CIO Strategy Enterprise Data and Analytics with Bruno Aziza of Google Cloud CXOTalk 730

Show video

This is about looking across all your workload,  so if you're using transactional or analytical   workloads, how do you bring all of that together,  reduce the cost, reduce the maintenance headache,   and create an environment where now you  can scale technically and also humanly?  Bruno Aziza is the head of data  and analytics for Google Cloud.  When I started, not many people cared about  databases and back-end issues when it came to   data. They felt that it was a necessary evil. Now,  organizations are realizing that if they're able   to harness data faster than their competition,  they really are able to do amazing things.  Great examples that you might be familiar with are  things like anomaly detection, fraud analytics,   or product recommendations. Think about, in your  daily life, you go on a particular website and   you're kind of shocked sometimes that a website  knows you more than you might know yourself.  

They're able to recommend amazing content and  products that you might not have thought about.  All these systems, all the work that goes behind  serving you content that's highly personalized,   highly relevant, and makes a great experience  for you typically is powered by our technology.  One of the significant challenges is choosing  the right problem and the right set of data.   How do you think about that? How do  you go about making those selections?  Think about the problems that are the  most related to the business value   that is driven for your organization. I think,  if you look at the average tenure of the chief  

data officer, it's about a thousand days,  probably a little bit less than a thousand days.  I think the reason for that is often the  opportunity for data is so big that you   tend to want to do everything. You end  up just focusing on the business value,   the business metrics. What is driving the bottom  line with data? There's a huge opportunity there.  The two areas that we see people not fail but  kind of lose their way is when you look at what   I call the "why nots." Why not would we look  at this use case? That sounds interesting but   actually might not lead to a specific value. Then there are the other use cases that sound  

interesting because they are highly  innovative, but they're really not   connected to some of the core issues  that your organization is trying to fix.  What I would advise every chief data officer  to prioritize their use cases with is not   fix the existing or find net new ones. It's  double-down on the use cases that your chief   financial officer, your CEO, your COO, your CPO  (chief product officer) is going to tell you,   "I can rally behind that because it's going to  drive bottom-line value to my organization."   I think this is really an important consideration. This seems easier said than done, and many  

data analytics folks, data science folks  have a real tough time getting into the heart   of the right problem. How do you winnow down,  and what advice do you have for folks who   are struggling with that? I think about what I call the five S's   of the data opportunity. If you break them  down, you realize that there are modern issues.  The first S is solve for speed. If  you look at the opportunity with data,   you're going to live in a world that is real-time.  The ability for you to empower your people,  

drive value for your organization, is about  delivering the right data at the right time   and using speed as an advantage. That's your  first S that you really should focus on.  This idea of discarding real-time  infrastructure is really not a good   idea. Real-time is now becoming just a must-have  for every organization. That's the first S.  The second one is scale. Some organizations  that we start working with tell us,   "I'm not in the big data space." The news  that I have for you is that everybody is. 

In fact, we just call it data because if you  look at what we're doing here, there's so   much data that we are creating. Two-thirds  of the data that any organization creates   actually never gets analyzed. In fact, the  economists, I think, came up with research   showing that less than 5% of the data actually is  a potential get in front of people that can make   decisions on. This idea of building for  scale is extremely, extremely important.  The third S is around security. We talked  about speed. We talked about scale.  

Security is really important to build in  the very first stages of your data strategy.  More data means more responsibility. We  know sometimes people don't think about   governance until they've started to  kind of build their systems. You need   to build with governance from day one. Now, you don't need to just be focused   on governance only. You don't want to be the  guardian of data and people fear using data.   But you do want to think about centralized  governance to enable your people. 

The fourth S is around simplicity. I know  we'll talk a lot about this today, but   Google was certainly late with this issue and  this opportunity to deliver simple interfaces   into your user community and still  get the sophistication of the issue.  Customers that we work with are challenged  with, "My environment and my problems are   highly sophisticated, but I need an interface  that's simple, so I can get adoption." If  

you're a data scientist, think about that. Then the only way to get from sophistication   to simplicity is my last S. It's what I call the  smarts. Artificial intelligence is going to be   the secret sauce on how you enable this. The world of data is growing. We're not  

going to talk about how the volume of data is  just out of control. But you have more data,   more people that want it, more use cases, so you  need to use artificial intelligence, automation   to augment the capabilities you have today  so you can deliver data with more simplicity.   That's the framework that I think  about when I work with organizations.  Can you give us an example of a data problem  that meets all of these criteria and tell us why?  We talked about the issue  of product recommendation,   for instance. It does meet all of that. For instance, you'd be a customer of,   let's call it, a retailer. You're going to have  a multi-channel relationship with this retailer. 

You're going to go to their store. You're going  to visit their site. You're going to visit their   partner sites. This ability of thinking at a  high level of scale is really important because   the data points I'm going to get about Michael  are going to be coming from multiple places.  Also, the ability that I have to project  an option for you is related to this speed   dimension that I talked about. It's not  useful, Michael, if I suggest you a pair of   shoes after you just went  through the cart experience   and you've already bought that pair of shoes.  It's not really interesting to do that for you.  Also, building a platform that controls the  way the data is being used. It's your data.  

It's not data that should be accessible by  anybody else. Having a strong governance   backbone to make sure that even though as  a retailer I have a relationship with you,   I've got to make sure that I've built an  infrastructure so your data doesn't get away.  Think about not just your clickstream experience,  but also your credit card information and so   forth. You've got to think about that dimension. Then artificial intelligence, like I just said,   I can look at your profile and maybe what  your clickstream information is, your cohort   information, people that are just like you  that I've made the right choices. Maybe they've   taken advantage of discounts or group products  in particular formulas where I can create   an interesting experience for you. Retailers, financial institutions,  

telecommunications organizations have  this huge opportunity to use data and   artificial intelligence to create compelling  experiences for their customers and that return   better value for them in the end. Hopefully, I touched on all the   dimensions that we just talked about here with  this one example of product recommendation.  How much do you care about and focus on the  infrastructure? The issue of prioritizing the   business problems is at a pretty high level.  Let's go all the way down to the other end,  

the infrastructure. How important is that  and where does that come into play for you?  Extremely important because we're now in a phase  where organizations are moving to this modern   data stack. You might have heard terms like  the data mesh, for instance, an important term   that has been popularized over the last few years. The way to think about it is to step back and   look at, what are the issues that your  organization is trying to solve? We see,   typically, three phases that organizations go  through. I'll try to break them down for you. 

The first one is what we call the data ocean.  Actually, it's not me that calls that. It's   customers like Vodafone, for instance, who  popularized the term a couple of years ago.  The data ocean idea is that you want to broaden  your perspective on where your data is as much as   you can. This is capabilities around multi-cloud.  Catalog technologies are important here.  This is about looking across all your workloads.  If you're using transactional or analytical   workloads, how do you bring all of  that together, reduce your costs,   reduce the maintenance headache, and create an  environment where now you can scale technically   and also humanly? As you have the opportunity  of looking at so much more data, you're not   going to hire a lot more people to match that  scale, so you want to solve for the data ocean. 

The technical infrastructure and the data ocean  is different from the other two phases. What are   the other two phases? The next one, as we just  talked about, is this concept of the data mesh.  The problem you're trying to solve here in the  data mesh is often people create data lakes. You  

hear this terminology saying they've become  data swamps. The reason for that is because   the data is stored, the data is observed, the  data is cataloged, but it's not really acted upon.  The concept of the data mesh is how do I  create federated environments so now I can   activate my business communities with data  analytics environments that are relevant to them   while I'm governing data century and that I can  make sure that the data that people start with   is the data that makes the most sense  for the organization. Data mesh is about   going from passive to active data. Data fabric  technology is important here. And this idea   of federation, access to analytics is important. Finally, the third phase is where you are starting  

to think about you're evolving to now bringing  new personas. In the first two phases, the data   architects, the data scientists are important. In  this third phase of what we call the data factory   (popularized again by McKinsey) is this phase  where now you're building data products.  Your chief product officers might come into  play here. You might use universal semantic   layer technology. It creates data-driven  applications so you can repeatedly create   products like the ones that we talked about,  like product recommendation, anomaly detection,   fraud analytics – all data products – all the  way to creating your customer data platform so   you can really have a 360-view of your customers. I think the infrastructure is extremely important.  

What's also important is understanding your level  of maturity and how you can accept this technology   not just from the technological standpoint  but also from an employee maturity standpoint.   You're going to have to do a lot of training,  a lot of communication to make sure people know   what is it that they're trying to achieve. Again, you're always linking everything,   all the technology aspects back to the  business problem that you're trying to solve.  It's the business problem and it's also the  organizational footprint. The reason for why I  

really like this idea of the data mesh is because  it describes what you're trying to achieve,   how you should line up your infrastructure, but  also how you should think about your organization.  We have lots of organizations that are asking,  "Hey, should my data people report centrally into   one organization? Should they be distributed?" When you have these goals of the data ocean,   the data mesh, and the data factory, you start  now thinking, "How do I align my organizational   footprint to best serve my business goals?" I think those are important considerations.  As a vendor, I know you're probably expecting  me to talk to you about products all day long,   but I really think that the success of  the data analytics and the data platform   strategy is highly bound by your ability to  galvanize your people, train them, and get them   to work with you on achieving your business goals. Let's take a few questions from Twitter. First,   we have Chris Peterson who asks, "In terms of data  security and privacy, how does Google navigate the   maze of different regulations internally?  How do you as a company manage this stuff?"  We internally probably have one of the most secure  platforms for just managing data. You can look up   this technology called Access Transparency and  so forth. We also have specific industry teams  

that normally are working closely with  customers on these issues, but also very versed   with the issues for every specific industry. I would say we build the security by design.   Just like I was advising any company to  think about it for their own data platform,   that's what we've been doing on our own platform. In fact, a lot of the issues that we're helping   customers with today are issues that we have  solved for ourselves. I think that's probably   one of our competitive advantages (if I could  talk about that). We relate very well with what   is it like to create compelling experiences  for the future of the analytics consumer.  If you think about your own organization, you're  going to want people to consume information   the same way that they go to Google.com today: A  simple interface. It doesn't require any training.  

Provides high-level sophistication,  high-level personalization, but still   through an experience that's highly simple. Another question that's just come up on   Twitter from Arsalan Khan relates to this  complexity, to another aspect of this complexity.   Arslan says, "Collecting this kind of  data at this scale has real cost." He's   wondering what smaller organizations can do  to take advantage of data despite the costs.  I forgot to mention that one of the important  considerations is this domain of financial   operations, particularly in the data mesh world.  In the data mesh world, if you can imagine,  

now you have centralized data governance.  You're creating your data hubs and your   data neighborhoods (as some of the  customers that we work with call them).  People start consuming and driving some  compute costs, so how do you manage that?   It's important that you think about choosing your  platform that has flexible financial ops options.  What that means is we can create reservations.  You could say, "Well, we're going to spending   up to this cap. We don't spend more than  that." Or you can allocate particular compute   capacity to specific workloads that you can either  predict or that you can kind of give a range to. 

What I would say are two things. One  is, there's the management of the costs,   but there's also, when costs increase, it's  not always bad news. It's also that your   people are actually engaged, and they are  actually using the platform to drive value.  I would not just look at costs by itself. I would  look at price and performance relationships,   and I would look at costs and value relationships  because, if you look at our industry   today – we've been working on it for the last 30  years – the adoption rates of technologies are   very, very low. In the business intelligence  space, we're talking about 30% adoption. In the  

AI space, we're talking about 35% adoption. In our case, I look at machine learning   that is deployed through BigQuery. We're  seeing 80% of our top 100 BigQuery customers   use artificial intelligence. What that means is  that they are getting value. They're getting to  

usability a lot faster, and that's not  always bad news for your organization.  This seems one of the core issues  that chief information officers   must grapple with because the mandate for many  CIOs is to innovate and, at the same time,   do more with less, right? We want you to be the  driver of innovation but do it with less cost.  It's a huge challenge, but  it's also a great opportunity.   Like I was just saying a little earlier, you don't  want to be the executive who restricts access   to data. You don't want to be the executive  that slows down innovation. You want to be   the executive who is lining up to the business  objectives of the organization and provides   a platform that is driving innovation. If you think about it, innovation in any   organization is going to come from the front-line  folks, the folks that are in closer contact with   the customers, and so forth. Enabling that model  – that's why I keep going back to the data mesh,  

I think it's a great model to follow –  is to really think about business goals,   technology stack, and organizational structure. Often, customers ask us, "Where should my   analytics folks report under?" You've got to think  about how you organize yourselves so you can get   to innovation faster than any other organization. I think, today, that's really the issue that   we see is people do a lot of POCs (proofs of  concept), but they're having an issue getting   into production and then innovate on top of  that production. We want to simplify that,   and we want to get to a more liquid (if  you will) relationship with your data.  It's very interesting that when you talk about the  data mesh, you talk about the business goals, the   technology, and then the organizational structure.  Why is the organizational structure so crucial?  Things get done through people, and they get done  through people that have shared goals. I could  

give you the best technology, but if it's deployed  in the wrong system, it can't really help you.  We did a survey a few months ago where we asked,  "Where should your data analytics powerhouse be?"   We asked people, "Should it be under your CTO?  Should it be under your chief product officer?   Should it be under the CFO? Should it be  under the CMO (chief marketing officer)?"  What we found is, one, the answer kind  of depends, but it also is related to   the types of executives that you have. About ten years ago, I wrote a book called   Drive Business Performance, and it was based on  interviews of organizations that had experience,   amazing success at driving a data culture  inside the organization. The key was  

finding the right executive and getting the  mandates into how we're going to make decisions in   our organization. You don't want to discard that. Sometimes, people look at their initiatives and   say, "Oh, I just have sponsorship," and  the sponsorship is good. It's necessary,   but the mandate from your top executive is  saying, "We will now make decisions based on data.   We will now go out and look for opportunities  to measure things that maybe we couldn't   measure before. But because we know the  business needs it, we're going to do it."  That's critically important and it's way more  important, I'd say, than any of the nicest and   latest technology you can acquire. If you  don't have the organizational footprint,   if you don't have the mandate from the CEO, you've  got a great Ferrari, but you don't really have   the keys to it, so why turn the engine on? You know it's kind of funny. What you're  

describing is almost a cliché and almost  extremely obvious that we need to, again,   align the work that we're doing to solve  the problems that we think are important.   But why is it so difficult to actually  achieve that very simple goal?  There are a few reasons for it. I  think one is we live in the time where   there's a lot of innovation, there are a lot  of buzzwords, and there are a lot of vendors. 

If you look at the data landscape that was just  published, it's something I wrote about in my   Forbes column, there is a high proliferation  of solutions. There's a lot of innovation   in technology. The cloud is a lot more available  than it might have been just simply ten years ago.  I think there's a flurry of  options being thrown in front of   CIOs or chief data offers. It's really hard to  kind of parse through that noise. Sometimes,   as technologists (and I'm a technologist myself),  we might get enamored by, "Oh, this is a cool   concept. What if I deployed this?" I think  this makes it a little bit hard for leaders. 

Secondly is that chief data officers don't think  about themselves as business leaders. They think   about themselves technically as technical leaders. What we work with organizations with is,   when you drive your initiatives, do you have  a brand for your initiative. Do you have a   communication plan for your initiative? As technical leaders, we really think that   there's a part of marketing the solution, if you  will, back into the organization that matters to   succeeding. I think the combination of those  explains why there's not a lot of success.  If you look at the latest research from  Accenture, I think it's 68% of organizations   can't find value from the data that  they have. That's a huge number. 

Also, it's just simply difficult for chief data  officers to stay in place. I think I said earlier,   I think the average tenure of the chief data  officer is less than a thousand days. Primarily,   it's related because we found that they rarely tie  their technical initiatives to business goals and   they might not think about the communication of  those business goals. And so, that kind of hurts   them in succeeding with their data strategy. All of this begs the question,   who should own these data science efforts?  When you say that there's a disconnect   between the technology problem that is being  solved and the business requirement, there   has to be a cause. Who should own these efforts? What we found in some of the surveys that we did  

was primary two executives that this falls under.  The First one, interestingly enough, is the CFO.  When we asked our community to tell us, "Where do  you think data science and analytics should roll   under?" a large percentage, 34% (not the majority)  of people said primarily the CFO. The reason for   that is because, I think, over the years, the  CFOs have gone more from the back-office, cost   retention type of role to innovating, using data  as a way to power their organization and drive the   operations of the organization. Data analytics and  data science have a great opportunity to do that.  The CFO might actually be. Again, it depends on  their objective. And so, you have to decide for   your own organization. Actually, it  might be a good group to own this.  The second one has been the CTO, the chief  technology officer. Again, it depends on their  

style, their team, and so forth. I  think the reason for that is because   you want, on your bench, a good amount of very  technical folks. We have seen, over the years,   the data scientist was the sexiest data role.  Now it's going to the machine learning engineer.  You think, wow, it's getting more and more  technical. I think the reason for that is because   the industry is innovating really, really  fast. And so, you want technically savvy folks  

to enable you to deploy, but you want  to couple them with your business folks.  It's probably harder to learn the business coming  from a technology background than the opposite   (at the moment) because we're now building  technology that is taking over a lot of the tough   steps that you might have needed to learn.  I think about auto data preparation or   auto data quality and all these steps that now  business analysts can come in and start using.  What I would say is, what we see organizations  do is they look at their blueprint. Say you  

have 100 people to handle all the data  analytics issues. They tend to put half   in the central business unit and then half  into a central corporate unit like the   CTO's organization or the CFO's organization. You said that CFOs could be the right folks to   manage or be responsible for these data efforts.  But my question is this. CFOs, in general,   may understand technology but certainly, by and  large, don't have the kind of deep expertise   that's required from a technology  standpoint. And so, how is it practical   for a CFO to manage this? By the way, why didn't  you say that the CIO should be responsible?  There are two aspects to the answer to this  question. Why not the CIO is because, at least  

from what we're hearing from our customers,  typically if your CIO is focused on internal   technology and infrastructure choices, data and  analytics tend to be an application business.  We don't think about this today, but they are  a business application consumed and directed   towards value creation. I think that's why the  CFO comes in here is because modern CFOs don't   think about just budget and reducing costs. They  think about opportunities for us to create value.  A great example of that is now organizations  creating data products that they will monetize.   I think about retailers, for instance, one  of the great organizations I'm working with.  Carrefour is one of the largest retailers  in the world. They've solved their data  

mesh issues for themselves and now have  built around it, and they are now starting   to create data products that they can sell back  into their community, which is now talking about   driving revenue for the organization. I think the mindset of the most   innovative organizations is that data is  not a liability. Data is not something that   I guard only. Data is something I build upon.  It actually becomes an asset for me to manage  

up to a point where I can create products  off of it and monetize these products.   I think that's why customers are saying what  they're saying around where it should fit.  We have another interesting question from  Arsalan Khan on an aspect of this topic. He says,   "As organizations become increasingly reliant  on AI and machine learning for decision-making,   are some executives resistant to accepting data  as the ultimate decision-maker?" In other words,   if I can rephrase it, what should folks do  if a business leader rejects the conclusions   that the data presents? "I don't think this is right. Yeah,   sure. Your data says whatever, but I know  from my experience it can't be right."  This is the typical gut feeling that we deal with.  The issue with a gut feeling is you never know if  

it's actual experience or if it's indigestion.  You don't want to just rely on gut feel, but   it is true that if you read a theory on this and  books from Malcolm Gladwell and other folks that   are very educated and well researched, the right  decision is going to come from the combination of   really good data and experience around the  mistakes that maybe you've made or maybe others   have made that you've been able to learn from. I think, in general, it's never a good idea to   decide 100% on your gut feel. You might get  lucky every once in a while, but now we have   technology that captures enough that you are  able to not just understand but, in many cases,   predict. And there many great stories like  this in baseball, in the wine industry,  

and others like this that we can all relate to. I think you're always going to get into a   conversation with an executive that maybe might  not believe the insights that you're bringing in.   This is why, in the last few years, you might have  seen the work from Nancy Duarte on storytelling,   so connecting with the emotional aspects of  how this executive might relate to the data.  In the book that we wrote a few years ago, Drive  Business Performance, we talked about the example   of Lego where the data analyst not only presented  the information, the dashboards, but they actually   had the voicemail left by the kids being played  to the executive. The executive, as a parent,   could relate to the customer feedback they  would get and actually did change the strategy.  What I would advise our friend  Arsalan here asking the question is   don't think about just the binary logic aspect on  how you're delivering the data. Think about the  

emotional aspect, the way people make decisions,  even executives with great experiences, how they   emotional connect with the data. That's really  important as well to build into how you present   your results to folks you're trying to convince. We have two interesting questions from LinkedIn.   This is from Prashant Motewar. He  says, "Number one, what about the   chief digital officer as the owner of data  and analytics? What do you think about that?"  Given the past couple of years here where  digitalization has really accelerated,   we see certainly in organizations that would  gather a lot of their information from physical   locations – retailers and financial industries  – nobody goes to the branches and nobody goes   to the stores, so the person that's in charge  of digitalization and taking this system into   the future is certainly going to be interested in  collecting and understanding data a lot faster.  What I would say, though, is it's not just the  title that you have to look at it. It's also   the organizational footprint. The people  under this leader, is it the right talent?  

Is this the right organizational  footprint? Do you have shared goals?  One of the important best practices that we see is  that it's not just the CTOs job to innovate with   data. It's the rest of the organization. We worked with CIOs, CFOs, CTOs who   share business goals that they actually  don't have the direct impact into it,   but they also have shared goals with the business  folks who do not have direct impact on the   metric itself. But the point here is to get  them to get together, align, and collaborate.  The chief data officer and chief digital  officer are great roles, but I wouldn't just be   wedded to the title. I would look deeper into the  organizational footprint of that organization.  Another excellent question from Prashant, an  important question. He says, "Data insights   are extremely valuable when delivered at the right  time to the right people with the right context.  

Any point of view on how to enable this?" A few years ago, I did a keynote at a data   summit. I came up with this acronym, and not a  very pleasant acronym but at least memorable,   called RAT. R-A-T because data needs  to be relevant, actionable, and timely.  You're absolutely right in your analysis.  If I told you here's an umbrella because  

it rained yesterday, it's not very helpful. There are a few best practices here. The first one   is actual data literacy across the organization.  It's one thing to deliver the data,   but it's also another thing for people to actually  understand the data. We also did a survey on how   many data employees should you have inside your  organization, so when you deliver the data,   people understand what to do about it. Keron Bourne, who is a member of the community,   had (I think) the best answer where he said,  "100% of your employees should be data literate."   What that means is they should be able to  recognize, they should be able to understand,   and they should be able to talk  data. I would make sure that at least  

all your employees understand the opportunity  they have with using data so that when they get it   in their context, they can use it. Then a third of the organization   should be data fluent. What does that mean?  That means they should be able to analyze.   They should be able to create arguments. They  should be able to present results visually,   emotionally to their management, to their peers. Then 10% of your organization should be data   professionals. A data pro here is someone  who is paid to create value from assets.  The reason for why I'm saying this balance of  roles matter is because the issue sometimes is you   might be presenting data in the right context but  to people that might not know what to do with it. 

What we tend to forget is we're not  in the business of building folks   to become data specialists. We're in  the business – you're in the business   of doing your business. And so, to be able to be  equipped with that, you have to deliver the data   on time to an audience that is willing to or is  equipped to act on that data. That's the most   important thing is how do you act on that data. We have another question now from Twitter from  

Lisbeth Shaw who says, "In a busy company,  should the person in charge of data science   repurpose their old BI data strategy for today's  data science needs? How should they do that?  Too often, we feel like we have to hire  outside folks to come and solve a problem   because the other folks have a data scientist  title and my folks don't have a data scientist.   What I would say, I just talked to a chief data  officer today who is looking at how she is going   to upskill her team, and she is absolutely  starting with the existing talent because   institutional knowledge of your organization,  knowledge of your customers, the knowledge of your   organizational process is critically important. Yes, you can bring outside talent that is   technically gifted and so forth, but you'll never  be able to hire enough of them so you can tackle   the problems that you need to tackle and often in  a very timely manner. The reality is that there  

are toolsets now that enable business analysts  to step up into a data scientist type of role,   and so I would never discard your existing talent.  I am sure many of them are capable of doing more,   either because they are motivated to get  the training or because the toolset that   is presented to them is making access to  data and working with data a lot simpler.  I'll just give you an example on our data stack  with this product called BigQuery. We have this   embedded machine learning capability  inside BigQuery, which means that   you don't have to move the data. You don't have  to set up an infrastructure. You can trigger  

models and run them with just a few lines of SQL. Just these few lines make business analysts able   to do work that in other platforms would require  a lot of code from a machine learning engineer.   And so, I think the good news here is that the  industry – cloud vendors like us and the rest of   the ecosystem – is really driving to making  tools easier to use, which means (for you)   you can use your people and upskill  them into where you want them to be. 

Certainly, when I talk with CIOs, the  idea of low code, no code products   is right, front, and center of how you can help  your organization innovate while reducing costs.  Absolutely. No code and no code is probably one of  the most disruptive trends of 2021. It's primarily   because it's enabling business users to create  business purpose, domain-specific applications   for themselves. I believe that the world  of package applications is going to be   disrupted over the next ten years because  now these business users are able to just   solve a problem by bringing in services, kind of  in a composable manner, to create an application   that's relevant to them and their community. Definitely, the tools are getting easier.   People are getting more skilled. There's more of  an acute need for working with more data. I think   the combination of those things doesn't mean that  you should look for the answer outside of your   organization. The answer often starts with you. We have another question again from LinkedIn.  

This is from Scott Beliveau. He asks, "What's  the next data thing that organizations are not   talking about and looking at but should be?" I look at a lot of technology. Recently, I bought   myself these glasses, these smart glasses that I  can talk to and they can do things for me and so   forth. I think what we're not talking about enough  is what is the future of the interface into data.  Today, we're used to our keyboards  and we're used to our phones.  

But the reality is that, in the future – and not  too distant of a future, I believe – our voice,   our eyes are going to be how  we interact with information.  There's a lot of investment going  into natural language technology. It's   a huge field around natural language  processing, natural language understanding.  It reminds me back when I was at Microsoft  where we shipped the Xbox. The Xbox had this   amazing camera called Kinect. It was one  of the most popular devices. It was this  

camera that would scan my body and the  tagline was, "You are the controller."  I think that's where we're going with  data. Ultimately, to broaden the appeal   of data and make data more engaging for more  people, we've got to change the relationship.  We're now in the mode where we have to talk  machine. I think, in the next ten years,   the machine is going to talk human, and that's  going to empower new use cases, new experiences   at home with smart home devices that you can  call upon to set up your alarm or get answers   and so forth. I think that's what's going  to happen in the corporate context as well. 

Bruno, we're just about out of time. Any final  thoughts or words of advice to folks who are   listening regarding being successful with the  data, aligning data efforts with the business,   the things that are really hard? I do have one thing that I think   is important for folks to consider  is this whole domain of data culture.  One of my good friends Randy Bean just  published a book. I think he calls it Fail Fast,   Learn Faster. I might be misquoting the  title, but I'll send you the exact link.  In it, he's interviewed data leaders and asked,  what is getting in the way of being successful   with data? It turns out that the majority  of them are saying it's the data culture.  What I did is create my data culture checklist.  We won't be able to go through all ten of them,  

but I'll give you a couple of best practices. The first one, we talked about branding.   Brand your initiative. Think about it  seriously. Create a logo. Have certified data,   so people recognize the quality of your data.  It's really important in making sure that   you take data seriously at your organization. The second is have what I call a decision   inspection. What I mean by that is  often we do a postmortem. Often, we  

look at failure, and we say, "Okay.  Let's try and understand how we failed."  Two best practices: Are you doing premortem?  Premortem is, if everything goes wrong,   what does that look like? Are we ready to react  to a situation when it's going to go wrong?  Then the second practice is, analyze your  successes. Often, you succeed and you move on,   but do you truly understand why something really  succeeded? Having this decision inspection   mentality, I know it's hard because everybody is  moving really fast. It's hard to prioritize. It is   the best way for you to build this data culture  that ultimately is going to make you one of the   most innovative organizations in the industry. Bruno Aziza, thank you so much for taking time to  

speak with us today. It's been an action-packed  45 minutes. Really, really appreciate it.  Thanks for having me, Michael. Everybody, thank you for watching, especially   those folks who ask such excellent questions. Now,  before you go, please subscribe to our YouTube  

channel, hit the subscribe button at the top of  our website so that we can send you our newsletter   and you'll get notified of our amazing upcoming  shows, check out CXOTalk.com, and tell a friend.

2021-11-24

Show video