is deliver more value.

phoenix:

Yeah, I do. And this has been, then, this has been ongoing discussion, ongoing challenge for all the players, right? In the financial service industry because we do know that if we have a shared service where we can hire the brightest, the best tech technical talent, right? Whether it's a AI developer, data engineers we actually have a more more solutions or more capabilities to build those solutions. The challenge is really how the different functions, how different line of business, quote unquote, compete on the services, right? Because if they cannot work, they cannot deliver everything at the same time for everyone. It's a shared service. So how to prioritize how they prioritize the work from different functions, that has been a challenging on them, right? So then. As a business, we are like, okay, if you cannot deliver at a speed for our business, then let's have the resource on our own so that we can do it. So I think it's, there's definitely pros and cons on, different operating models here.

ben parker:

Yeah, definitely. I think it's any, any change, you've got opportunity and you've got challenges. It is, as soon as you move the goalpost, new things happen, new challenges are gonna hook, occur. It's just, how you adapt and it's just, it's the way it is.

phoenix:

Yeah, it is.

ben parker:

So I guess then, if you're leading a data team today, what's the one thing you would change in the next 90 days? To prepare? To prepare, sorry for this future?

phoenix:

I think about this question all the time. I ask this question to myself all the time. How would I do differently? I think number one, I would immediately.

ben parker:

I.

phoenix:

Align the team around a small number of high impact business decisions that we're responsible for influencing which means that we may need to stop something else that doesn't support those outcomes. The reason is that I want to make sure the executives, the business leaders, see the data science team as. The driver, as the collaborator, not just the supporter. As I mentioned before, I don't want our analysis to justify the business outcome, right? I want to our data to be able really tell the bank how we are doing and what we did right or wrong. The second part, as important as the first one, is I would intentionally engage a small set of business s at the executive level to. Really to gen who genuinely believe in data-driven decision making, right? Have the authority to act on insights. I would engage them and then work with them. Pitch our work, work with them, have them be our champion to move forward a lot of the work because we, if we cannot create an impact on the business, our team, our work will continue to be. The s the supporting document, the artifacts. So when we have the executive champions who ask the right questions, who are willing to challenge the old status quo, right? To challenge the intuition and the visibility, use the data to make decisions, it changes the whole dy dynamic. It echoed with one of the que one of the thing we discussed earlier, it's the leadership alignment, right? The culture, the leadership alignment, how the leadership is willing to learn, adapt. I think it really make a difference. So that combination, the clear ownership, the decision ownership, and a strong executive sponsorship and. The ruthless right. Prioritization, I think that will turn analytics from a reporting function into a strategic capability. And I, I. I came up with it, and I think it's a part of, because of the work we did a few years ago with my previous leader we had a data analytics council meeting with the executive every month and we present our analytics project to them and it really helps the, an, the executive leaders to understand the day in, day out what we're doing right. To view us not as a reporting function, right? And not as like just a data pool team, but we actually strategically think about how the work can help them make better sales, save more operating costs, things like that. I. Hope and I really hope that more analytics leaders can do that, can be a champion on their own, and also engage more business champions in this process to continue to bring forward the work we are doing.

ben parker:

Brilliant Phoenix. I've really enjoyed your insights and your conversation, so thank you.

phoenix:

Absolutely. Thank you as well, ba I enjoyed this conversation as well.

ben parker:

Brilliant. And hopefully listeners you've enjoyed today's episode please do make sure you follow or subscribe at Data Analytics Chat, so you do not miss future episodes. Thank you for listening and we'll see you in the next episode. Okay.