datamaster summit 2020

Building Data Teams for Tomorrow’s Challenges

 

Salema Rice

CDO @ Geometric Results

When building successful companies, there are generally three elements to this success: great people, streamlined processes, and key core and integrated technology. Salema Rice, CDO at Geometric Results, and her team are focused on answering the question: How to have the right team and processes to win our data challenges?The answer is simple, it’s in the data.

By delivering better hires to companies through innovative data analysis, companies can ensure future success. Coupled with the right processes and tech stack – an organization can turbo-charge its data and analytics outcomes. Join Andy Palmer, Tamr’s CEO, and Salema Rice as they dive into the people part of data.

Attendees will learn:
The key elements for building effective data teams
How agile methodology plays a role in modern data organizations
How the combination of people, processes, and technology accelerates outcomes

Transcript

Automated Voice:
Data masters Summit 2020, presented by Tamr.

Andy Palmer:
Hi, this is Andy Palmer and I’m thrilled to be here at Data Masters with Salema Rice. And Salema is the Chief Data Officer for GRI, Geometric Results Inc, which is a Bain Capital company. And she’s a member of the company’s exec management team. Salema is a senior exec with over 25 years of experience directing data management strategies. She’s truly a pro and she was recently listed as CDO Magazine’s 2020 Global Data Power Leader. And it’s really a thrill to see you again Salema and really excited to spend some time talking. Thanks for being with us.

Salema Rice:
Thank you, I’m excited to be here today.

Andy Palmer:
So, one of the things that we’ve talked about in the past, and that is so compelling in my mind is your integrated role in, not only managing all things data related, but also how data integrates into product at GRI. Could you talk a little bit more about how your role is not a purely technology job, but it’s really integrated into the core of GRI’s business and products?

Salema Rice:
Sure. In fact, I was just thinking about that today. And we were talking about organizational structures and while my first passion and love obviously, is data and that’s my, quote unquote, official title. I’m really rather the Chief Technology Officer and the Chief Product Officer, as well as the Chief Data and Analytics Officer for GRI. So, data is really the core of everything we do. It’s our differentiator in the industry. And I was actually brought in by Bain Capital specifically, to differentiate GRI as a managed service provider with our data and analytics product offering. So, product development, managing our data teams from an agile perspective to really drive a difference for our customers is my number one goal and our opportunity at GRI to really support the customers that depend on us.

Andy Palmer:
It’s amazing. Can you talk a little bit about how… I know agile is really important to you and most people associate agile with software. Could you talk a little bit about how to manage data using agile principles?

Salema Rice:
Sure. So, agile methodologies plays a huge role for me in my organization because it’s really a combination. When you think about data engineers and data analysts and data scientists together, create really that core development team. The engineers, really having the capability to ingest any data from anywhere, in any format and the analyst really at the front end of that, as well as throughout the entire process, being both the liaison to the business, as well as having the core business knowledge of the data, and then the data scientists really being there to provide the value that we’re going to drive home. But those three, to this day, even prior to COVID, we are a complete remote shop, my entire data organization. We still hold daily stand-ups, we do it remotely every single day. We’re reporting, what were the things you worked on yesterday? What are you working on today? What risks or impediments? And that way, we’re not waiting until the end of the week to find out, “Somebody didn’t call me or I had to wait for this piece of information.”
And from my organization, it’s been an absolute game changer. I have from just a data perspective, I mean, not just technology, but from a data perspective, I have full responsibility for all data. So, that’s master data, data operations, data management, the technology of the data, the storage of the data, how we architect data. So, each level of our programs methodology has to be very agile. We often, will deliver products on an MVP solely to get something out there and then look at opportunities to spend the next several sprints really refining it, so that we’re constantly providing a value to both our internal customers, as well as our external customers.

Andy Palmer:
Wow, it’s fascinating. You’ve mentioned MVP, right? Is there a similar idea for data, where it’s like an MVD, like your Minimum Viable Data required?

Salema Rice:
Absolutely. So, when you think about, what’s the minimum amount of data versus all the data, right? So, for many products that we develop, there are specific pieces of data or specific types of master data that we might need to develop a product. I mean, one of the challenges in the human capital area, which we are, is that the lack of standardization, the lack of taxonomies, the lack of mastering things like job titles, mastering things like skills. At one point, I think we had 35,000 skills in our catalog. So, I mean, when you think about what you need to master in order to deliver an MVP, versus if you’re pulling data, you might have larger sets of data that you’re going to bring into your raw zone, but then what actually gets transformed and mastered and ultimately gets put into the semantic layer, is something much more refined. But, at the lowest level of granularity, we always start with a very large data set and through an MVP process, we can actually get precisely to just the data that is needed to solve a specific business problem.

Andy Palmer:
It’s incredible you say that because this idea that there’s either small data or big data, but what you’re implying, it’s like, “You have to manage all of it because you never really know which subset you’re going to need in any given situation.

Salema Rice:
That’s right.

Andy Palmer:
Yeah.

Salema Rice:
Yeah, just when you think you’ve got it all figured out, they’ll come up with something else. Especially in our world, a lot of our data comes from third parties, right? So, when you think about the large vendor management systems or acquiring data from third parties, these are huge data sets that you’re bringing in to work with and you don’t always want to take the time to transform it, but to go all the way back to the source, would be very difficult to do on a regular basis. So, having that raw data available to you, on an as-needed basis, is really important.

Andy Palmer:
That’s great. But, you imply… When you said, “Mastering” and that always near and dear to my heart, at Tamr. So, some of these mastering challenges that you described are just massive mastering skills, I think is a great example. How do you think… When you think about tackling these data mastering challenges, what are the methods that you look to and what are the things that you try to do when you’re doing big mastering projects?

Salema Rice:
Well, one of the keys for us is really not reinventing the wheel. So, we really look to organizations like SOCK and Oxnard, and the Department of Labor, the Bureau of Labor Statistics. Because, a lot of times when we’re looking at rate data or market data, things we want to bring in from other sources, even metrics we want to bring in from other sources to be able to do benchmarking. It’s really important that our data be consistent, that our reference data specifically for mastering, is consistent. When you think of sourcing methods or something as simple as an end reason code, right? I mean, in order to really have a proper benchmark, you need to be comparing apples to apples. So, research upfront is always a key component of that. Another would be really having the business partake in that, being a part of the business. We always talk about the importance of, “Data is not an IT problem.” Right?
I can’t stress enough to organizations, when I hear that the CDO works for the CIO, and I think, “Data cannot be an IT problem.” IT can own the processes sometimes that manage the data, but the business has to take ownership for the data. Otherwise, you just got garbage in, garbage out, right? So, I think that really… The other big challenge I think, for mastering the data is making sure you have sound governing principles in place. One of the things we learned back in, I don’t know, probably dating myself. But I guess, ’08, ’09, during the financial crisis, was the importance of a taxonomy and really having standard taxonomies for things like skills or even job titles. When you think about what happened from the financial side and really having that documented, it’s been a huge differentiator for GRI. When we go in somewhere that had one of our competitors previously, I hear all the time, “This is night and day” and that’s because we apply all those principles.
We can actually show our customers and use it as a competitive advantage to say, “Here’s where the data came from, here’s how it’s being mastered.” Here’s the lineage of data, how it starts from a system of record, all the way to the target. And it’s a concept that in my industry, has never been done before.

Andy Palmer:
It’s amazing how compelling the quality of data and the lineage of that data can be a competitive advantage, but we see this playing out across all industries and you guys are really at the cutting edge of it. Could you talk a little bit about… So, I know one of the other things that I love about what you’ve built is you really relied on the cloud, right? Really heavily, it’s core, right? These core principles of agile and cloud, could you talk a little bit more about how you’ve embraced cloud and how that has enabled you to do more faster trade-offs with regards to on-prem?

Salema Rice:
Sure, everything we do is a hundred percent cloud-based. So, we were very fortunate that when I came on board, Bain really gave me the leash. I joke with my team all the time and say, “I’m the CEO of the data team.” But, the reality of it is they were like, “What do you have to do to be able to deliver at the right people, the right processes, the right technology?” And it needed to be fully cloud enabled. It needed to be able to be able to leverage updates and be able to access data, access our portal. We’re not just delivering data. So, while we are delivering data, we needed our data repository to be incredibly hybrid, right? We’ve got all kinds of data.
We’ve got data that we ingest from tables and columns, but we also have data that we’re ingesting from blobs of text and PDFs and SOW’s or service procurement documents and images, sounds. So, there’s a little bit of everything that you need to be able to do, as well as really architecting that for analytics. Back in the day, we started this completely from scratch. So, this was a manual process that we knew we had to really differentiate how anybody else was doing it. So, to be able to take the big data and the small data and combine it together, it’s been a lifesaver, right? I mean, to have data scientists as part of our regular scrum team, to be able to build out the products that are really changing the industry. Things like our talent cost estimator, right?
The one tool that in a matter of a couple of clicks, you can see where to find purple squirrels. So, think about, “I want a software developer that has MDM, but also has Python and GPS. Well, where are there concentrations in the country with that type of skill?” And within seconds you have both the supply and the demand. And then, you want to know from a small data perspective, “Well, how long does it generally take to fill that position? What should I be expecting? How long they’re going to stay.” So, within one model, we can actually see, where to find that talent? How much you might expect to pay for that talent, based on the level of expertise that they have. And then ultimately, what’s it going to take for me to get that person? And what can I really expect long-term? In terms of how long they’ll stay with me.

Andy Palmer:
So, not to extend the metaphor, but it’s a super-powered sniper rifle for purple squirrel hunting.

Salema Rice:
That’s one way to look at it. I think the other thing that… Architecting the data for analytics in the cloud also gave us the ability. When you think about reporting and business intelligence, right? I think that there’s always been this concept of the rear view mirror, right? Here’s what happened. And until we became fully cloud-based and really started architecting the data for analytics, we couldn’t get to where we could show a customer, “Here’s what happened, here’s why it happened.” And then, using our data science, machine learning and AI, we can actually tell them, “Here’s what you can expect to happen next.” Or “Here are some actions that you could take right now, to change the trajectory of where you’re heading.”

Andy Palmer:
So, it’s really enabled you to become more predictive and prescriptive, as opposed to just-

Salema Rice:
Absolutely. I mean, everybody does descriptive and some companies do it really well, but being able to deliver predictive and prescriptive, we really had to be able to have all of the data available to the data scientists, coming up through and up from an analytics perspective. I mean, it used to take 85, at least 85% of the time, just figuring out how to put the data together, to build a model for analytics or statistics or anything like that. So, to be able to have that already pre-done and ready for our data scientists, they’re able to produce models, and ultimately we can deliver more products in a much shorter time period than I personally, have even ever witnessed in my career.

Andy Palmer:
Right. That’s amazing. So, switch topics a little bit, talk about talent on your team. And I would imagine one of the benefits of being so fully remote, is you can recruit people that are anywhere. Talk to me about the people you’re recruiting, where are you finding the best pockets of talent and what are the skills that you like and prefer today? And maybe contrast it to all the data management people that we used to hire and staff with all the time. Yeah.

Salema Rice:
So, I mentioned three of them, our three cores are really data engineers, data analysts and data scientists. We look for engineers that are capable of ingesting any data from anywhere. So, I like to have data engineers that have a good combination of Oracle and SQL, but also have big data, right? So, they can use Python, maybe they… I’m less concerned with what tools, and more-so in how they think. I always tell people, I want to know, what are you doing? What are you working on when you’re not at work? Right? Are you using data in your personal life? Right? Because, I need people who are creative and innovative, and those are soft skills that we look for across the board, right?
We’re an incredibly collaborative team. So, if we branch off into discussions, I often tell people like, “Put your job title to the side, and you’re problem-solver number one and you’re problem-solver number two and you’re problem solver number three.” Right? Because, it’s really that collaboration and innovation and creativity, is why they stay. So, I can find engineers and analysts and even my directors and product managers, but at the end of the day, when they’re getting ready to call it a day, and they get called by recruiters all the time, they stay with us because one, we really think of each other as family, right? So, while we have great respect and trust for each other, but on the other side of that, they also know that this is an opportunity to have true creative freedom and collaboration. And so, it’s not just about the technical skills, right? It’s really some of those soft skills that make them a good fit for my team.

Andy Palmer:
Cool. Well, that’s great. Are there geographies where your team is concentrated or are they all over the map?

Salema Rice:
They are all over the map. We have pockets though, I would say. We’ve got quite a few on the East Coast, just where we’ve been able to find some really good talent, Mid-Atlantic and East Coast, but we have some in Denver, Texas, Detroit, London, India. So, we’re all over, but we support customers all over too. So, it’s nice to have… It makes daily challenge, a little challenging sometimes, trying to get a good time frame for everybody. But, other than that, it works really well for our organization because we’ve got people working around the clock.

Andy Palmer:
What do you guys use for online collaboration tools? Are you a Slack shop or are there other tools?

Salema Rice:
No, we actually use Microsoft tools. So, we’ve been using Microsoft Teams and Microsoft Planner.

Andy Palmer:
Oh cool, what else?

Salema Rice:
We were heavily in using Jira, but this year we transitioned to Microsoft and we’ve really been trying to… All of our data now, where we actually store the data is in an Azure Repository, that we built out. It’s actually more of a data factory, so that we have that hybrid flexibility. And they’ve been a great partner with us through this to really provide good tools and good collaboration for the team.

Andy Palmer:
Yeah. We feel the same way at Tamr. We’re a team that runs natively on Azure now. And Microsoft is also a customer of ours, doing customer data mastering inside of Microsoft. And so, I’ve been really impressed by the Azure platform and how far it’s come in the last couple of years.

Salema Rice:
I completely agree. Previously, I used a lot of other independent tools to do independent things, right? For small data versus big data and it’s really been wonderful to have everything in one environment.

Andy Palmer:
Oh, that’s great. So, as you start to think forward, right? The next three to five years, what do you think are the biggest opportunities, both on the business side and the technical side?

Salema Rice:
Well, I think that we’re going to see more big data, right? I mean, one thing that COVID showed us is, when my governor got on TV and started talking about big data and data sets and how much data he had, I was like, “Look at this.” I think that people are really looking at data as the differentiator in every industry there is. So, I think that we’re going to see data embedded everywhere. One of the things that really fascinates me around the people process technology in our industry is that, we’ve been in business for probably 25 years, I guess. And in those 25 years, it really was about putting the right processes in place, putting the right expertise in place, right? To drive those processes. And then, the end result was getting data out that we could do something with.
And for the first 24 years, we’ll say 23 years, that’s where it stopped. And so for us, it was taking all that data and turning it into something, right? So with data, what information are you making out of it? And that’s really how we’ve been able to turn the corner on the future of this industry, by taking that right process, the right talent and the data, and building a data science, mastering everything on top of that data to ultimately get to the best results. To be able to… Every customer is at a different place on the maturity scale, right?

Andy Palmer:
Mm-hmm (affirmative).

Salema Rice:
But I think that more than ever customers are seeing the value for… I mean, this industry has changed so much Andy. When I first joined this industry, I would say talent was like HR or procurement, right? HR was your FTE data and procurement was your contingent data and it stayed there. Well, when we started building our products like this, now I have a monthly meeting with a global CIO of the largest OEM in the country, right? So CIOs, I meet on a quarterly basis with CFOs and that’s very different than what we ever saw, but what we can tell about data, what we can tell about where you’re hiring? Who you’re hiring? How you’re hiring? How are you retaining talent? What is your real cycle time? There is a war on talent. And so, finding the right talent, knowing where they’re at, [inaudible 00:23:20] mine. Everybody had the first question of COVID, is where are they all? Right?

Andy Palmer:
Right.

Salema Rice:
Where are all my employees? Where are they sitting? That’s been a… I just think that, we’re going to see more companies wanting to look at CDOs and think, “How can my CDO mentality really differentiate my revenue stream down the road?” And that’s something that… I think that even my organization looked at early on was the opportunity, we just think differently. I think we didn’t come from the business side, we came from a side that was more relational, right?

Andy Palmer:
Mm-hmm (affirmative).

Salema Rice:
And I think respect, right? My team is my family, right? And they look at me like that and I look at them like that and we have great respect for each other. And I think transferring that to the broader businesses is… It’s a game changer in this evolution, right?
Working remotely, being able to work remote successfully. We’ve been doing it for a long time and I think companies are going to be looking more towards, “How do I build teams of the future that are remote? That are still creative and collaborative and innovative.” And I think that, those are some of the challenges, but I think that there are so many organizations like Bain out there, where we are doing this really well. And that you don’t have to start from scratch and do it all on your own anymore.

Andy Palmer:
No, that’s cool. Hey, going back to the product role and you mentioned the talent cost estimator, right? Really cool stuff. When your customers interact with the data that you provide them, how much of that includes analytics and models and things and in visualizations, versus do they want raw data or APIs to data? What’s their level of sophistication and what are they most interested in?

Salema Rice:
So, we don’t give APIs. We don’t give servers anymore, they still ask sometimes. So, one of the things I love about my organization is that I have a seat at the table. So, when we have a business review with a customer, or we have a internal business review, having somebody at the table to talk about data and talk about what is the business problem you’re trying to solve versus giving you the data. If I’m going to give you the output of a model and you’re going to download it to Excel, I didn’t do my job, right? So, we really look at our models to solving a pro… How did we help you solve a problem? Right? Probably, one of the first products that we produced was a rate health analyzer.

Andy Palmer:
Hmm.

Salema Rice:
And really the goal of the rate health analyzer was for you to be able to look at all the talent in your organization and be able to see how your rates are, compared to the market. And then, based on how you compare to the market, how quickly are you able to find talent? And how long are they staying with you? And that one motto was number one for, I don’t know how long. Just because, it very quickly allowed you to draw your eyes to where something was out of color, right? So, if it’s red or if it’s dark blue, it draws your eyes right to where you want to go and then you can see, “Okay, well, is this a job that is part of a job family that is really critical to the success of my organization? If it is, then I need to look at this further.
Maybe I’m okay for some positions to take 50 days to fill, but for critical positions, I want a time to fill of less than 20 days and where they used to stay with me for four years, it’s not acceptable if they’re leaving after 14 months. So, we really want to keep the talent for as long as we can at a competitive rate.” And rate’s not all of it, right? I mean, just like with my team, and it’s not just about the money, it’s about providing an environment that they want to work in because otherwise, you’re going to continue to have attrition and turnover like you see on the EMT side.

Andy Palmer:
So, it really sounds like, I mean, the data is the foundation, but you’re delivering real analytic outcomes to your customers all the time and-

Salema Rice:
Absolutely.

Andy Palmer:
Yeah and-

Salema Rice:
Yeah, every model we deliver is a step on the maturity scale to optimization. And each customer is at a different place right? I mean, there are times when I’m super excited to go to an organization that I think is really going to be, just on the cuspid of they’re going to love data and instead they’re like, “Oh, just tell me what my average time to fill is.” And then, there are other organizations that are just like, “We’re going to develop the world, right? We’re going to create new models with you, we’re going to challenge you.”

Andy Palmer:
That’s awesome.

Salema Rice:
You never know what you’re going to expect because like I said, “Different.” And some of it is historical legacy. I mean think about, you’ve got a lot of companies in the country that have been out there for a hundred years, right? I think though, that what we saw this year with COVID and how they see data now versus a year ago, is a game changer.

Andy Palmer:
Amazing. Well, Salema this has been amazing. Really appreciate you joining us for Data Masters and it’s always so inspiring to talk to you because you’re truly on the cutting edge and teaching the rest of the world how to do these things. And it’s really an honor to have you with us today. Thanks for taking the time.

Salema Rice:
Thank you, the feeling is mutual. Anytime Andy, you got me.

Andy Palmer:
Thanks Salema.

Salema Rice:
All right, take care.

Andy Palmer:
Bye-bye.

Salema Rice:
Bye.