Driving a robust data analytics engine in an enterprise is not much different than a team sport. Business, data, and IT constituents need to develop mutual trust, bring forth ideas and challenge norms, take ownership as appropriate, and work cohesively in one direction. How to secure sponsorship? Who all should be recruited in the team? Who should be the coach? And, how should the playbook be practiced to consistently win the games?

Contributor

• • Deepa Soni, US Chief Information Officer (CIO) and US P&C Technology Chief Information Officer (CIO), BMO Financial Group

Download Podcast
Apple Podcast, Google Podcast, Spotify, Pandora, iHeartRadio, SoundCloud, TuneIn, and Stitcher. Find other syndication channels here or search CIO Talk Network podcast on any other app.

Explore More

• • Is Data Analytics helping increase the impact of K12 Education?
  • Creating a Winning Data and Analytics Strategy
  • Building High-Performing Data and Analytics Team
  • Steps to Future-proofing Data and Analytics Environments
  • Building Data and Analytics Driven Culture
  • Have Enterprise Data and Analytics Programs Delivered on the Promise?
  • Trusting Big Data through Extractive Analytics
  • How to be an Effective Analytics Leader?
  • How to Enable Real Time Analytics
  • Analytics in Retail – The New Breed of Loyalty Programs
  • How to Make Predictive Analytics Effective

Transcript

Sanjog Aul [00:00:22]:

Hello, and welcome to this segment on CTN. To learn more, please visit ciotalknetwork.com. And the topic for today is data analytics as a team sport. And we have Deepa Soni, who’s the CIO at BMO Financial Group. Hey, Deepa. How are you?

Deepa Soni [00:00:37]:

I’m doing alright.

Sanjog Aul [00:00:38]:

Very good. Very good. Hope life and business is going well with the interesting roles that you guys are all playing in the digital world. How’s that going?

Deepa Soni [00:00:47]:

Absolutely. There’s never a dull moment.

Sanjog Aul [00:00:50]:

Yes. There you go. So talking about never having a dull moment, let’s talk about data analytics. And frankly, this, we were just talking before we got started, is in quite a few conversations I’ve had with leaders globally, when it comes to data analytics, people are not exactly struggling, but they are not able to take the maturity level of the data analytics, function above 2 or 3 out of 10. And they say it’s not that we don’t want to, but it’s just the nature of the beast. And when it comes to taking it to the next level, of course, it’ll be requiring that you have the business, the data group, and the IT group to all come together and work on it. And that’s why we brought up this concept of team sport, and we have to see who’s the coach, how do you assemble the team, what’s the sentiment that should prevail for data analytics to truly add value and be successful for an organization. So that said, Deepa, when you were to look at, of course, at BMO and any other experiences you’ve had, what do you see as the current state of data analytics function in enterprises? I quote unquote say enterprises because that’s where people struggle.

Deepa Soni [00:02:04]:

Yep. I would agree. I think at that enterprise level, the perception may be, you know, closer to maybe 4 out of 10 or whatever, you know, depending on the enterprise. But I do think that there have been some great pockets of value generation that have been realized by the data analytics functions and the whole practice. There are definitely some good business problems that have been solved. But I think as we look towards where we are today, I think in the last 5 to 10 years, we’ve spent a lot of time on the data ingestion, and bringing it together, gathering the data, creating data lakes, creating enterprise data warehouses, right? And now the focus is really shifting on data consumption, which is where I think the pace question becomes even more important. I also think the technology capabilities are growing by leaps and bounds. And what we have today in our hands gives us even a greater opportunity to really leverage data. And then the last, I think, factor in my mind that’s driving the need, or the PACE question is really around digital transformation. So pretty much, you know, a lot of the organizations that we talk to today are in some form of digital transformation. And that puts a lot more pressure on the pace issue as well. And I would say CIOs, CDOs, business leaders, we’re always trying to come up with more ways to deliver value faster. But I think the perception at an enterprise level is probably still not catching up to the business demand, triggered by some more factors that are contributing to that pace, gap, if you would.

Sanjog Aul [00:03:43]:

So given that and thanks so much for giving your perspective on where things are. Now as per many discussions I’ve had, people say that we created a data lake, for example. But then people took it loosely and kept throwing data in it, so it became a swamp. And then there was not much done or there was no clarity on which portion of data we should keep or I mean, the data lake concept started with the whole idea that, okay, you can put things in there and then later on use it. But then after some time, it becomes a very fuzzy definition of what we should keep, what we should work on, what we should leverage. And that’s why the data engineering function is the one where the struggle starts. So have we done anything with that? Because that’s the very source you’re banking on.

Deepa Soni [00:04:37]:

That is correct. So, like I said, I think we’ve put on a lot of focus on data aggregation, getting the data to a centralized place. But I think what we lost in my mind in that context is the data comes with a business context. So as we start to leverage that data, I think there has to be more specific, more business oriented orientation to that usage. Because that’s how we will start to unwind the swamp into a lake, right? So I think, for me, the most important, data engineering is part of it. But even then, you can’t tackle it at a broad level, at an enterprise level. You need to start putting in some domains or business context into how do you start solving the problem and start making that data more consumable for the enterprise.

Sanjog Aul [00:05:31]:

When, like, you know, almost a decade or so ago, people used to talk about master data management. And that was to some extent also, okay, one version of the truth. We gotta get all constituents together. And that’s where we struggle because the ones who are the originators or the creators of data were not willing to take responsibility and or be ready to be a custodian of the data. Do you think that has changed?

Deepa Soni [00:05:56]:

I think absolutely, I think that has evolved a lot more. I think data governance functions, as we used to talk 10 years ago, have come way further in my mind. I think there’s a lot more maturity, a lot more ownership than I see at least in banking world that we used to see before. You know, master data management, again, is one part of the data that we create. And so the more we start using the data, I think the greater focus starts to bring on the quality of the data and upstream value chain and, you know, authoritative sources of data to really drive that analytics that’s going to drive the business. Right? So I think there’s a value chain. I’ve definitely seen a lot more maturity, a lot more work to be done. But I think, if you start tackling it in a piecemeal in a business context, I think those things are being tackled, I would say, today. I wouldn’t say at an enterprise level, but definitely in the context of a business strategy or business context. I see that happening way more than what we saw last 4, 5, 10 years.

Sanjog Aul [00:06:57]:

So divide and conquer is a good strategy, of course. And as you mentioned, you assign a business domain or a context, and that will make sense for that particular data element to be handled properly. Now at the same time, people also say, hey. Why are we tackling this data issue in a silo mode? Because what you’re doing here could also have value in some other part of the business. But if we try to make those silos to solve that point problem, we are moving away from holistic, which is the holy grail, of course. And we are creating another, challenge which you’ll have to tackle later. Should we go this route? Is this the only route for us to go? Or could we have not exactly boil the ocean, but could have started with a holistic manner?

Deepa Soni [00:07:43]:

So let me clarify. When I talk about divide and conquer, I think it’s important that we don’t go to the lowest denominator. You do have to have a domain, as I call it, or a business full business. Cross channel, cross products, cross customers. The domain has to be big enough so that it doesn’t become too siloed. And in banking world, for example, we look at a customer risk, financial, all those things holistically. So you don’t make decisions or you don’t set up the infrastructure based on that. But I think enterprises even is a sum of multiple businesses and most of the companies, right? And so I would say the domain or how we divide and conquer has to be big enough, has to be large enough tied to a business strategy. So it can’t be at functional level, it has to be at a higher level. So it’s a balance between going too narrow or too broad.

Sanjog Aul [00:08:37]:

And so that is a gut based decision?

Deepa Soni [00:08:43]:

In my mind, we do it based on lines of business. So if you have a certain segment of business that’s a big P&L, that has large customers, products, distribution channels, that would be a holistic way to look at that whole business and look at analytics that could drive every aspect of the business through a centralized, through a consolidated set of analytics.

Sanjog Aul [00:09:07]:

So essentially, some sort of an ROI or what makes the most money is the one which gets the most influence or gets that their specific vantage point. Is that the approach being taken today?

Deepa Soni [00:09:21]:

Yes. I think that I’ve seen with stuff the analytics that drives the business, meaning, as a business, I’m worried about acquiring customers. I’m worrying about not losing my customers. I’m worried about introducing new products. I’m worried about making sure my cross channel customer experience is consistent. Those are the kind of things that you look at holistically and start to drive more value in each of those dimensions and not to mention the risk and the financial aspects that come as part of being running a business.

Sanjog Aul [00:09:52]:

So imagine you took a business case or you built you created a business case where you took one of the biggest P&L areas and allowed that to drive your decisions on what the rest of the, you know, context would also, the pattern that they will follow. Now that said, the business changes at a speed which is happening at a much faster rate than any one of us would like. Right? But it’s a reality. And then what you have today as your growth catalyst or as an area which is most promising, that equation could change. But then you started. You made decisions. You started this data analytics engine in motion keeping that particular business segment in mind. I’m just playing devil’s advocate, but things could switch. But when they do switch, do you have a fallback, or you’re saying, you know what? The ship has sailed. We gotta live with it.

Deepa Soni [00:10:46]:

So I think I would say two things. When you talk about data lake, setting up the infrastructure, the platforms, you do have to take an enterprise view. Right? It’s not a P&L view. It’s an enterprise view. So you do set yourselves up for an enterprise view with the consumption being driven by a P&L unit. Right? So that’s a different so that you’re setting yourselves up for always success and consumptions on different paradigms. Things could change, but your infrastructure, your platforms are architected and set up in a way that it does help with the changing strategy or changing P&L or changing business units.

Sanjog Aul [00:11:28]:

Yeah. So that can answer, Sid, is because it’s not the foundation being built in a siloed mode. It’s more the consumption. Okay. Fair enough.

Deepa Soni [00:11:36]:

So your foundation always has to be broader than any one particular area, particular P&L, particular domain. Right? That’s why that’s where you get the lake and all that centralized data being utilized by same data being utilized by multiple constituents consistently.

Sanjog Aul [00:11:54]:

Let’s take a quick break. Listeners will be right back. And let’s talk about the collaboration that happens or should happen between business data and IT teams because there is a lot of reporting around that where these constituents, either they don’t exist as clearly in terms of the data team separate from IT team or business and or the collaboration that’s happening. Everyone is pulling out their agenda and trying to push their agenda versus looking at it as a team. What’s the current state? What are we doing to tackle these gaps? Let’s, discuss that more when we come back. Please stay tuned.

Speaker 0 [00:12:37]:

Predict your company’s future by creating it. Is your workforce able to connect, exchange ideas, and share brilliance simply and securely? Create tomorrow, today. Empower your people to innovate anytime and anywhere with secure BlackBerry Enterprise Mobility Management and document sharing solutions. To learn more, visit blackberry.com/enterprise.

Speaker 0 [00:13:04]:

Patient centered care requires a connected enterprise. Are you ready? If you’re looking to scale your health care IT efforts, visit redmane.com/health today. Whether it’s to connect data from multiple partner solutions or developing software for unique needs, Redmane can help. To find out how Redmane can help your company deliver on the patient centered care promise, visit redmane.com/help or call (773) 693-3919. Visit today.

Speaker 0 [00:13:38]:

Your growing business needs a highly productive workforce, effectively communicating and collaborating without exposing corporate data to cyber attacks. Are you looking to balance security and workforce productivity? Move beyond short term measures and securely scale your business with BlackBerry Enterprise Mobility Management Solutions. To learn more, please visit blackberry.com/enterprise. You are listening to CTN, CIO Talk Network with Sanjog Aul. To learn more about our program, please visit ciotalknetwork.com. Now back to the show.

Sanjog Aul [00:14:26]:

Welcome back. So when we talk about the data analytics, the engine, the function that we want to build, we do talk about the business constituent, the IT, segment, and then a data organization. Many companies don’t even have that separated independent data organization. But even if they do, they always have these three pillars, if you will, or three team members or constituents are not really seeing eye to eye at all times. So who becomes the Pied Piper? Who basically reins them in and say, guys, if we have to make this happen, the data analytics aspect, we’ve got to get pull our socks and get friendly with each other and move forward. What’s holding us back, and how has that been tackled? Deepa.

Deepa Soni [00:15:18]:

Sure. So I think, I would say, you know, data journey is not a trivial journey. And, we’re used to technology projects. We’re used to data projects. But I think now we’re at a point where it’s not it’s never a pure technology project. It’s never a pure data project. And so the cross functional stakeholdership is much higher as we look at the consumption of the data. So I would say, again, going back to, I think, on the setting up the infrastructure, setting up something that’s leverageable by the enterprise is definitely, I call it the enterprise function between data and technology teams, right? So it serves all the businesses. When you come down to the consumption side of the data analytics, I think the paradigm is shifting towards that, that cross functional team with clear operating model. What’s holding us back today, in my opinion, the biggest factor in all through my career, this has been many times true, is what I call it the organizational operating model, which means who’s leading the ship, who’s following, are the stakeholders aligned on a collective vision of what we’re trying to accomplish and how we’re trying to accomplish? What outcomes are we looking for? What scope? What expertise? And who’s going to play that role? So those questions, if aren’t clarified in an operating model, that’s where that slows down the pace of execution because everybody’s trying to figure out what other is doing, either it’s a coordination problem or it’s an overlap problem. So from my experience, and I’ve seen this multiple times, once you put together cross collaboration cross functional teams, and set the operating models in terms of the role of clarity, who’s so couple of examples I would give from my career is we were trying to send very qualified leads, analyzing the leads and sending it to the branches for distro for converting them into sales. The data team had a different view. The marketing team had a different view. The line of business who was receiving the leads has a different view. And so we had to set up a model where the actual business owner became the lead in terms of defining what’s going to actually sell in the distribution channels. And that led to technology, data and marketing teams coming together on figuring out what domain knowledge did they bring to the table to really improve the quality of the leads and converting them into sales. So that paradigm, as soon as we clarified who’s defining the what and who’s defining the how and how each of the pillars is going to play a role, the pace of acceleration that we saw on that initiative was tremendous. I’ve seen it multiple times happen. So I think the we’re not going to get away from business data technology and actually multiple stakeholders within business being part of this, I think clarifying that operating model, to me, has been the biggest impediment. And there’s other factors like lack of vision, too broad of a charter, too narrow of a charter. But all those things start flush out once you get that operating model, clarified and agreed upon.

Sanjog Aul [00:18:41]:

So I would 100% agree that if you put an operating model and you’ve got a process in place, that at least gives a blueprint on how this should run. Right? That that’s a good start. But then at the same time, while we have the clarity, but you are also having people with different intents and even, to throw a wrench in the system, different agendas.

Deepa Soni [00:19:06]:

Yep. Yep.

Sanjog Aul [00:19:07]:

And that’s what holds people back. So where do we lose out in the process of building an operating model? The effort to align the intent and the agenda, not every time the agendas are very explicit.

Deepa Soni [00:19:22]:

Right. So I would say it’s not just one time activity in my mind. You set up the model, but then every piece of execution needs to be shepherded towards that model. So you’re right. You’ll get distracted. That team could get distracted by, oh, this is important or that’s important. But I think that that’s the operational mindset that leaders have to pull in, is, you know, we all agreed we’re going to move down. And then there’s a business leader that is empowered to make changes. If things become important or distracted, I think there is a framework around decision making. But it’s all in the context of that operating model, meaning decisions don’t get made outside of that operating model. And so I go back to we’re also starting to see the shift towards more product centric teams, which are more empowered, higher caliber teams, at least in data world for us, that are clearly aligned to the business outcomes, business strategies, where when these distractions come and these other factors that may jump in, those teams are empowered and they have the adequate leadership to align things that are coming from outside into their model. They can prioritize things. They can de prioritize things. So I think we’re going to so it’s not just a one time setting up with the framework. It’s also making sure that you operationalize it. You get buy in from across the core stakeholders and toward and us moving towards that model. And it’s an evolution. It does take some time to get to that operating model. But once you are executing in that operating model, I think the maturity and the pace just increases tremendously.

Sanjog Aul [00:21:01]:

So in your example of making it product centric, which is a creative idea so that the P at least the business side starts seeing it as, oh, it’s what’s in it for me. Yes. There’s a lot in it for me, so I should support. But then the data organization, which is trying to shepherd all different product teams, for example, towards a common version of the data or a vision of the data. Then IT is saying, I want you to work with a standard set of plumbing, if you will, so it flows properly. So when you do that, just by moving the strategy I’m just trying to deconstruct this. If you move this product centric strategy, do you think the data and IT teams’ respective agendas are met automatically?

Deepa Soni [00:21:46]:

They are part of the team. So in the model where we have product centric teams, we have technology and data folks lined up to that product owner. And so I think the data teams sort of solve two problems. Right? They are also the horizontal custodian to making sure that, you know, none of the verticals are doing things that are counterproductive. But they also understand what the vertical teams are trying to accomplish. So they bring that horizontal vertical discussion to that model to that product owner. And in some cases, the enterprise wins, and in some cases, the vertical wins. But that discussion happens as part of that operating model. So in this operating model that I’m talking about, data teams and the technology teams are dedicated, along with the product owner, to driving a certain set of data analytics, for a particular business domain.

Sanjog Aul [00:22:40]:

And they negotiate with the product owners across different products to make sure that the universal form of data and universal form of technology foundation, that’s not jeopardized.

Deepa Soni [00:22:53]:

That’s correct. They do.

Sanjog Aul [00:22:55]:

Oh, okay. So just yeah. Go ahead.

Deepa Soni [00:22:58]:

The horizontal view and the enterprise, that custodian exists outside of these data analytics teams that is keeping that master data management and all of the enterprise level assets at an enterprise level asset. But when you think about the consumption, if the Enterprise Foundation is set up well, you don’t run into a lot of those issues. It’s 80/20. 80% of the work can be done with the data that has been already set up well. And then there might be 20% or stuff that needs to go back and be retrofitted or reconciled with the enterprise. But most of the consumption side, if the foundation is well built, and in my scenarios, they have been, it becomes less of a big friction point. And there are processes to take care of that.

Sanjog Aul [00:23:48]:

In the eyes of the folks, the executive sponsors or the management who’s supposed to fund these, are these data analytics initiatives seen as building a better mousetrap, or it’s a cost of doing business which will position them forward?

Deepa Soni [00:24:07]:

In my experience, what I’ve seen is all the business leaders or most of the business leaders that are transforming their business think of data and analytics as a core pillar to that transformation. So they don’t think of it as something as an afterthought. In fact, last year here in BMO, we actually made data and analytics a big part of the technology and data funding as part of the business strategy enablement. And we are using data as a core pillar. So it’s not a separate set of initiatives. It’s really tied to how we’re driving business outcomes. And I see business leaders providing tremendous support and executive sponsorship. In fact, they’re using a lot of the output from the data analytics to shape the business.

Sanjog Aul [00:24:57]:

So in your case, would you say you’re lucky comparing to many other industries where they say, I want to build that Bentley, but I’m getting the funding for a Chevy?

Deepa Soni [00:25:10]:

I would say I think we’re building between Chevy and Bentley, right size for the business. But I I’ve seen, in my banking experience in the last 18 years, I’ve seen a lot of business leaders driving, leveraging data and analytics in a very positive way.

Sanjog Aul [00:25:26]:

So you feel that the executive sponsors are not gonna be a stumbling block if an organization chooses to go this direction?

Deepa Soni [00:25:35]:

Not at all. I feel they are the ones who are championing for this. And that’s what we’re you know, if you think about going back to our first question of, there have been pockets of success. I think the pockets one of the pockets of success factor is always a business executive who’s leveraging the data, right? And so I’m not sure if we’re going to ever see 110 out of 10. But, I think now that a lot more business leaders are transforming their businesses, I think the breadth of data analytics is expanding in terms of what it’s driving the value.

Sanjog Aul [00:26:07]:

Let’s take a quick break, listeners, and let’s talk about the way the data analytics to be handled at an organizational design level. So we got people from business and from data and IT coming together to get that outcome. Then you got some people who work in that analytics team, and then now we are introducing the chief analytics officer. What is the what are the different flavors of organizational design to enable this data analytics function we should be looking at or exploring given your situation and which one would be a recommended route given the context? What would all look like all that look like? Please stay tuned, listeners. We’ll be right back and explore.

Speaker 0 [00:26:59]:

Your growing business needs a highly productive workforce, effectively communicating and collaborating without exposing corporate data to cyber attacks. Are you looking to balance security and workforce productivity? Move beyond short term measures and securely scale your business with BlackBerry Enterprise Mobility Management Solutions. To learn more, please visit blackberry.com/enterprise.

Speaker 0 [00:27:30]:

Patient centered care requires a connected enterprise. Are you ready? If you’re looking to scale your health care IT efforts, visit redmane.com/health today. Whether it’s to connect data from multiple partner solutions or developing software for unique needs, Redmane can help. To find out how Redmane can help your company deliver on the patient centered care promise, visit redmane.com/help or call (773) 693-3919. Visit today.

Speaker 0 [00:28:02]:

Predict your company’s future by creating it. Is your workforce able to connect, exchange ideas, and share brilliance simply and securely? Exchange ideas, and share brilliance simply and securely? Create tomorrow, today. Empower your people to innovate anytime and anywhere with secure BlackBerry Enterprise Mobility Management and document sharing solutions. To learn more, visit blackberry.com/enterprise. You are listening to CTN CIO Talk Network with Sanjog Aul. Now back to the show.

Sanjog Aul [00:28:43]:

Welcome back. So as in any team, you want the sponsor, so you have it. The good news as, Deepa, you mentioned that most executive management is very willing to sponsor these initiatives. So that said, the next step is who should be brought onto the team, which talks about the organizational design. You got business, you got data, and you have an IT group, and you mentioned that BMO went for a product centricity in their organizational design. How was that, how did you reach that decision? What were the parameters that you looked at? What did you uncover as a need for you to say, okay. Let’s go with this organizational design?

Deepa Soni [00:29:25]:

So I think, I would go back maybe a couple of years. You know, we’ve been doing data work, more in a project centric way, if you would. And each of the projects has been delivering value. But when you look at the higher level domain that we were talking about, at some point, you really need to start thinking about, I’m doing all these siloed projects. They’re adding value. But is there a way to step up the function, and have a top down approach, as I would call it, rather than a bottoms up to the data and analytics and the value it can drive? And so we had a new business strategy. We were at a step up function point, and it made sense that we would start to take a top down approach. And the top down approach really comes from that product mindset, where it’s persistent teams having persistent knowledge of the data, persistent business knowledge that sits in those teams that can better can come with more innovative ways to drive the outcomes. So we moved, I would say, from a project to product sort of a paradigm or we’re still moving, I would say, based on how the business evolved, how our strategy evolved. And also, as we started to see a little bit downside of this project based approach, it wasn’t one plus one wasn’t adding up to two. And so that’s how we moved. You know, I would say, depends on the organization where they are in their life cycle. Initially, setting up some stuff could be a project based, could be a very, very narrow scope tactical. But as you start to leverage harness the power more, or you have more desires, to be aggressive in leveraging data, you know, that model could move. So I think there are a lot of variables. Where is the domain knowledge, and we’ve had to shift some of that. Where are we going to build that domain knowledge, for the years to come so that we can get even more value as machine learning and other things are coming into this space? So I think there were some explicit discussions and decisions that triggered this move towards the model.

Sanjog Aul [00:31:32]:

And it wasn’t a rolling of dice. It was more deliberate set of discussions than decisions. And subsequently, you thought this is the best route to go.

Deepa Soni [00:31:40]:

Absolutely. And also the PACE question, right, that we started off with. How do you accelerate the delivery? And it comes from persistent knowledge, persistent knowledge of the data, persistent knowledge of technology, persistent knowledge of the outcomes that the business is trying to achieve. So I think all those things, you know, come together in a product model more efficiently and more elegantly than in a project model. Yes.

Sanjog Aul [00:32:06]:

And how is it going? Is it proving your decision to be right every day as you work?

Deepa Soni [00:32:14]:

I would say yes. There’s definitely less friction in the system. So we set up the right operating model. We’re moving towards product ownership where we have stronger, more empowered, a more high caliber team that’s driving this. I would call this again, we’re talking about the consumption side. So, yes, we’re starting to reap the benefits. We’re starting to reap the benefits, feeling better. And again, I think we’ve done it in other domains, not in data. But it’s just a matter of time before this operating model will start to result in some higher, bigger, step up, you know, value that we could drive from data.

Sanjog Aul [00:32:55]:

So you being the CIO, I’m assuming you tackle the IT side of the three pillars, right, the business, data, and IT. Do you have your counterpart who is leading the data group?

Deepa Soni [00:33:05]:

That is correct. So we have a CDO, Chief Data Officer, and we work very collaboratively with the business, together to set up these models and paradigms. And, you know, we all were all on this together, pretty much, coming to the same conclusions that we needed to move towards more the product ownership model, to help the business get more value.

Sanjog Aul [00:33:27]:

And did you feel the need of a Chief Analytics Officer to be added as the fourth pillar?

Deepa Soni [00:33:32]:

So analytics is covered under our data ownership model. But the actual execution of the analytics actually happens in the business. So in terms of the tools, the infrastructure, the enterprise COE, center of excellence, we have that all under the data.

Sanjog Aul [00:33:52]:

And when you do that, do you think there would be a risk of business seeing in the direction it wants to see versus allowing analytics the independence that it shows patterns? You see where I’m going?

Deepa Soni [00:34:08]:

Yeah. So in the way we’re set up, the business analytics lead is independent of the business functions, if you would. They’re part of the business, but they don’t own any functions, so they can’t skew the results, if that’s what you’re talking about.

Sanjog Aul [00:34:26]:

Exactly. So sometimes, you have a vision. You want to go. So you got, like, rose colored glasses, so everything you see is that color. But analytics by design is supposed to be an independent function. But because you’ve added it to be a reporting relationship under business, then because of job security or whatever other fear, would people not do something which they would have done if they were completely independent of any of these three functions, other groups?

Deepa Soni [00:34:51]:

Yeah. No. I don’t worry about that stuff at all.

Sanjog Aul [00:34:55]:

So you’re looking at more people motivated to do the right things, and they will go ahead and bring patterns through analytics to the business to say, hey. Look. You may be missing out on this particular opportunity.

Deepa Soni [00:35:08]:

Absolutely. And the whole culture is really leveraging analytics to shape the business. So it’s really an asset. It’s looked at as an asset. So it’s a different culture, if you would. Right? It’s a different mindset.

Sanjog Aul [00:35:23]:

So now let’s talk about because you you’ve shown that BMO has gone a long way in the journey, and it’s fairly mature, at least, based on how you describe. As part of your journey when you were not as mature and as you were moving along, what were the pitfalls, the gotchas, the mistakes that were made? And it’s your discretion, what you can share, what you cannot. But what can the listeners learn from a journey which led you to where you are today?

Deepa Soni [00:35:54]:

So I would take it outside of BMO. I would take it at a general level. Throughout my career, these teams have repeated themselves. So the gotchas and the pitfalls that I always encounter is siloed approach, which is one part of the business doing something, second part doing something, and it doesn’t come together. And then we’re reconciling between various different analytic reports that are going out. The second pitfall that I’ve always seen is lack of prioritization. So is it value driven? Do we have the right prioritization framework to be working on things? Because going back to the pace issue, if we’re not working on the most burning platform and working on something else, then everything seems to be slow and not of value. So that’s where the right prioritization to me is very critical. And there are more problems to solve always than we have the appetite or the money for. But I think getting the highest burning platform issue for execution becomes that pitfall to really avoid or actually go for in terms of getting the clarity of data. The third one I’ve seen is either too much focus on one or two use cases without consideration of that domain of the bigger domain, which is where you were going initially. If you set up the foundation, if you set up the use cases that’s very narrow, then you miss out on the piece where this could become a bigger value driver, but then you have constrained yourself. So either too narrow a scope or too much focus on one or two use cases. And then the last one, or at least one of the last ones is, sometimes there’s too much or unhealthy emphasis on what technology can do, versus how does it solve the business problem? And so these are the pitfalls that I’ve seen, which either dilute the business value of the data, or people start to say, am I really going to get value out of these efforts or not? So rightsizing all these dimensions to figure out what teams should be working on, where do we invest, where do we accelerate, I think it helps give data and data analytics a good name.

Sanjog Aul [00:38:10]:

Now given that you’ve got analytics embedded within business, and there is an expectation that analytics group or data analytics function will keep wowing the business to say, hey. This is the next frontier. But do you think we are ever seeing a vacuum that, okay, we are out of ideas. We had five use cases that I started with. As a team, we invested in all that restructuring. And then after a few proof of concepts or use cases, which you then eventually implement, you do not know what to do next. So are these guys still twiddling their thumbs?

Deepa Soni [00:38:48]:

I don’t think we’ll get there. I really don’t think, at least, I haven’t seen it in my career. I think there’s a couple of things. We’re never short of ideas. The question becomes which ideas to execute so that there is ROI on that, right? So that becomes a constraining factor of always able to prioritize. I do think, secondly, it’s not a static world. You may have five use cases today. But, two things. As the business evolves, those analytics diagnostics can take on a different flavor. So I think that drives the evolution. And the last part is technology is changing. So we used to be doing analytics on our own data. And now there’s so much outside the enterprise data that could be a big part of how we would look at analytics. So that is going to give us a whole new world of how this analytics could be used. So no. I don’t see that in the near future we would have done everything.

Sanjog Aul [00:39:54]:

So yeah. So maybe I should rephrase the question. So you will never be out of ideas, but when you really try to put, go sharp with your pencil and say, okay. I will get the most value out of this. I don’t find as many use cases. So a lot of people will give a lot of ideas, but there should be someone who’s doing the due diligence and say, nope. I’m not gonna invest in this. So do you see that there are always enough ideas which are viable and are worthy of the investment?

Deepa Soni [00:40:36]:

My experience would be organizations, when the businesses again, there is upfront larger investments that are needed. And as businesses get more mature and leveraging these capabilities, those investments start to shrink. They never go to zero, but they become a little bit less, right? Because you don’t need to catch up, and you don’t need to set the infrastructure and things like that. What is the steady state? I think would depend on each organization. Does that help answer the question?

Sanjog Aul [00:41:13]:

Yes. It does. It does. I just wanted to basically drill down a little bit because many times, when you do any of these innovative efforts, you talk about AI and IoT and many other initiatives. A lot of people start with a big fanfare and say, we’re gonna change the world, transform what you what we do for you. And after some time, that well of ideas dries up, and they don’t know how to move forward, how to continue to get funding from business to keep investing in that function. So data analytics is also seen in a similar way to say, okay. If I’m going to continue to have dedicated people on it, am I creating enough ideas out of the data set, which, yes, it’s changing, but is it changing enough? Am I done yet? So that’s why I actually asked you the question. Now let’s take the analogy of the sports team concept where we need a coach. We got these three constituents. But is there someone outside of these three who’s watching and shaping and shepherding them for data analytics function?

Deepa Soni [00:42:19]:

I think there’s quite a few variables in an organization to sort of answer that question, right? It could be the leadership. It could be the domain knowledge. It could be what exists in an enterprise versus at a business unit level. Different organizations probably will have to tackle this differently. But from my perspective, there are two or three core tenets, whether we have one coach or there’s a team for certain elements. So I see a lot of value, for example, in a centralized group being the COE, the center of excellence for best practices, bringing the setting up the infrastructure, the data platforms, and ensuring that the technology is accessible to whosoever wants to consume the technology. I think that is a role that should exist. The execution could be closer to the businesses we’ve been talking about. In the product ownership mindset that we talked about, the notion of a coach sort of starts to fade even more, because it becomes more of a high level ensuring that the standards are being used, the data is being consumed right, with the right governance, right structure. But in the product centric world, that team sort of becomes a self propelled team. So the notion of coach I don’t know if I have a very specific answer or a specific opinion about that. But I would think at an enterprise level, there should be somebody ensuring that the data is leveraged across the various constituents consistently and in a good manner.

Sanjog Aul [00:44:07]:

Let’s take a quick break, listeners. We’ll be right back. And then, we should reconsider this whole coach question because as a coach of a team, we are supposed to have this individual or maybe if you want to take an enterprise setting, a group, maximize the potential and performance of that team versus will they be functioning versus not functioning? So if we were to evolve them to be at their self actualization level, all different team members and leaders who are part of the business IT and data teams to do the best. Who could, someone from outside, someone from inside, could act as an individual or as a group, act as a coach? Please stay tuned, listeners. We’ll be right back and dig further.

Speaker 0 [00:45:01]:

Predict your company’s future by creating it. Is your workforce able to connect, exchange ideas, and share brilliance simply and securely? Create tomorrow, today. Empower your people to innovate anytime and anywhere with secure BlackBerry Enterprise Mobility Management and document sharing solutions. To learn more, visit blackberry.com/enterprise.

Speaker 0 [00:45:28]:

Patient centered care requires a connected enterprise. Are you ready? If you’re looking to scale your health care IT efforts, visit redmane.com/health today. Whether it’s to connect data from multiple partner solutions or developing software for unique needs, Redmane can help. To find out how Redmane can help your company deliver on the patient centered care promise, visit redmane.com/help or call (773) 693-3919. Visit today.

Speaker 0 [00:46:01]:

Your growing business needs a highly productive workforce, effectively communicating and collaborating without exposing corporate data to cyber attacks. Are you looking to balance security and workforce productivity? Move beyond short term measures and securely scale your business with BlackBerry Enterprise Mobility Management Solutions. To learn more, please visit blackberry.com/enterprise. You are listening to CTN CIO Talk Network with Sanjog Aul. Now back to the show.

Sanjog Aul [00:46:45]:

Welcome back. So digging further, Deepa, is the coach’s role to maximize someone’s potential versus have them do the job as expected. So when we have this rather complicated but later on streamlined team of business IT and the data team members, they’re working together, moving towards a given direction, trying to make the best they can, but do they have blind spots? Are there places where they could be thinking fundamentally differently, or one idea that spark will allow them to take a quantum leap in their effectiveness and efficiency? Is that even considered a role which would stay independent of the function itself and all people involved in it and still give them outside an input? It could be an individual or a group. Has that idea occurred in an organization?

Deepa Soni [00:47:49]:

So I would say, not in a very clear, transparent form, but the banks or the organizations that I work for always are looking for that calibration. I call it calibration with the rest of the industry, rest of your peers, how you’re doing, what are you leveraging the most, what are you not leveraging. So, the way I would think is the CDO, one part of the CDO role is to make sure that we’re calibrating ourselves, and they do that independent of what’s being executed at each of the business units. At an enterprise level, they are calibrating us as compared to where we are. And there is also a group of executives from each of the businesses that’s a virtual group that is looking at this stuff.

Sanjog Aul [00:48:39]:

Sure. So there is a specific instance where an organization, and I know about them, built this data analytics group, and they built a COE. And when COE went to business and said, hey. We want to start proactively, and we want you to adhere to certain standards. If you would do that, that will help you. Not that they came with as an order, but they said it will help you. So they took the meeting. They had the discussion, did not do anything in relation to that, and came back a month later, going the old way or the way they should not be going. Back to the COE that this is what we wanted to do, and that COE was bundled into a data analytics group. So what went wrong there?

Deepa Soni [00:49:33]:

I think this is, again, that operating model discussion. We, as data and technology leaders, have to convince the end users that there’s a bigger good in doing things a certain way. There’s enterprise value. And this is really going back to how do you operationalize that paradigm and things you do on a daily basis. So could we help you set up doing the things in a new way so the burden is not on the business? There are multiple ways you could go about it. And I think there is a change management that we always have to be very cognizant of. Things don’t change in one go. So I would say, in general, it’s a change management paradigm. It’s leadership. It’s top down executive support. A lot of these things help make that change perpetuate in the organization. Otherwise, it’s a harder change. And sometimes, you do have to realize where change would be adding value and where change wouldn’t be adding value. So I think that really boils down to how the CIOs, CDOs and the business leaders come together and shape a strategy.

Sanjog Aul [00:50:42]:

So if COE was not listened to, that could also mean lack of respect or there was a different agenda. So would somebody from outside come and intervene in this, or would you think the COE leaders are the ones who have to step up their game so that they earn the respect so that other people start listening to them?

Deepa Soni [00:51:01]:

I would be in the camp of COE leaders need to drive that change. And I’ve seen it on multiple occasions where they’ve done it in my 18 years of working in technology. It doesn’t start with first time everybody agrees to doing it a certain way. But slowly and steadily, you have to get the buy in, get the value clear, clarified on what COE is asking for, why is it important, why is it important for the enterprise, and then what are the downside ramifications of not sticking with that standard? At some point, it’s gonna hurt the end user. And what does that mean? So I think it’s a lot of change management and the buy in and operationalizing the way we work with the businesses.

Sanjog Aul [00:51:48]:

One last question, a minute or less. What would you think should be the change in the leadership style or the leadership mindset of the business IT and data functions so that they are able to help take this data analytics group to the heights it can reach?

Deepa Soni [00:52:09]:

I think it’s no different than technology or any other topic that we talk about, and the mindset is really having that clarity of outcomes, having the agility, having that collaboration. And I would say figuring out a way where there’s not a lot of friction in the system. And that doesn’t come naturally. But we have to get to a point where the friction between the teams becomes less, and the focus is higher on the outcomes and agility and delivering value. Easier said than done, but to me, doing that is the only way to get to the next level of value driving.

Sanjog Aul [00:52:54]:

On behalf of the show and our listeners, thanks so much, Deepa, for sharing insights on how organizations can best develop data analytics teams with the very required playbook to win in this game. Thanks so much.

Deepa Soni [00:53:07]:

Thank you.

Sanjog Aul [00:53:08]:

Please like us on Facebook. Search for CTN. That’s CIO Talk Network. Be sure to follow us on Twitter and join our LinkedIn group. Thanks again for listening to this segment on CTN. This is Sanjog Aul, your talk show host. Till next week. Take care, and God bless.