December 18, 2024 • 30 mins
In this episode of Approaching Infinity, we're honored to host Maria C. Villar, co-founder and managing partner at Business Data Leadership, a trailblazer in enterprise data strategy, governance, and transformation. With a distinguished career leading data initiatives at SAP and IBM, Maria brings a wealth of experience to the conversation.
Maria shares her journey as a data strategy leader, reflecting on the pivotal moments that shaped her approach to data management and business transformation. Her career path offers invaluable lessons on the power of resilience, adaptability, and visionary leadership in the ever-evolving world of data and analytics.
Listeners will gain actionable insights on crafting a winning data strategy that aligns with business goals, the role of people and culture in successful data initiatives, and the importance of continuous data management. Maria also dives deep into the role of AI in data governance, sharing practical advice on balancing innovation with ethical considerations. Her perspective on "managing data as an asset" challenges traditional mindsets and provides a new lens through which organizations can view their most critical resource, data.
Join us as Maria unpacks her wealth of knowledge on enterprise data strategy, offers fresh perspectives on AI's role in governance, and delivers timeless lessons in leadership from her tenure at some of the world’s most influential organizations!
Support and Connect with Maria C Villar
● LinkedIn: https://www.linkedin.com/in/mariacvillar/
● Business Data Leadership: http://www.businessdataleadership.com/
Follow Approaching Infinity
● YouTube: @approachinginfinityshow
● Instagram: @approachinginfinityshow
● TikTok: @approachinginfinityshow
● X: @Justin_777_
● LinkedIn: linkedin.com/company/approaching-infinity
● Facebook: Approaching Infinity
Hosted by Justin Roopnarine
● LinkedIn: linkedin.com/company/approaching-infinity
Thank you for tuning into Approaching Infinity! Stay engaged for more thought-provoking discussions that redefine finance, technology, and innovation.🚀
Disclaimer: Approaching Infinity is an independent media production for educational and entertainment purposes only. Nothing in this episode is investment, legal, or tax advice, or an offer or solicitation to buy or sell any security. The host and guests may be principals of, or affiliated with, investment firms and may hold positions in the securities or instruments discussed; their views are their own as of the recording date. Mention of any company, security, or strategy is not a recommendation or endorsement. Do your own research and consult a qualified professional about your circumstances. Past performance is no guarantee of future results.
Listen
Watch
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:00):
Welcome to Approaching Infinity,where we focus on the latest in
finance, technology and innovation.
We present you with interesting and unique stories, news and
interviews to help you understand your limits and how
to exceed them. Now here's your host Justin
Rochner in it. Welcome back to Approach
(00:24):
Infinity. The show refocus on all things
finance, innovation and technology.
Today I am pleased to introduce Miss Maria Villar.
She's the Co founder and managing partner at Business
Data Leadership, which is a firmcommitted to promoting effective
data management, governance and analytics.
She has over 25 years of experience as a Chief data
(00:46):
officer from serving as Sapp's first Chief data Officer to
pioneer IB, Ms. first enterpriseData Center of excellence.
Finally, she serves as an advisor to numerous VC funds,
industry boards, and startups. And to top it all off, she's a
University of Miami grad. Miss Velar, thank you for
joining us again. Anything I missed?
(01:07):
No, you got it. I would just call it SAP instead
of SAP. Yeah, that is that kind of the
official name of the company. Yeah, I've heard it called the
SAP. No, when we were originally
looking for guests, right. And noting that this is more or
less Season 2 of our show, we want to guess who more or less
(01:28):
aligned or more senior in data positions to more or let's go
over various topics and creatinga data strategy going over what
that means. What are some best practices and
kind of the vision of setting that forward.
And you were the essentially thetop pick for that.
The fact that you were AUM grad more or less sold as on it as
well. So we brought you on to more or
(01:48):
less share your insights on a variety of topics.
But first one is essentially just for our audiences.
Knowledge is what is a data strategy, How do you craft one?
And then essentially how do you align it to grow the business
but as well manage data? Yeah, thank you Justin and thank
you for inviting me and and for the reach out.
It did come through LinkedIn, which is always interesting how
(02:10):
how well that that network does work to find the right
connections. And yeah, I spent 25 years of
my, of my professional career inthis topic of data when it
wasn't really a thing, when it wasn't really even a topic that
people and executives really cared about.
It was all kind of buried under IT.
(02:30):
But now with AI, now with certainly with analytics and
business transformation, the topic of data is much more
important than it's ever been. And it's, it's, it's great to
see it for myself, having spent 25 years trying to explain why
this is important to now get to a point where I think most, most
(02:52):
companies understand how important it is.
They just don't know how to manage it effectively.
And hence we come back to your topic, to this topic of a data
strategy. So to me, a data strategy is a
perspective, it's a point of view, it's a document that gives
direction on what is the data that most matters to the
(03:13):
company, what are the kind of risk and opportunities around
that data and how do you manage it effectively.
And you made a really good point, which is the way that you
start a data strategy is you look at your, the business
outcomes for a company. So I've got this master class on
YouTube, which is, I did about five years ago and it's still
(03:38):
relevant today. And it's a free asset that's
available to everyone. You don't have to be a, you
graduate just anyone. You can go to Google and, and
just search for business outcomedata strategy in my name.
And you'll see these 12 videos that really outline how to
create a data strategy. But the most important part, and
this I've learned through my three times of being a chief
(04:00):
data officer, that it starts with the business and the
business objectives. And you could be in government,
in which case if the business objectives are more mission
based, but nonetheless, right. If you look at where does your
organization want to go? What does it say publicly about
its direction in terms of eitherits annual report or what it
(04:23):
tells its stakeholders or what it tells outside investors or
what it tells, you know, the, the country, right?
What are the main objectives that they are trying to achieve
within that organization? And then you look at that and
decipher from that, well, if that's the most important areas,
then what's the data that is most important to those
(04:46):
objectives? And, and it's a process.
It's a process of really understanding how do you go
about affecting change in that transformation.
And transformation is the keyword.
You said it, most companies lookat data when they want to
transform their business, right.They're doing a digital
transformation there. And that could be a sales
(05:06):
process transformation that they're going digital or a
marketing or a supply chain or, or you know or something else in
their company or M&A that's verydata centric.
And then they'll go say, wow, I'm going to, I can't do that
transformation if the data underneath it, the foundation
(05:28):
from which all of this has to goand be converted isn't grounded
in high quality good data. And so really aligning those
business outcomes, those missionoutcomes to the data that
matters most is the start of a data strategy.
And that's the most important start of the data strategy.
(05:48):
So you do a scoping exercise, right, to define the that data
that's most important. And then you start to really
define what about that data has to change.
Is it the quality level? Is it that you need more data of
a certain category because you need to run analytics or AI?
So you kind of analyze the dimensions of data that have to
(06:12):
change and then the processes around data that have to change
and then the technology. So all of those things encompass
the data strategy. And that's, I think we're all
coming to really understand thatas as business professionals.
But often times, and I would saythe first, the last probably 10
years or so, everyone thought a data strategy was just a bunch
(06:33):
of databases and it was all technology.
But more and more, I think we'reall having the realization that
you can't just throw technology,build some databases and build
some reports. And that's the job that that's
not the job, right? The job is getting the results
that you need in the in the business with that data and
(06:55):
getting people to understand howto use data effectively.
And all of that encompasses muchmore than technology.
And all of that is what should be in a data strategy.
Excellent. Well said.
And just to summarize too, so essentially the first step is a
discovery process. What are your objectives or what
is the company's objectives? What data do you have on hand,
(07:18):
the quality level? What do you hope accomplish so
to say? Exactly.
Exactly. Let's say there's a case where
they're they have objectives or clearly defined objectives, but
their data quality isn't up to par.
What would you suggest in that case?
Like just go acquire new data orhow would you more or less
advise? Or does it truly depend on the
(07:39):
company and what they're gettingafter?
Well, ultimately it does depend on the company and the maturity
level and what they're willing to invest.
But if you, if you know that youhave a data quality problem, I
think the most and, and you, thefirst step is making sure that
you scope the data because oftentimes companies say, yeah,
I have a data data quality problem.
They go, and then they'll go offand try to solve every data
(08:01):
quality problem they have, right?
Oh yeah, my customer data is bad.
Let me go fix all my customer data When certainly that's a
noble cause on who can argue that you don't want good
customer data? But This is why going back to
the outcomes that you want to accomplish matters, because
those outcomes will tell you what about that customer data
(08:22):
has to change. Because if you try to handle
every piece, every customer field that there possibly exists
for a customer, you could be trying to clean 200 fields when
you really only needed 10 or 15.Because the outcome that you
needed, whether it was a marketing outcome, campaign
outcome is so you needed the good emails and maybe you
(08:44):
needed, you know, the good titlefor a come for a particular
person. You see, it depends on the
outcome. And the outcomes is so important
in the prioritization scheme because there will never be
enough people and resources to clean all the data and to do
everything that needs to be done.
So that's the scoping going backto scoping, scoping, scoping.
(09:05):
So once you know that you have adata quality problem and you've
scoped it down and you, you know, getting new data, look,
you only want to get new data ifit's something a real immediate
need and you don't have the timeto clean that data up.
But the problem is, but so get it and to do what needs to be
done, whether that's running a campaign, whether that's doing,
(09:26):
you know, special analytics or so if it's a one time data
acquisition, go for it and get the results and the credibility
you need as an organization. But oftentimes the problem that
you're really solving for is that the processes for managing
your data in your company are not right.
(09:46):
And so you get all this data in at one time and in six months
your data quality for that data is bad because you know the data
has decayed, the data has changed.
If you don't also include that data management process
transformation to constantly be managing your data to the right
quality level, you're going to end up in the same place.
(10:09):
So that's why a data strategy includes a lot of also process
transformation, data process transformation, you know, new
roles and responsibilities for people to manage that data
effectively and then the IT thatgoes with it.
So yeah, you got to start by getting that base clean, as
clean as possible, but you can'tkeep it clean if you also don't
(10:31):
do the transformation and have the right, you know,
architecture and and and roles and responsibilities to keep it
clean. Understood.
And would you say oftentimes that's easier said than done?
Or is the customer able to more or less all right once they get
their discovery meeting works out of the way, they set their
objectives, they figure out the more or less the data they need.
(10:54):
Do you think usually customers are able to more or less turn it
around, get the data and essentially gain insights,
analytics from it? Or is it truly just a a
continuous process where Oh my. Gosh, yeah, it's a, it's a
continuous process. It's an organization, right?
I mean, I, I've built a career around getting an organization
ready for maintaining that data.You can do a lot of things one
(11:15):
time, Justin, for sure, right? And, and yes, if you need that
special analytics, right, especially if you were during
Covad, a lot of things happened really quickly, right, because
you had to get all that done quickly to get the results that
you needed. I think a lot of a lot of COVID
pot organizations that they, themore ready you were from a data
perspective, the faster you could respond to all these
(11:39):
changing requirements until yoursupply chain being at risk and
to your employees aware all youremployees and their health and
well-being and all that. So, you know, I have built a
career and I coach and advise C level executives like CFOs and
Coos and CEO's on the importanceof, you know, commit to this
(12:00):
because this is an asset that you want to manage like you
manage people like you manage finances, like you manage your,
you know, supply chain. It is an asset that requires
resources, it requires processes, it requires
technology. And you got to put it up there
in terms of levels of investmentand it is the hardest part.
(12:20):
Going back to your question, what is the hardest part of all
this? It is the people part because
the technology, hey, there's a lot of technology out there that
will that, that will get you what you need, but really
getting the results and that result could be the business
outcomes that you're driving for.
People have to change their behaviors around data.
(12:41):
They have to make a decision. I think many executives that are
in data, they'll issue out, they'll have these great
dashboards and they do all thesegreat reports and do all these
great AI things and they give itto the business teams and the
business teams don't know what to do with it or they don't
trust it enough to make make thedecision.
So the last mile of data transformation is taking the
(13:04):
data and doing something different in the business that
creates the outcome that you're looking for in the business,
right? Whether it's improving sales,
whether it's improving your supply chain or whatever.
Data people can give you all thereports and data in the world,
but the business has to then take that data and make a
decision. And they won't make a decision
(13:27):
on that if they they don't believe the data is credible,
the report is credible, the organization is credible, and or
you know, they won't want to just trust their gut, not
believe the data. That whole aspect of
transformation is the hardest part that any data executive
will have. Excellently well said.
I agree on that point. The technology, at least what
(13:49):
I've seen is there. It's getting the people to adopt
that continuous mindset of like,hey, are you making your debt?
Are you cataloging? What objects are we trying to
meet and how can this data be used to more or less show
metrics or helps achieve those objectives?
It starts with the people. I partly agree with that.
Point is, and that's. Also kind of why I'm excited
about AI, right? Because I think AI from a data
(14:13):
management perspective, right? AI, if we can start applying
more and more artificial intelligence and automation and
digitization to all of the sort of tasks that you have to do
around data. I think you know the data
quality is going to get better the credibility is going to get
better and business owners are going to feel more comfortable
(14:35):
with the data that that they get.
And so I'm, I'm excited about AIin specifically the use of it in
data management and then. On the note too, like utilizing
AI, so we discussed data strategy, right?
I do want to pivot on to more orless inform our audience.
What is data governance? How does it relate to the
strategy data cataloging? Can you come more or less
(14:56):
untangle this web? First, you know all the
different data, so to say categories.
What is data governance and thenhow would AI essentially help in
this? So data governance often times
is kind of a, you know, a bad word in the mind of
organizations. They think of it as well, It's
the data police, right? So look, I, I think data data
(15:17):
governance is 1 segment 1 aspect1 chapter of your data strategy.
And that chapter of that data strategy is around what are the
policies and the standards around that managing that data.
I prefer to use the word data management rather than data
(15:38):
governance because just think about it, right?
A lot of what we're trying to dois to change behaviors and have
organizations, especially, you know, employees and business
people and managers think of data as an important asset in
their company. Well, you don't go around your
company and say, well, I'm goingto govern my people or I'm going
(15:58):
to govern my, you know, projects.
You say you manage them. And so management is the word I
like to use around this topic. And, and so data governance is
to really break down how you're going to manage data and what
are the standards that you're going to use, like whether it's
a data quality standard, whetherit's a standard for how you
(16:20):
create data. So there are process standards,
whether it's, you know, a standard for who has access.
So all of those kind of parameters around who and how
and what to manage is governance.
It's also your data policies. And then of course, it'll be the
technology that helps you to do that.
But that to me is, is what governance is.
(16:45):
And again, it's not, you know, it's not everything that's in a
data strategy. So it wouldn't cover like, for
example, all the training that you have to do, right?
That's not governance. That's training and culture and
culture training and literacy trainings.
That's what I mean. There is a place and a role for
data governance, and it's reallyimportant, but it also has to be
(17:07):
done in a smart way, Justin. So, and what I mean by smart
governance is that it has to be flexible.
It has to be tied to the problemyou're trying to solve.
So yeah, you could govern every aspect of data and you make it
so onerous that nobody wants to use it and nobody can do it.
(17:28):
This goes back to if you know what the outcome is that you're
trying to achieve with that data, then make sure the
guardrails are appropriate to towhat you're trying to get done.
So be smart about the standards,be smart about the guard rails
to make sure that they're flexible, make sure that they
(17:48):
can be executed and done, especially with automation.
If you can automate governance, that makes everybody's job
easier and and then, you know, be willing to change.
I think we need to go from the old fashioned governance was
very rigid, a very rule based, avery almost, you know,
governmental in a way. But the new governance,
(18:10):
especially as we think about AI governance or governance around
even analytics, there's also going to be more flexibility
there. There will be some hardcore
governance that you can't get out of because you need that for
let's say financial data or other kinds of data and privacy,
right? So there be some hardcore stuff,
(18:31):
but there be other opportunitieswhere you want to manage
analytic data or AI data a little bit differently.
Let me give you one example of that.
One example is, is having an AI,having AI principles, not just
AI strong rules. So a lot of companies start by
saying, the first thing I want to do is I want to give every
(18:53):
employee our AI principles so that they can start as they
develop AI or use AI. They understand what's important
for us as a company. For example, you know, our, you
know, we will be, we will have safe diverse AI.
We will have a human in the middle in AI.
(19:14):
You see, those are all principles, right?
I mean, you would not think of that as governance, but that is
a form of governance. It's a way in which you're
establishing guardrails for the company.
So you know, there are guard that that's what I mean by
governance is going to be changing and it's getting
smarter. Another example of how it's
changing is, for example, in, you know, we want to move
(19:37):
privacy and security almost at down to the to the edge level to
the phone, but you won't just, you know, your phone should have
more really security and privacy, not just in the cloud.
That's a different kind of governance.
And so how do you do that effectively when you have a
small footprint like a phone? So you see how it's, you know,
(19:57):
the, it's, it really is so important, but it's got to be
done in the right business perspective with the right
flexibility and scope and, you know, the right way of
communicating the importance so that everybody complies because,
you know, a lot of times, you know, people want to kind of go
around the rules. What, when, where you want the
(20:18):
organization to, to realize thatthese, these guardrails are
there for a reason and we want to make sure that everyone can
comply to that. And you, you know, it's all
possible, right? I've seen it done being done
effectively, but you've got to do it thoughtfully.
And again, this becomes an important part of your data
strategy to think through not just the governance rules and
(20:40):
standards, but how can we do that in a flexible, automated,
intelligent way that, you know, that then encourages compliance.
I think, I think to just add on to that, as far as the vision,
so to say for like overarching data management, would you agree
with once you have the data strategy mapped out, you see how
(21:04):
it ties you determine the data quality, how it is related to
those objectives, determining for meeting those and then
utilizing AI or their software practices to automate as much as
the governance compliant as possible, while at the same time
ensure our training or the training enables the people, so
to say to be thoughtful, intelligent and data literate.
(21:26):
And you would you say that's more or less a good broad vision
for the majority of companies seeking out the data strategy?
Absolutely. You got it right.
And I think the other one is, you know, is investing in the
resources that go with it. And you know, do you need a head
of data? Do you need a head of data and
analytics and AI? There's some interesting roles
now where it's all combined. You have one who's a head of AI,
(21:51):
data foundations and data analytics.
Some other organizations have split this role up into two or
three. I think in all cases, if you
really are serious about this topic and the reason and you
realize that you cannot achieve the business transformation that
(22:12):
you want to achieve without managing that data first, then
you, you have to commit intellectually and then resource
wise to having a person that's going to lead that work.
But that person alone can't do all the work, right?
They don't have a magic wand andgoing to make all this go away.
And, and, and it's not going to be easy.
(22:35):
They're going to have to enlist the organization.
So it's a commitment of, of people, of processes of some
dollars and technology, but do it.
And This is why you want to go back to this business outcome.
You do it because you need to doit, and you need to do it to
drive your outcomes. You don't just do it for data's
sake, for sure. And you don't do it on every
(22:56):
single piece of data that you have.
You do it on the data that matters most.
Excellent. Understood.
I do want to pivot for a moment too and kind of go over more or
less how you got into this passion for data and data
strategies. More or less going over the
hero's journey where you startedoff from kind of lessons learned
along the way that more or less shaped your passion for data
(23:19):
management and strategy and key lessons learned along the way
for upcoming up and coming CD OSor just companies interested in
creating a data strategy. So how did I get involved in
this? I actually kind of, I got
volunteered into it. I would say my first job
straight out of college was at IBM and I spent many years at
(23:42):
IBM on the technology side and then I into in IT and I had a
very, very good boss who we had been going through a major
business transformation inside the company in sales and the
business transformation was taking very, very long.
It was ACRM customer relationship management transfer
(24:03):
and the, and the, the project was delayed because the
underlying data was very bad. And it's so there was a lot of
data cleanse cleansing that wasn't done.
Our customer data was just inconsistent, outdated, bad
quality, all of those things andit was causing more and more
(24:23):
delay. So I had a very insightful boss
at the time who said, Maria, youknow, I've kind of was known for
being able to turn projects around.
They said, well, can you go hereand fix this data problem for
CRM? And I said, OK, never really had
done specific data programs, right?
Had kind of all part of a major transformation, said sure.
(24:45):
And so we did. We got the project back on
track. And then from there, the same
very smart executive said, you know, what you're doing here
makes a lot of sense for other kinds of data.
Why don't you just start, why don't you take product data and
then finance data? And it's little by little.
We created this first center of competency for data had 1000
(25:09):
people in. I mean, it was a very large
organization had it had technology, it had business, it
had architecture and had operations.
I mean, it was very, very big. Of course, everything in IBM is
was big back then. But so that's when we got
started and I went back to school.
That's when I got, went back to University of Miami and I, so I
had an undergrad and computer science, but I went back to
(25:31):
school and got my MBA and my MISat University of Miami graduate
program. And then I love the marriage of
the technology with the businessand driving business outcomes,
right? So I know how to can, I can read
an annual report and I can not only understand it from a
financial perspective and what that all means, but I can hear,
(25:53):
I can listen with data ears and I can hear when companies say
they want to do XY and Z, that XY and Z is very data intensive.
And that has come from this kindof combination.
Where my career is really headedis this combination of data with
business and why I'm such a big believer that that is the way to
(26:17):
success in these programs and, and why did the master class,
why I've done, you know, a lot of other kind of work on this
topic. To me, AI is just another tool
in the toolbox of data and data management and another reason to
manage data because AI is all based on data, right?
(26:39):
Whether you're creating the new AI data models or the large
language models or whatever it is, if your underpinning is not
right from a data mod, from a data perspective, you won't get
the results you need. So to me, AI is exciting because
it's elevated the importance of managing data, and it's also so
(27:00):
you know where the money is going, so you have to follow the
money. In corporations, everyone seems
to be really excited about this,this new technology and doing
the necessary investments, but they're, they're waiting for
results. I mean, a lot of companies, what
they're doing is they're all piloting AI and they're all in
kind of that mode. But they want to see outcomes.
(27:22):
And again, you're not going to go achieve outcomes until you've
got all the pieces there. Right.
No, that's, that's an incredibletalent though that you mentioned
where you could gleam like a company's financial statements
or earnings pull and kind of determine from that what are
their objectives and what data they need to go after to more or
less hit those objectives. That's credible.
(27:44):
Yeah, yeah. It comes with your practice,
right. But it's, that's what I'm
saying. You kind of have to listen with
data ears and it's all possible,right?
It's definitely possible for those, for those of us that have
been in this profession for a while to develop those data
ears. But but it needs a little
practice for sure. Awesome.
No, thank you for sharing all those insights.
(28:05):
I do want to leave off on, you know, we like to do this thing
with our audience. If you had to give some words of
advice or words of wisdom to ouraudience, regardless of topic,
yeah. Do you have any insights or
lessons learned you'd like to share?
Well, I look, I, I, I want to goback to the AI topic because I
do think this is going, this is a controversial topic.
(28:25):
I mean, there are some who say AI will be the death of, you
know, the death of all workers or, you know, AI will solve
every single problem. And I think at the end, it's,
it's another tool in the toolbox.
It is important to understand how to use AI because it's going
to be in every single business process and every single company
in some way, shape or form. So I, it's really important to
(28:48):
know how to use AI in the right way and, and it could be a tool
for you, whether it's ChatGPT orsome of the other great AI
tools, but also understand it's real limitation and, and there
are limitations, technology, false positives, all sorts of
things. I am, I am on team human, right?
So I'm, I don't believe AI is going to take over the world.
(29:12):
I think that there's quite a bitof difference between artificial
intelligence and human intelligence.
And what makes us human is all of those other skills around how
we learn around communication and social and, and, and just
watching other people behave andothers, those those things are
not going to be easily taken over.
(29:34):
And that's what makes it special.
So as much as you want to use all this technology, you also
have to follow your gut and yourgut will tell you.
And following your gut, your gutand your brain are connected and
you will use that tool as well. So yeah, it's a great time to be
in this topic, but also everyoneshould really understand the
pluses and the minuses. Really well set up context to I
(29:56):
also am on TV human sodas. Good.
Miss Bullard, thank you again for joining us on the show.
If our audience members wanted to reach out to you or what
should they do? Where should they go?
Yeah, LinkedIn. I am on LinkedIn, so just invite
me to connect. I'm happy to connect.
Well, to our listeners, thank you again for joining us on
(30:18):
another episode of Approach Infinity.
Hope you all have a good rest ofthe evening.
Take Care now. Thank you for listening to
another episode of Approaching Infinity.
Make sure to follow us to be notified of new episodes, and
feel free to reach out directly on LinkedIn at Justin Roopner.
Read on Instagram at Jr. Live 7 or Twitter at Justin.
(30:42):
Under score. 777 under score.
Welcome to Approaching Infinity,where we focus on the latest in
finance, technology and innovation.
We present you with interesting and unique stories, news and
interviews to help you understand your limits and how
to exceed them. Now here's your host Justin
Rochner in it. Welcome back to Approach
(00:24):
Infinity. The show refocus on all things
finance, innovation and technology.
Today I am pleased to introduce Miss Maria Villar.
She's the Co founder and managing partner at Business
Data Leadership, which is a firmcommitted to promoting effective
data management, governance and analytics.
She has over 25 years of experience as a Chief data
(00:46):
officer from serving as Sapp's first Chief data Officer to
pioneer IB, Ms. first enterpriseData Center of excellence.
Finally, she serves as an advisor to numerous VC funds,
industry boards, and startups. And to top it all off, she's a
University of Miami grad. Miss Velar, thank you for
joining us again. Anything I missed?
(01:07):
No, you got it. I would just call it SAP instead
of SAP. Yeah, that is that kind of the
official name of the company. Yeah, I've heard it called the
SAP. No, when we were originally
looking for guests, right. And noting that this is more or
less Season 2 of our show, we want to guess who more or less
(01:28):
aligned or more senior in data positions to more or let's go
over various topics and creatinga data strategy going over what
that means. What are some best practices and
kind of the vision of setting that forward.
And you were the essentially thetop pick for that.
The fact that you were AUM grad more or less sold as on it as
well. So we brought you on to more or
(01:48):
less share your insights on a variety of topics.
But first one is essentially just for our audiences.
Knowledge is what is a data strategy, How do you craft one?
And then essentially how do you align it to grow the business
but as well manage data? Yeah, thank you Justin and thank
you for inviting me and and for the reach out.
It did come through LinkedIn, which is always interesting how
(02:10):
how well that that network does work to find the right
connections. And yeah, I spent 25 years of
my, of my professional career inthis topic of data when it
wasn't really a thing, when it wasn't really even a topic that
people and executives really cared about.
It was all kind of buried under IT.
(02:30):
But now with AI, now with certainly with analytics and
business transformation, the topic of data is much more
important than it's ever been. And it's, it's, it's great to
see it for myself, having spent 25 years trying to explain why
this is important to now get to a point where I think most, most
(02:52):
companies understand how important it is.
They just don't know how to manage it effectively.
And hence we come back to your topic, to this topic of a data
strategy. So to me, a data strategy is a
perspective, it's a point of view, it's a document that gives
direction on what is the data that most matters to the
(03:13):
company, what are the kind of risk and opportunities around
that data and how do you manage it effectively.
And you made a really good point, which is the way that you
start a data strategy is you look at your, the business
outcomes for a company. So I've got this master class on
YouTube, which is, I did about five years ago and it's still
(03:38):
relevant today. And it's a free asset that's
available to everyone. You don't have to be a, you
graduate just anyone. You can go to Google and, and
just search for business outcomedata strategy in my name.
And you'll see these 12 videos that really outline how to
create a data strategy. But the most important part, and
this I've learned through my three times of being a chief
(04:00):
data officer, that it starts with the business and the
business objectives. And you could be in government,
in which case if the business objectives are more mission
based, but nonetheless, right. If you look at where does your
organization want to go? What does it say publicly about
its direction in terms of eitherits annual report or what it
(04:23):
tells its stakeholders or what it tells outside investors or
what it tells, you know, the, the country, right?
What are the main objectives that they are trying to achieve
within that organization? And then you look at that and
decipher from that, well, if that's the most important areas,
then what's the data that is most important to those
(04:46):
objectives? And, and it's a process.
It's a process of really understanding how do you go
about affecting change in that transformation.
And transformation is the keyword.
You said it, most companies lookat data when they want to
transform their business, right.They're doing a digital
transformation there. And that could be a sales
(05:06):
process transformation that they're going digital or a
marketing or a supply chain or, or you know or something else in
their company or M&A that's verydata centric.
And then they'll go say, wow, I'm going to, I can't do that
transformation if the data underneath it, the foundation
(05:28):
from which all of this has to goand be converted isn't grounded
in high quality good data. And so really aligning those
business outcomes, those missionoutcomes to the data that
matters most is the start of a data strategy.
And that's the most important start of the data strategy.
(05:48):
So you do a scoping exercise, right, to define the that data
that's most important. And then you start to really
define what about that data has to change.
Is it the quality level? Is it that you need more data of
a certain category because you need to run analytics or AI?
So you kind of analyze the dimensions of data that have to
(06:12):
change and then the processes around data that have to change
and then the technology. So all of those things encompass
the data strategy. And that's, I think we're all
coming to really understand thatas as business professionals.
But often times, and I would saythe first, the last probably 10
years or so, everyone thought a data strategy was just a bunch
(06:33):
of databases and it was all technology.
But more and more, I think we'reall having the realization that
you can't just throw technology,build some databases and build
some reports. And that's the job that that's
not the job, right? The job is getting the results
that you need in the in the business with that data and
(06:55):
getting people to understand howto use data effectively.
And all of that encompasses muchmore than technology.
And all of that is what should be in a data strategy.
Excellent. Well said.
And just to summarize too, so essentially the first step is a
discovery process. What are your objectives or what
is the company's objectives? What data do you have on hand,
(07:18):
the quality level? What do you hope accomplish so
to say? Exactly.
Exactly. Let's say there's a case where
they're they have objectives or clearly defined objectives, but
their data quality isn't up to par.
What would you suggest in that case?
Like just go acquire new data orhow would you more or less
advise? Or does it truly depend on the
(07:39):
company and what they're gettingafter?
Well, ultimately it does depend on the company and the maturity
level and what they're willing to invest.
But if you, if you know that youhave a data quality problem, I
think the most and, and you, thefirst step is making sure that
you scope the data because oftentimes companies say, yeah,
I have a data data quality problem.
They go, and then they'll go offand try to solve every data
(08:01):
quality problem they have, right?
Oh yeah, my customer data is bad.
Let me go fix all my customer data When certainly that's a
noble cause on who can argue that you don't want good
customer data? But This is why going back to
the outcomes that you want to accomplish matters, because
those outcomes will tell you what about that customer data
(08:22):
has to change. Because if you try to handle
every piece, every customer field that there possibly exists
for a customer, you could be trying to clean 200 fields when
you really only needed 10 or 15.Because the outcome that you
needed, whether it was a marketing outcome, campaign
outcome is so you needed the good emails and maybe you
(08:44):
needed, you know, the good titlefor a come for a particular
person. You see, it depends on the
outcome. And the outcomes is so important
in the prioritization scheme because there will never be
enough people and resources to clean all the data and to do
everything that needs to be done.
So that's the scoping going backto scoping, scoping, scoping.
(09:05):
So once you know that you have adata quality problem and you've
scoped it down and you, you know, getting new data, look,
you only want to get new data ifit's something a real immediate
need and you don't have the timeto clean that data up.
But the problem is, but so get it and to do what needs to be
done, whether that's running a campaign, whether that's doing,
(09:26):
you know, special analytics or so if it's a one time data
acquisition, go for it and get the results and the credibility
you need as an organization. But oftentimes the problem that
you're really solving for is that the processes for managing
your data in your company are not right.
(09:46):
And so you get all this data in at one time and in six months
your data quality for that data is bad because you know the data
has decayed, the data has changed.
If you don't also include that data management process
transformation to constantly be managing your data to the right
quality level, you're going to end up in the same place.
(10:09):
So that's why a data strategy includes a lot of also process
transformation, data process transformation, you know, new
roles and responsibilities for people to manage that data
effectively and then the IT thatgoes with it.
So yeah, you got to start by getting that base clean, as
clean as possible, but you can'tkeep it clean if you also don't
(10:31):
do the transformation and have the right, you know,
architecture and and and roles and responsibilities to keep it
clean. Understood.
And would you say oftentimes that's easier said than done?
Or is the customer able to more or less all right once they get
their discovery meeting works out of the way, they set their
objectives, they figure out the more or less the data they need.
(10:54):
Do you think usually customers are able to more or less turn it
around, get the data and essentially gain insights,
analytics from it? Or is it truly just a a
continuous process where Oh my. Gosh, yeah, it's a, it's a
continuous process. It's an organization, right?
I mean, I, I've built a career around getting an organization
ready for maintaining that data.You can do a lot of things one
(11:15):
time, Justin, for sure, right? And, and yes, if you need that
special analytics, right, especially if you were during
Covad, a lot of things happened really quickly, right, because
you had to get all that done quickly to get the results that
you needed. I think a lot of a lot of COVID
pot organizations that they, themore ready you were from a data
perspective, the faster you could respond to all these
(11:39):
changing requirements until yoursupply chain being at risk and
to your employees aware all youremployees and their health and
well-being and all that. So, you know, I have built a
career and I coach and advise C level executives like CFOs and
Coos and CEO's on the importanceof, you know, commit to this
(12:00):
because this is an asset that you want to manage like you
manage people like you manage finances, like you manage your,
you know, supply chain. It is an asset that requires
resources, it requires processes, it requires
technology. And you got to put it up there
in terms of levels of investmentand it is the hardest part.
(12:20):
Going back to your question, what is the hardest part of all
this? It is the people part because
the technology, hey, there's a lot of technology out there that
will that, that will get you what you need, but really
getting the results and that result could be the business
outcomes that you're driving for.
People have to change their behaviors around data.
(12:41):
They have to make a decision. I think many executives that are
in data, they'll issue out, they'll have these great
dashboards and they do all thesegreat reports and do all these
great AI things and they give itto the business teams and the
business teams don't know what to do with it or they don't
trust it enough to make make thedecision.
So the last mile of data transformation is taking the
(13:04):
data and doing something different in the business that
creates the outcome that you're looking for in the business,
right? Whether it's improving sales,
whether it's improving your supply chain or whatever.
Data people can give you all thereports and data in the world,
but the business has to then take that data and make a
decision. And they won't make a decision
(13:27):
on that if they they don't believe the data is credible,
the report is credible, the organization is credible, and or
you know, they won't want to just trust their gut, not
believe the data. That whole aspect of
transformation is the hardest part that any data executive
will have. Excellently well said.
I agree on that point. The technology, at least what
(13:49):
I've seen is there. It's getting the people to adopt
that continuous mindset of like,hey, are you making your debt?
Are you cataloging? What objects are we trying to
meet and how can this data be used to more or less show
metrics or helps achieve those objectives?
It starts with the people. I partly agree with that.
Point is, and that's. Also kind of why I'm excited
about AI, right? Because I think AI from a data
(14:13):
management perspective, right? AI, if we can start applying
more and more artificial intelligence and automation and
digitization to all of the sort of tasks that you have to do
around data. I think you know the data
quality is going to get better the credibility is going to get
better and business owners are going to feel more comfortable
(14:35):
with the data that that they get.
And so I'm, I'm excited about AIin specifically the use of it in
data management and then. On the note too, like utilizing
AI, so we discussed data strategy, right?
I do want to pivot on to more orless inform our audience.
What is data governance? How does it relate to the
strategy data cataloging? Can you come more or less
(14:56):
untangle this web? First, you know all the
different data, so to say categories.
What is data governance and thenhow would AI essentially help in
this? So data governance often times
is kind of a, you know, a bad word in the mind of
organizations. They think of it as well, It's
the data police, right? So look, I, I think data data
(15:17):
governance is 1 segment 1 aspect1 chapter of your data strategy.
And that chapter of that data strategy is around what are the
policies and the standards around that managing that data.
I prefer to use the word data management rather than data
(15:38):
governance because just think about it, right?
A lot of what we're trying to dois to change behaviors and have
organizations, especially, you know, employees and business
people and managers think of data as an important asset in
their company. Well, you don't go around your
company and say, well, I'm goingto govern my people or I'm going
(15:58):
to govern my, you know, projects.
You say you manage them. And so management is the word I
like to use around this topic. And, and so data governance is
to really break down how you're going to manage data and what
are the standards that you're going to use, like whether it's
a data quality standard, whetherit's a standard for how you
(16:20):
create data. So there are process standards,
whether it's, you know, a standard for who has access.
So all of those kind of parameters around who and how
and what to manage is governance.
It's also your data policies. And then of course, it'll be the
technology that helps you to do that.
But that to me is, is what governance is.
(16:45):
And again, it's not, you know, it's not everything that's in a
data strategy. So it wouldn't cover like, for
example, all the training that you have to do, right?
That's not governance. That's training and culture and
culture training and literacy trainings.
That's what I mean. There is a place and a role for
data governance, and it's reallyimportant, but it also has to be
(17:07):
done in a smart way, Justin. So, and what I mean by smart
governance is that it has to be flexible.
It has to be tied to the problemyou're trying to solve.
So yeah, you could govern every aspect of data and you make it
so onerous that nobody wants to use it and nobody can do it.
(17:28):
This goes back to if you know what the outcome is that you're
trying to achieve with that data, then make sure the
guardrails are appropriate to towhat you're trying to get done.
So be smart about the standards,be smart about the guard rails
to make sure that they're flexible, make sure that they
(17:48):
can be executed and done, especially with automation.
If you can automate governance, that makes everybody's job
easier and and then, you know, be willing to change.
I think we need to go from the old fashioned governance was
very rigid, a very rule based, avery almost, you know,
governmental in a way. But the new governance,
(18:10):
especially as we think about AI governance or governance around
even analytics, there's also going to be more flexibility
there. There will be some hardcore
governance that you can't get out of because you need that for
let's say financial data or other kinds of data and privacy,
right? So there be some hardcore stuff,
(18:31):
but there be other opportunitieswhere you want to manage
analytic data or AI data a little bit differently.
Let me give you one example of that.
One example is, is having an AI,having AI principles, not just
AI strong rules. So a lot of companies start by
saying, the first thing I want to do is I want to give every
(18:53):
employee our AI principles so that they can start as they
develop AI or use AI. They understand what's important
for us as a company. For example, you know, our, you
know, we will be, we will have safe diverse AI.
We will have a human in the middle in AI.
(19:14):
You see, those are all principles, right?
I mean, you would not think of that as governance, but that is
a form of governance. It's a way in which you're
establishing guardrails for the company.
So you know, there are guard that that's what I mean by
governance is going to be changing and it's getting
smarter. Another example of how it's
changing is, for example, in, you know, we want to move
(19:37):
privacy and security almost at down to the to the edge level to
the phone, but you won't just, you know, your phone should have
more really security and privacy, not just in the cloud.
That's a different kind of governance.
And so how do you do that effectively when you have a
small footprint like a phone? So you see how it's, you know,
(19:57):
the, it's, it really is so important, but it's got to be
done in the right business perspective with the right
flexibility and scope and, you know, the right way of
communicating the importance so that everybody complies because,
you know, a lot of times, you know, people want to kind of go
around the rules. What, when, where you want the
(20:18):
organization to, to realize thatthese, these guardrails are
there for a reason and we want to make sure that everyone can
comply to that. And you, you know, it's all
possible, right? I've seen it done being done
effectively, but you've got to do it thoughtfully.
And again, this becomes an important part of your data
strategy to think through not just the governance rules and
(20:40):
standards, but how can we do that in a flexible, automated,
intelligent way that, you know, that then encourages compliance.
I think, I think to just add on to that, as far as the vision,
so to say for like overarching data management, would you agree
with once you have the data strategy mapped out, you see how
(21:04):
it ties you determine the data quality, how it is related to
those objectives, determining for meeting those and then
utilizing AI or their software practices to automate as much as
the governance compliant as possible, while at the same time
ensure our training or the training enables the people, so
to say to be thoughtful, intelligent and data literate.
(21:26):
And you would you say that's more or less a good broad vision
for the majority of companies seeking out the data strategy?
Absolutely. You got it right.
And I think the other one is, you know, is investing in the
resources that go with it. And you know, do you need a head
of data? Do you need a head of data and
analytics and AI? There's some interesting roles
now where it's all combined. You have one who's a head of AI,
(21:51):
data foundations and data analytics.
Some other organizations have split this role up into two or
three. I think in all cases, if you
really are serious about this topic and the reason and you
realize that you cannot achieve the business transformation that
(22:12):
you want to achieve without managing that data first, then
you, you have to commit intellectually and then resource
wise to having a person that's going to lead that work.
But that person alone can't do all the work, right?
They don't have a magic wand andgoing to make all this go away.
And, and, and it's not going to be easy.
(22:35):
They're going to have to enlist the organization.
So it's a commitment of, of people, of processes of some
dollars and technology, but do it.
And This is why you want to go back to this business outcome.
You do it because you need to doit, and you need to do it to
drive your outcomes. You don't just do it for data's
sake, for sure. And you don't do it on every
(22:56):
single piece of data that you have.
You do it on the data that matters most.
Excellent. Understood.
I do want to pivot for a moment too and kind of go over more or
less how you got into this passion for data and data
strategies. More or less going over the
hero's journey where you startedoff from kind of lessons learned
along the way that more or less shaped your passion for data
(23:19):
management and strategy and key lessons learned along the way
for upcoming up and coming CD OSor just companies interested in
creating a data strategy. So how did I get involved in
this? I actually kind of, I got
volunteered into it. I would say my first job
straight out of college was at IBM and I spent many years at
(23:42):
IBM on the technology side and then I into in IT and I had a
very, very good boss who we had been going through a major
business transformation inside the company in sales and the
business transformation was taking very, very long.
It was ACRM customer relationship management transfer
(24:03):
and the, and the, the project was delayed because the
underlying data was very bad. And it's so there was a lot of
data cleanse cleansing that wasn't done.
Our customer data was just inconsistent, outdated, bad
quality, all of those things andit was causing more and more
(24:23):
delay. So I had a very insightful boss
at the time who said, Maria, youknow, I've kind of was known for
being able to turn projects around.
They said, well, can you go hereand fix this data problem for
CRM? And I said, OK, never really had
done specific data programs, right?
Had kind of all part of a major transformation, said sure.
(24:45):
And so we did. We got the project back on
track. And then from there, the same
very smart executive said, you know, what you're doing here
makes a lot of sense for other kinds of data.
Why don't you just start, why don't you take product data and
then finance data? And it's little by little.
We created this first center of competency for data had 1000
(25:09):
people in. I mean, it was a very large
organization had it had technology, it had business, it
had architecture and had operations.
I mean, it was very, very big. Of course, everything in IBM is
was big back then. But so that's when we got
started and I went back to school.
That's when I got, went back to University of Miami and I, so I
had an undergrad and computer science, but I went back to
(25:31):
school and got my MBA and my MISat University of Miami graduate
program. And then I love the marriage of
the technology with the businessand driving business outcomes,
right? So I know how to can, I can read
an annual report and I can not only understand it from a
financial perspective and what that all means, but I can hear,
(25:53):
I can listen with data ears and I can hear when companies say
they want to do XY and Z, that XY and Z is very data intensive.
And that has come from this kindof combination.
Where my career is really headedis this combination of data with
business and why I'm such a big believer that that is the way to
(26:17):
success in these programs and, and why did the master class,
why I've done, you know, a lot of other kind of work on this
topic. To me, AI is just another tool
in the toolbox of data and data management and another reason to
manage data because AI is all based on data, right?
(26:39):
Whether you're creating the new AI data models or the large
language models or whatever it is, if your underpinning is not
right from a data mod, from a data perspective, you won't get
the results you need. So to me, AI is exciting because
it's elevated the importance of managing data, and it's also so
(27:00):
you know where the money is going, so you have to follow the
money. In corporations, everyone seems
to be really excited about this,this new technology and doing
the necessary investments, but they're, they're waiting for
results. I mean, a lot of companies, what
they're doing is they're all piloting AI and they're all in
kind of that mode. But they want to see outcomes.
(27:22):
And again, you're not going to go achieve outcomes until you've
got all the pieces there. Right.
No, that's, that's an incredibletalent though that you mentioned
where you could gleam like a company's financial statements
or earnings pull and kind of determine from that what are
their objectives and what data they need to go after to more or
less hit those objectives. That's credible.
(27:44):
Yeah, yeah. It comes with your practice,
right. But it's, that's what I'm
saying. You kind of have to listen with
data ears and it's all possible,right?
It's definitely possible for those, for those of us that have
been in this profession for a while to develop those data
ears. But but it needs a little
practice for sure. Awesome.
No, thank you for sharing all those insights.
(28:05):
I do want to leave off on, you know, we like to do this thing
with our audience. If you had to give some words of
advice or words of wisdom to ouraudience, regardless of topic,
yeah. Do you have any insights or
lessons learned you'd like to share?
Well, I look, I, I, I want to goback to the AI topic because I
do think this is going, this is a controversial topic.
(28:25):
I mean, there are some who say AI will be the death of, you
know, the death of all workers or, you know, AI will solve
every single problem. And I think at the end, it's,
it's another tool in the toolbox.
It is important to understand how to use AI because it's going
to be in every single business process and every single company
in some way, shape or form. So I, it's really important to
(28:48):
know how to use AI in the right way and, and it could be a tool
for you, whether it's ChatGPT orsome of the other great AI
tools, but also understand it's real limitation and, and there
are limitations, technology, false positives, all sorts of
things. I am, I am on team human, right?
So I'm, I don't believe AI is going to take over the world.
(29:12):
I think that there's quite a bitof difference between artificial
intelligence and human intelligence.
And what makes us human is all of those other skills around how
we learn around communication and social and, and, and just
watching other people behave andothers, those those things are
not going to be easily taken over.
(29:34):
And that's what makes it special.
So as much as you want to use all this technology, you also
have to follow your gut and yourgut will tell you.
And following your gut, your gutand your brain are connected and
you will use that tool as well. So yeah, it's a great time to be
in this topic, but also everyoneshould really understand the
pluses and the minuses. Really well set up context to I
(29:56):
also am on TV human sodas. Good.
Miss Bullard, thank you again for joining us on the show.
If our audience members wanted to reach out to you or what
should they do? Where should they go?
Yeah, LinkedIn. I am on LinkedIn, so just invite
me to connect. I'm happy to connect.
Well, to our listeners, thank you again for joining us on
(30:18):
another episode of Approach Infinity.
Hope you all have a good rest ofthe evening.
Take Care now. Thank you for listening to
another episode of Approaching Infinity.
Make sure to follow us to be notified of new episodes, and
feel free to reach out directly on LinkedIn at Justin Roopner.
Read on Instagram at Jr. Live 7 or Twitter at Justin.
(30:42):
Under score. 777 under score.
Popular Podcasts
Hey Jonas!
Hey Jonas! The official Jonas Brothers podcast. Hosted by Kevin, Joe, and Nick Jonas. It’s the Jonas Brothers you know... musicians, actors, and well, yes, brothers. Now, they’re sharing another side of themselves in the playful, intimate, and irreverent way only they can. Spend time with the Jonas Brothers here and stay a little bit longer for deep conversations like never before.
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com