Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
025 - The Rise of the Data Product Manager…
379 Plays10 months ago
The Rise of the Data Product Manager…
How often have you joined a data team where data is poorly understood by the business, siloed, and churning out tickets with little impact? Or found that the Analytics team doesn’t communicate with Data Engineering, leading to project delays due to missing data?
These common challenges underscore the need for a new role that’s gaining traction: the Data Product Manager. This week on The Stacked Data, I'm joined by Karen Francis, a Data Product Manager at B&Q, and Rianna Kelly a Head of Data at Zeps.
Karen and Rianna have firsthand experience with these issues and many more. They’ve seen how powerful and transformative the role of a Data Product Manager can be. In this episode, they share insightful stories and discuss why this role is crucial for high-performing modern data teams.
We discuss:
- The responsibilities of a Data Product Manager
- The typical challenges they overcome
- How Data Product Managers drive value and efficiency
- Case Studies and Success Stories
- Why more data teams need to adopt this role
Tune in to gain valuable insights and understand why the Data Product Manager is essential for any data-driven organization.
Recommended
Transcript
Introduction to Stacked Podcast
00:00:02
Speaker
Hello and welcome to the Stacked podcast brought to you by Cognify, the recruitment partner for modern data teams hosted by me, Harry Golop. Stacked with incredible content from the most influential and successful data teams, interviewing industry experts who share their invaluable journeys, groundbreaking projects, and most importantly, their key learnings. So get ready to join us as we uncover the dynamic world of modern data.
00:00:34
Speaker
Hello everyone and welcome to today's podcast episode. We are going to delve into some of the biggest challenges data teams face when it comes to communicating value and demonstrating impact to the business, as well as discussing strategies in how to overcome them. a key part of this that I've been seeing is the emergence of a new role as the data product manager. I think many of you can relate to that feeling of the data team often being siloed a department within an organization, maybe not as well understood as it it can be. and And I think that's where it's
00:01:11
Speaker
really evidence of data teams not being able to to clearly communicate their value. We've definitely seen this over the the last 18 months leading to to hiring freezes and in data teams at at best and significant downsizing at worst.
Meet the Guests: Karen and Rihanna
00:01:27
Speaker
Today, I'm joined by the wonderful Karen Jean Francis and Rihanna Kelly, who are going to dive into this topic with me. and They've both got excellent experience in the in the space. Karen is a data product manager at B and&Q, as well as the host of the Women in Data podcast. And Rihanna has been leading data teams at Zeps. It's great to have you on, guys. Thanks for for joining me. Thanks for having us.
00:01:53
Speaker
No worries at all. I suppose for the audience it'd be great to get a bit of an intro to you both. I don't know who wants to go first. who I think I'm going to go first because after that Rihanna is going to make me feel self-conscious about my introduction.
Karen's Journey and Role
00:02:04
Speaker
and So I'm Karen, as you said, and I grew up in Guadalupe in the Caribbean, which is an island most people would not have heard of, unless you watch Death in Paradise and in that case you will know. But I'm talking about, as this is where it's filmed, I've been in London for just over 10 years now. Before that, I spent some time in Paris where I started my career in data. And I think I see myself as an analytics translator. um And that's because I'm really passionate about making data and analytics as accessible to everyone in the business.
00:02:41
Speaker
and I also help data analytics professionals thrive. so you We are talking about you know a bit of a career journey. I feel like I've been all over the place at the start of my career. like A lot of data professionals are there. I spent a lot of time really searching what it meant for me to be a data professional and what was my definition of that. and That's because there is a lot of noise going on out there with buzzwords and then shiny job titles that are really trying to distract you from the essence of you. and That's how I ended up going to data product management, which is around all the trial and errors that I've made, trying to find the space where I would really add value to a business and feed full fields. In the past, I've worked as a data analyst, I've also worked as an analytics manager, mainly in financial services,
00:03:36
Speaker
So in organizations like Carlitics and Monzo. And as you said, I am now a data product manager at BNQ. So I'm going to be supporting their analytics and data as science team from that space. And I host and produce the Women in Data podcast, which I've been doing for over four years now. And yeah, that's about, that's mean and nutshell, I want to say. Noice. Rihanna? Karen, you are always so hard to blow it out of your introductions. Slightly, I do not have quite an exotic start to life.
Rihanna's Career Insights
00:04:08
Speaker
Not so much sunshine, but hi, I'm Rihanna. I currently work in the data team at a payments company in both an analytics engineering and analytics capacity.
00:04:18
Speaker
It's a payments company that's currently focused on remittance. So that's sending money across borders. And it's actually the second remittance company that I worked for. I seem to have found something that I really, really enjoyed doing. And for the v majority of my career, I've been in data, starting out in a traditional finance company and then very promptly finding my way and into FinTech and always focusing on organizations that have some sort of social positive aim. in mission. And within that, I've worked in mainly analytics roles. So sort of insights, starting out with insights leads, head of product analytics, head of data, working my way sort of around that side of a data team, really resonating with what Karen says around
00:05:00
Speaker
you It does take some time, I think, to find your feet. And what I really love about my current role at Zeps is that spending a lot more time on the engineering side for about the past year, I've been working very heavily with an analytics engineering team and really seeing a different perspective. And I know that we'll talk a lot more about that today. So thank you for having us. Really excited for this conversation. No, as I said, it's ah it's a pleasure to to have you both on. I know that you've got done a lot of the recent blogs together and yeah share share some great insights with the with the community because you I think you both come with different perspectives which clearly complement each other and are very in touch with with really what's going on within data teams. So I definitely think yeah people should be be checking them out.
00:05:45
Speaker
Okay, let's dive
Communicating Data Value to Business
00:05:46
Speaker
right in then. I suppose, Karen, I started off the episode by ah outlining how you know we're seeing data teams really struggling to to demonstrate their value and and really communicating that with the the business, particularly in recent times, but I think it's been systemic for much longer. Is is this something that you've you've witnessed as as you've transitioned in across your career? I mean, i absolutely. And I think we've all been there, right? I do believe this is a ah deep problem that we all have across the organization, and that needs fixing. So you would have seen this start around 85% of data science slash data analytics projects fail to deliver the value. And it is important for us to address that. Now, how they measure this 85%, I do not know. But I feel like it's an interesting start to to share.
00:06:36
Speaker
and As data professionals, we really want to work on interesting projects, and people talk a lot about having the good balance between proactive analytics, and I'm going to talk a lot about analytics because that's where I've started for most of my career. I know Rihanna, we've talked a bit about seeing the other side of things, and I was thinking, no, no, no, I just want to be analytics. So these are the different perspectives we're talking about, our breakfast conversations, basically. ah And so as I was saying, trying to find the balance between proactive analytics, ah interesting pieces of work where we see we can add value, and also finding the balance with that with ad-doc requests and repetitive tasks that data professionals might feel like they're not able to play by their strengths when working on that, but are necessary for the business and stakeholders as well.
00:07:29
Speaker
And I remember a few years ago, I was working as a data analyst. And I think Rihanna likes to describe the kind of situation I was in as JIRA ticket hell. And I think a lot of people can relate to that. So you're going to have In my case, it was account managers and account partners, but you're going to have loads of people from the business coming, putting in a ticket, saying, hey, can I have this? And then the data team will go work and then provide whatever was was requested.
00:08:04
Speaker
And it felt like we were just pulling numbers. We were really not adding that much value because we could not use these keys where we were the strongest at. And sometimes we were, I don't know if you've been in that situation, Rhianna, but sometimes we were feeling like people were not actually looking at the results before a few months, even if they had said, this is urgent. So someone would come and say, this is super urgent. Can you give it to me night tomorrow and then three months later, they come back, hey, you remember this number you gave me? It looks a bit odd. And I'm thinking, but I thought it was urgent. So this is not an ah a good outcome for anyone. And I think talking to loads of people in the ah in the field, we've seen that quite ah quite a bit. And this is definitely, as I said, a problem we need to address.
00:08:54
Speaker
Yeah, I think the the interesting part is there's many organizations that you mentioned there. They were trying a data team. So anyway, trying to move away from being the that ticket churner, which I think makes you become very much ad hoc. you' You're seen more as a you know software engineer and a technical team. And I think that the whole point of data is not to be a ah ticket churner, it's to be strategic. How have you moved away from ah from that sort of ticket? driven approach because that's that I think the the key part of one of the key aspects of it.
00:09:27
Speaker
Yeah, I think being a service team is not something we we want to be as data professionals.
Understanding Stakeholder Needs
00:09:33
Speaker
And there are many ways to walk away from from that. And it's done to you know how different organizations will have different solutions. But for me, in that particular case, it was very much about creating deep relationships with the stakeholders that were not causing the problem, but that we're working in in that way. So as I said, helping people understand the value of data, thinking more strategically is very important. So spending time with stakeholders, understanding what it is that they need, why they need that, and how we can support them is very important, and showcasing the value of data.
00:10:15
Speaker
So showing, okay, this is the problem you're trying to solve. This is what we've done in the past. These were the results. And then from that, really getting them on board and understand what it is that we can do. Because a lot of the time, what it is is... They think they need something. They think this is how it's going to be solved. But it's not always the case. And that's when we end up oh with them saying, can I have this number? And then them coming back and thinking, this is not quite the number I was looking for. Can I have this number instead? And then you stuck into a loop of pulling numbers and pulling numbers. So sitting with them and then getting them to understand from the start.
00:10:55
Speaker
what is the problem you're trying to solve can help quite a lot with that. But Rihanna, what's your view on that? I was thinking this, I'm just nodding sort of furiously here, I feel super cramped, see it, you know, what you you say around people kind of bringing you the tickets and you're saying, oh, this is really urgent. The fact that it is all ticket based and you're not necessarily, you know, you're just receiving, you're not doing as you say, which is going out and actually understanding what the needs are, what the problem is actually. trying to be solved is sort of leads you to just being this sort of providing team rather than a ah questioning team. and And you just do exactly as you said, as opposed to you really understanding what what it is ultimately trying to achieve, what that actually means for the business as well in terms of the prioritization angle of it. And when you sort of step back
00:11:47
Speaker
And look at all this, all the work that's being done. Is it actually going to add together to an overall sort of value? Is it going to enable you to communicate some sort of impact with the organization, which is obviously something that a data team really, really struggles to do. Because when you come to sort of quantifying things, it feels like sometimes in these service teams, all you can actually quantify. is how long it takes you to respond to a ticket or how many tickets you do. And I feel like if you are in a situation where those are the discussions, you know it's it's taking that step back and actually understanding you know all the things that that Karen noted around what it is you're trying to achieve and why. I think in terms of this Jira ticket, hell, or I remember doing a survey when I started a new role and and asked people
00:12:32
Speaker
I needed that for open feedback about the team. Someone said, you know, it's a black hole. I send Enduro tickets and they just never come back. So, you know, unless I think people have kind of learned that unless they say it's urgent in a lot of these teams, it doesn't ever get done, whether it is genuinely urgent or not. But I feel like we also do that within the data team as well. So, you know, take us in a little bit of a different direction to pull me back if this this isn't right. But another scenario that I see is where the data team themselves aren't communicating. And yeah I started my career in analytics, and up until the past year have spent all of that time in the analytics team. And Kara and I talk a lot, as she said, around the wider teams. And
00:13:12
Speaker
I think it's so easy when you're in analytics to think that the analytics team is all that you need. You know, I have done this a huge amount up to about the past year, so but that's rarely the case when you think about what we do in analytics, we talk and promote about the benefits of speaking to the business stakeholders and you're understanding their needs, building output that helps them achieve their outcomes. But it's only once you start the project that you then start to ask the questions of do I actually have this data? Is this data reliable? um Because these are things that the analytics, this is data that the analytics team likely doesn't ah notice to consumer of. And if the answer to either of those questions or both of those questions is no, you then pivot to the other sort of juror ticket experience that you know, which is the service desk of the data engineering team. You throw in a ticket and you essentially do exactly to them.
00:13:59
Speaker
ah what people are doing doing to you. It just continues upstream. and I remember sort of starting a new role in analytics, coming in, having some really great workshops with end users around kind of metrics that they needed and wanted. We had a fantastic discussion doing all the things that you know Karen just suggested we should do and promoted around what's the business need, what's the prioritization.
Unity and Communication in Data Teams
00:14:23
Speaker
But there was no communication with any sort of engineering team there. you come out and you realize that that data isn't there and in this case it really genuinely was oh we'll just you know we'll throw in a ticket over the fence, stay to engineering and you know see when it can be done you know you push for the urgency you push for the prioritization usually with an unreasonably short timeline but no real proactive sort of discussion around what it is for and and why it's just it needs to get done I don't know what you're working on this is really important. So
00:14:55
Speaker
I think there are the two angles to being a service team, both what that means for you and your communicating value, but then also what... sort of how you continue that and perpetuate that across the data team. And I think there's a real, um that's really like what the business sees when they see this lack of cohesion across the team, they see this sort of, I don't even know if it's a lack of relationship, but just a lack of sort of unity. It really sort of does not lay a foundation to enable you to communicate impact and and work efficiently in any way. and And really understand as an overall data team, what is it that you're trying to achieve for the business, you know, in terms of business priorities, but also the technology priorities that we have as a very technical team for the platform. You know, it's very, very difficult to communicate that. I know that we've spoken about this before, Rihanna, and I completely agree the fact that we have silos within our own data teams. We're talking, you know, how data can be a silo part of the business. But when you actually look within the data team, it can be very similar, especially in larger organizations, enterprises where they are
00:16:00
Speaker
the engineers The data engineering has no idea what analytics are are doing. And as you said, you you're throwing the ticket over over the fence. and And I think one of the problems that causes is that is ah lack of empathy between engineers, analysts about what they're both doing. And and that, as you said, adds to this time to delivery and the whole point of you know being able to work quickly to ship insights I think is all you know having a a minimum viable product to your your stakeholder and for that I think everyone needs to be at that conversation from the get-go so
00:16:33
Speaker
if I've got an engineer saying, yes, we've got this data or we can get this in X amount of of days, weeks, et cetera. So I think that then if we can sort our out internally and have much better processes in communicating the value of each teams internally, that can then subsequently hopefully lead to to better communication of of the value and and quicker work to to the business as well. naturally, as we I think we get quicker at delivering and deliver better products and and i have this better communication internally. I think that naturally leads to so more results and and subsequently more more investment in the in in the data team and and what's possible because I know that's something which leaders often struggle with.
00:17:16
Speaker
So what sort of solutions are there for these problems around silos, data teams, silos within data teams? What approaches, if our audience are seeing this in their current organization, what what solutions do you guys have which could could help them to so break
Adopting Product Management Practices
00:17:35
Speaker
these down? I feel like it's easy to kind of see, see it just the data team who's sort of stuck in these permanent discussions on how to demonstrate value. Although I think I'm probably quite biased in that I listen to a lot of data podcasts and I read a lot of data blogs. Maybe I should ah listen a bit more broadly, but I think fortunately many have gone gone before us. I appreciate the start, you know, saying that we shouldn't exactly liken ourselves to a software engineering team, but I do think that there is,
00:18:02
Speaker
you know a lot that we can understand from how product and engineering have approached communicating technical value and and working on sort of the most important things and really understanding problems. I think as data as a field matures, it's maturing in a really positive direction. um And Karen and I came to discuss this because we actively seek out the same role, basically. We have very different approaches to it, which I think is what what leads to such great conversations between us. But and we we actively seek out roles where there is that really significant element around helping teams move away from these sort of service desk approach.
00:18:42
Speaker
building our effective teams that focus on answering the business questions, creating that voice, all that Karen was talking about. And where we we feel like we're seeing these profiles most recently is actually in what I feel like is a new role, although I'm sure a lot of people will tell me it's it's not. Um, it's a new role to me at least, which is the data product management. And Karen is actually, ah has just started, I guess, when this comes out a role with this official title. So really, really excited to see what that goes. And I know that you'll talk a little bit more about it later. Um, I somewhat stumbled although very happily into a role that I would typically describe as a PM role for for a lot of my responsibilities at least. And I do feel like it's a role.
00:19:24
Speaker
that could really help teams emerge from the scenarios that we discussed earlier. So I'm just going to talk, I think it just a little bit about what I think a data product manager is, and then maybe Karen, you can follow up with your thoughts on what that, maybe yeah how that helps these scenarios. So yeah, the question is then like, what what is a product data product manager? and And in its simplest form, my interpretation is it's the person who determines what should get built and why. So this is the person that will build up those relationships around the organization within data and will understand what are the business's strategic aims, what should the priorities be, what is it that we are all actively working together to achieve. With that, they will then go and generate that vision.
00:20:13
Speaker
for the data team that they work in should be creating. And obviously, they'll be creating data products. And that's something where, I mean, I don't really know what the definition of this is, but again, well, my interpretation is like, is it something where like data itself is the output? And this is where, you know, I guess I'd really love to hear what you think Aaron and Harry is. It's like, for example, like a recommendation system because the output of that is actually the piece of data that then goes and tells you, you know, this is what we think you should watch or, it's a forecast, you know, in in my world of of payments, you know, this is the number of transactions that we would expect. And that's what the business then takes and goes and makes a decision on. And obviously, it's something that isn't just a one off, it's something that's, it's a product, it's maintained, it's got a ah long life, you know, same would be thought of as
00:21:03
Speaker
and the automation of self-serve insights and dashboards, or the creation of something like an experimentation platform. I think you can you can think about all of these in a ah product sense, and they would be built um and led by the data team. And a really key part of the role that I think is what really helps overcome the scenarios is the fact that they have a huge platform, in my opinion, to encourage the adoption of data, to Our re-advocate for trust in the output, which I know is something that yeah we struggle a lot with, and facilitating the education around it. So, you know, what what is the data literacy of the organization? How can that be improved? And how can that, you know, encourage adoption of the output and sort of creating that data drive into a culture internally, all of which
00:21:52
Speaker
I think it gives you the foundation to really start communicating what you're doing as a team and and what your impact is. But as I say, very intrigued to hear your take Karen, especially someone who is officially doing this role. And maybe after Harry, you know someone who you couldn't could be hiring for this role. Yeah, I think I socially agree with what you said and to me it's about building bridges and you know when you're talking about trust and education, this is absolutely key in these areas. So very often what happens is, and unless you're in a startup where people are already
00:22:28
Speaker
very close to data and engineering and all these kinds of things. Very often what happens is we come into an organization that is well established and then they have their own ways of doing things and then the data team comes and say we have to do things differently because leadership said that. So before we can actually do things differently and and encourage people to do things differently, we do we need to create this trusted relationship. With our trust we can't actually work together and to me, it's about basing bridges. And as I said before, it's really understanding the why and the what. And I know we don't have a lot of time to do things at work, but spending time with key stakeholders, people who are going to make decisions, people who you are going to work with on a daily
00:23:14
Speaker
basis who are not part of your direct team is very important because that will help with the trust element, that will help with them understanding data a bit more, so with all the literacy, etc. But also, it will help you understand their processes. How is something working from the start to the end? How can that be optimized? And Riana, when you were talking about the definition of a data product, I see data as being an enabler, where I've started in the most of my career. It's about data for decision-making. If you don't understand how people are going to make the decisions,
00:23:54
Speaker
what's the driver to making one decision, then you can very easily give them the wrong product, the wrong data points, the wrong definition of a metric. and Understanding all of these is very key. I feel like what it processes, is it's probably the least sexy of of the the whole thing we described, but it gets me very excited. i When you understand really, how I don't know, if you take the example of a customer service, you have a message that is urgent. How is urgency defined? When is it defined in in the whole process? And that will just help you sort out your SLAs. For example, I have a really good example of that when someone defined urgency. So
00:24:40
Speaker
what is the service level agreement for urgency three years ago, and then loads of things change on the engineering side. And then that meant that that definition was not relevant anymore. And then we go and we're like, oh, why does it not work? Why are we not able to prioritize urgent messages correctly? And that's all because the definition was outdated. Loads of things happen and no one understood the processes of how do we go from a customer sending a message to us understanding that is urgent? And then once you get that, then the data flows so easily. So understanding processes might not be the first thing that comes to your mind, but it's very important in this kind of situation. And I do believe that
00:25:22
Speaker
A lot of these types of responsibilities, so building the relationship with the stakeholders, understanding you know the whole what the strategy should look like, what it is that we should be focusing on, what it is that they're trying to achieve, and also understanding the processes and managing the team, relies a lot with the head off. But i'm just I just want us to ask ourselves really, how much realistically can we expect from the person heading up the team? Can they really do all of these things as being one person and then proving the value of data, pitching for more money, more hiring, and and all these kinds of things? And that's when, for me, the PM role becomes very important.
Challenges in Hiring Data Roles
00:26:06
Speaker
I completely agree with you, Karen. I think, as Rihanna said to to me as well, the the data product manager for your role has definitely become more popular and it's become is is there a hot new role on on the market. And we're seeing a lot more of these roles pop up, but they tend to be in larger organizations. And the skill sets that you've mentioned, Karen, about you know what makes that role, I think that the skill sets that analysts should be taking into their into their toolbox naturally anyway and and playing some part of their skill set and their toolbox. But as you said, the amount of times that we see smaller organizations who are looking for a head of data to play, they want them to be hands on. They also want them to work with the C-suite to identify opportunities. They want them to manage the team. The list goes on there. You you have these unicorn roles, which is is a lot to ask for people. and And what the reality is, is what you get is is someone that is
00:26:59
Speaker
either a jack of all trades or it's just completely stretched. And and i don't think I think it's unfair to to ask that to to be done ah of one person. So I think the the interesting piece is about how how do you as ah as a data leader get that buy-in from the the leadership to be able to demonstrate the value of this data product role because I think it can be placed into more organizations because that at the moment, heads of are doing it themselves. But it's not something which I think I'm seeing ah adopted by smaller organizations at all. it's It's really focused in on
00:27:37
Speaker
the large organizations, large data teams where we're we're talking sort of 100 plus data professionals. So it's it's it's a much easier to to get by. And is that something you that you've seen with the smaller organizations and the the the over-utilization of ah of a head of or managers? um For me, i absolutely. so and and What is sad sadly, the head-off would be more focused on one area rather than another because they can't do everything. right we they They will need also to be able to do their job, demonstrate the value of their work without burning out. and
00:28:14
Speaker
What I'm seeing sadly is that what I see very often happens is that the part of their role that I guess gets the least attention is the developing of the talent in their team because they're so focused on working with the business to make things happen, deliver the work that then analyze data scientists, data engineers, that their developments. Not it's forgotten, but it's a bit deprioritized. And that's when when you do engagement survey, you see all the the team members saying, I don't see what the future looks like for me in my organizations and all these kind of things. We don't want this to happen. We don't want people to leave. We want them to feel like they can thrive in an organization, grow, learn new skills, and apply them. But you're absolutely right. so I've worked, many apart from in the last
00:29:04
Speaker
year and a half. I've worked mainly in smaller organizations. It's not there was no space for the data product manager, but it was very difficult to justify the money and the hiring the extra present because everybody would be thinking about, yes, but what I need is the analytics or what I need is the data engineering engineering and the product, sorry. And having an extra role for them is not about having that. They're thinking, yes, but the analyst can give me that. Why would I need a product manager, data product manager, to then ah help with the delivery? And what I believe is,
00:29:47
Speaker
really all the things that the analysts are doing. It would be so much easier, so much faster, and you would get so much more value from it with a data product manager. So it is an investment. It's just like, I mean, we we could still be i I don't know, communicating with letters and saying, why do I need an email when I can send a letter? So you know it's loads of different things like that where you really have to see it in action, understand how it works, what value it can add for your team so that you can then pitch for the budget for it. I remember chatting about that with Rihanna a couple of days ago, thinking about well where do we find the money for it in smaller organizations. But yeah.
Treating Data as a Product
00:30:33
Speaker
I think it comes, it links in nicely obviously with breaking down the barriers between the business and the and the data team as as well. as That's really the, it sounds like that one of the key roles. it One of the key responsibilities is this role plays and I think you're able to if you're able to to have a role like that which is able to spot opportunities and um treat data as a product and keep all the ah team ah aligned. it Product development has been around for a lot longer than than data teams and and why do you think there are product managers there? They are the ones that are orchestrating it and ensuring that product gets delivered. So I think the more data teams can transition to treating data as a product and the products that they build, I think the use case will become more and more just of the obvious choice.
00:31:20
Speaker
I think because I'm sure you've probably got some examples of that in Zeps, Rihanna, is the role that you've played and and the how pivotal it's been and in ensuring when you've got however many deadlines to to hit without without someone like yourself orchestrating that, the the wheels fall off. They do, you end up very much in the scenarios though that Karen and I sort of outlined at the start, and but just to to go back to the point of kind of the organizations that we work in, I too have really only ever worked in small organizations. Zeps is probably not considered very small anymore since it's the merger of two startups into one sort of getting very quickly large with scale up. But for me, I've really challenged that point of it's only for the big companies because, yes, Zeps is not that small, but it's also not that big. It's not as big as, say, B&Q where Karen is or some of the much, much larger
00:32:13
Speaker
sort of institutions. And it really comes down to ah not focusing on this sort of, this need when you're designing teams or hiring, you know, people with crime say, you know, you just want the analytics, they just want to see the output. And, you know, I have seen that firsthand in nearly every team I've worked in that if you don't really understand what it is you're doing and why, you end up in that scenario of 85% of the the output that gets built, whether it's a product or not, doesn't ever really get used, or it won't be maintained, or it's not really utilized to its full capacity. Because there's simply no one there really advocating for it or maintaining it, or actually treating it like you would a product. And it feels like data is
00:32:56
Speaker
is that area where we haven't embraced that, I i would call it a product way of working, but development ways you called it, Harry was still trying to hire a lot into one person. And when we think about engineering leaders, like they are almost always working alongside a product leader. and data engineers, they scientists, knew they are very technical people, they are essentially the equivalent of a software engineer or developers and the engineering lead and that product lead, they have complementing roles with complementing skills and they pair that what and the why that we've sort of put under the product with
00:33:33
Speaker
with the how, and that's what creates this transparent roadmaps and these products that we use every day, but also continue to use every day. so If it's a recommendation system or maybe more is in the case of Zeps, a volume forecast or a transaction forecast, that's something that isn't just something that we produce once and then move on. It's something that is actively maintained and is productionalized by a team of engineers led by that by a product manager. and It's fairly early on when you get back to the point of a small company, it gets really early on that you'll be hiring a product manager to drive that product vision. That's probably one of your first hires if it's not the engineer that they would be supporting. and I really do think that there's a need for someone to do the same but for data because you've hired your yeah it's not going to be your first hire. You've hired a data engineer, you know you've hired your data scientist, um but getting that buy-in is really difficult for data. Communicating isn't easy. And challenging existing ways of thinking really isn't easy. you know All these points that we've we've talked about, and if you then at that point hire someone who can do that, you are enabling the data engineer that you've just hired and the data scientist to actually you know do what they want to do to focus on what they need to do for their sort of technical skills, but actually have them working on something that ultimately supports the business. And I would consider our sort of, we have an ML and AI team, they've previously called the data products team,
00:35:01
Speaker
I'm not their PM, so very much someone else doing a fantastic job here, but I would consider them just one of the most high performing teams in the organization because they genuinely understand what the business needs, how that crosses across different domains, produce products for them, typically around an hour, so more of the forecasting. side and then maintain them ongoing as a sort of full stack team. And the success is really, you know, I think not just in obviously the really talented individuals that produce the work, but that the PM who is able to really understand what the business needs then and how we can achieve the most impact across the most number of sort teams or people or or financial figure for the time that we have.
Applying Data Strategies to Small Teams
00:35:47
Speaker
yeah I feel like I'm going really quite deep into something, but it's part of that maturing that I think the data is doing generally. And so, you know, overall, I just really challenge like, this shouldn't be something only for for larger companies, it should be, I think it can help make smaller teams actually much more lean and efficient. I agree on on that and I think engineers and and technical professionals, I think it's fair to say, do you maybe have a tendency to over engineer, to over develop, to chase perfection in what they're doing and and quite often when you're over engineering, it might be excellent and very scientific and and a very detailed, but it it might not be driving the the value that the business needs. so
00:36:28
Speaker
you need someone that can reel these professionals in and and make sure and just keep on ah aligning that that what they're doing is is touching base. And I've got a real world example of that. we've We've been recently working with ah a startup who've got ah building a new data platform and and they wanted a head of data platforms to to build out a new platform, but also work with the with the C-suite to spot opportunities and you know identify what use cases for data or all of the rest of it that that we've discussed. And we we we sat them down and said, you're looking for two very different types of of profiles here.
00:37:05
Speaker
you either going to need to find one of these unicorns that maybe do exist but they're they're very hard or we can split this down into into two roles, maybe bring down the seniority of your your engineer and I have the addition of of someone that has a bit more of this commercial where was like the the data product manager and and they actually after after a few weeks of realizing that they weren't finding the right profiles, they that they did adopt that strategy and and actually pivot to to hiring someone that was much more of a, they didn't actually call it a data product manager, but that was very much the the remit of their role. So I think it's just about that sort of coaching. And I think sometimes organizations have to go through the pain when they are hiring to really sort of understand it, especially when these are the first data hires, you know,
00:37:51
Speaker
we're we're struggling as state people within the data industry so to outline this. So we say it's only going to be harder for for people that aren't involved in our industry. And and I think it's our job to to sort of coach them along the way and and show them why that's possible. I love that. I was thinking, so you know how obviously the the economic situation is a bit tough at the moment and so is the market. So when you can hire, if you think about it, that that would be great and having partners like like you that can support and help really understand what kind of roles you need to have and what can you realistically find is great. but
00:38:28
Speaker
If you're in a position where you cannot hire right now, there will be someone in your team that will have the skills. So thinking about maybe how do you restructure your teams to make that role happen. And it does not mean having to change the titles and all these kinds of things. It might just be around responsibilities and then looking at how do we make the workflow and how do we define ah a piece of work and create a product in the organization from from that perspective would be a good way of doing it as well. and I mean, that's that's actually what I did, essentially. You know, I say I sort of stumbled in very happily. I'm still technically, in fact, I actually don't know what my title is. I do a lot of different things and that's what I personally enjoy, but also know that you know a lot of what I do does seem to fall under this data data product manager sort of title, whether it actually is or not. It's just acknowledging the the need for for differentiation in those skills and sort of this move away from hiring everything in ahead of. So really appreciate the the work that you're doing there.
00:39:32
Speaker
Yeah, know I think, Karen, your point is spot on there. are People, as as I said earlier, analysts, even engineers that have to have this that naturally have this this product mindset, and I suppose it's just fostering that a bit more in in in everyone and and maybe aligning someone to to look after that. that mindset a bit more and and focus in on and also that that type of roles and and responsibilities because as you say not everyone in the the current space is is ready to hire but it's obviously as we set out at the beginning so important to be aligning what the data team is doing with the business and and communicating that value to to the business so yeah I think that's been some great great
00:40:11
Speaker
thoughts and and examples from both of you. I think that's ah and a nice place to to wrap up unless you've got any final thoughts. Nope. so Well, look, thank you ever so much guys for for joining me. I've really enjoyed it. And yeah, I think obviously it's going to be a role that we're only going to see grow and and hopefully the listeners have taken a ah lot away and can maybe start looking at how they can maybe not introduce this whole new role. But as as we've said, just introduce some of the the responsibilities and and maybe draw their focus to to this. Thanks for having us, Harry. Bye everyone. See you next week.
00:40:50
Speaker
Well, that's it for this week. Thank you so, so much for tuning in. I really hope you've learned something. I know I have. The Stack Podcast aims to share real journeys and lessons that empower you and the entire community. Together, we aim to unlock new perspectives and overcome challenges in the ever evolving landscape of modern data. Today's episode was brought to you by Cognify, the recruitment partner for modern data teams. If you've enjoyed today's episode, hit that follow button to stay updated with our latest releases. More importantly, if you believe this episode could benefit someone you know, please share it with them. We're always on the lookout for new guests who have inspiring stories and valuable lessons to share with our community.
00:41:34
Speaker
If you or someone you know fits that pill, please don't hesitate to reach out. I've been Harry Gollop from Cognify, your host and guide on this data-driven journey. Until next time, over and out.