[ad_1]
The world of data has changed significantly, with data today having become highly distributed. It’s stored on-premises, in data warehouses and data marts, and in the cloud – both singular, and plural.
Data is created by devices and applications that didn’t exist a few years ago, and today, needs to be transformed, augmented, analysed, and acted upon in real time.
Bruno Aziza, head of data and analytics at Google Cloud, is focused on this in his role, supporting customers to navigate these challenges.
In the video interview embedded below, Bruno discusses several pertinent topics around data and analytics, including:
- The challenges today’s CIOs are faced with when making sense of the data market
- The biggest shifts in the big data landscape since the pandemic
- Learnings we can take away to inform business strategy in 2023
- What businesses need to consider when building their “modern data stack” in 2023
- How recent developments from Google Cloud (announced at last month’s Next event) are helping arm CIOs with the tools they need heading into 2023.
The video interview is embedded immediately below, after which is a transcript of our discussion, so please watch, and read on!
Below is a transcript of the video done with Otter.ai and tidied up, apologies if there are any errors, please see the video above if there is any confusion!
Alex Zaharov-Reutt 0:07
Well, hello, and thank you for joining me for our iTWireTV interview. I’m joined today by Bruno Aziza. He is the head of data and analytics at Google Cloud – welcome to the program!
Bruno Aziza 0:19
Well, thanks for having me, Alex.
Alex Zaharov-Reutt 0:20
Thank you for taking the time! Now Bruno, it’s nearly 2023. How has data changed significantly over the past decade or two? What is normal today that was once just science fiction
Bruno Aziza 0:33
I’ve been in this space for about 25 years, and the industry has changed quite a bit. When I started, I won’t say nobody cared about data as much it was a back office type of set of actions. And then these kinds of tasks that you need to do take, I think now organisations are seeing it, as the engine for innovation and organization, their relationship with their customers is changing, it’s multichannel.
It’s the digitalisation of their relationship with their customers. If you’re a retailer, if you’re a financial services, organisation, you have the opportunity now to know a lot more about your customers and create those compelling customer experiences. And so I think what we’re seeing is, first of all executives pay attention to it.
Second, it is an engine of growth for organisations, I suppose as 25 years ago, and it was one of the things you needed to do. And then I think third is probably the expectations of consumers in general, where they expect a better relationship with the organisations that they, , interface with, they expect them to understand what they need, they expect customisation at very large scale.
And so all of that, I think, is creating a great environment for the space, I’m really happy to be in this space. Because I really think in the next 20 to 30 years, it’s going to be the technology that the technology executives need to focus on if they want to bring value to their organisation.
Alex Zaharov-Reutt 1:46
Look, I certainly know the value of when I first installed the Google Desktop Search program, which would have been probably nearly 20 years ago.
And suddenly, all that information on our computers was searchable. I mean, that was the revolution that sparked Of course, Google itself with the search engine having to the ability to fine tune all that data, but the granularity of the data that we can collect that that was always there. Accessing it all has become second nature.
And you’re working right at the pointy end with the biggest of data sets. And obviously, it’s a very exciting time for the world.
And as you said, 25 years, you must be overjoyed at how far we’ve come. And yet how far we still have to go!
I always like to talk about this as the black and white era of computer technology. I mean, sci-fi writers have done incredible things. So how has our ability to interpret, transform, augment, analyse, and act upon this data in real time, but also different timescales, some data doesn’t have to be acted upon in real time. How has that improved over the last few years, based on everything you’ve done, and all the companies you’ve worked for, in this space?
Article continues below ad, please read on!
|
|
Bruno Aziza 2:49
I’ve been very lucky to work with some of the largest innovators in the space over the last 25-30 years, I’ve worked with all types of organisations.
So it’s really small organisations that I helped launch in the data analytics space, like, I’ll find the labs inside sands and at scale. So I had the opportunity with work to work with our initial like business objects that really grew and went IPO very quickly, and really kind of help organisations with reporting and analytics.
And then, most recently joining Google, I also worked at Microsoft for seven years. And I think what’s really happening in the industry is, there’s a kind of convergence of a few, a few big trends that people need to pay attention to. The first one is this idea that you need to have systems that allow you to have limitless access to just about any data.
And you talked a little bit about it here, real time data, but also data of all types, structured and unstructured, semi structured, you really need to move away from this idea of having silos of multiple data mart’s for a function, I think customers expect now that if the data exists, they should be able to query it, and not have to worry about capacity or setting up environments like we used to, in the on-prem world.
The second area, that is a big focus for customers, this idea of governance, and I call that governance with a big G, because often, to be able to output, great value out of data, you also have to be able to trust that data. And so the idea of having high data quality, having high data completeness and freshness, and the ability to kind of trace the lineage of the data from its creation, if you will, all the way to its activation is a really important capability.
And often it’s a second thought for organisations that are building this muscle. And so we’re seeing organisations accelerate their migration to a really data driven organization by starting with governance, rather than just starting with data.
And then finally, it’s about activation, right?
I mean, the world that we live in here, and we’re not here to just create the dashboards, we’re here to provide insights, just like you said, in the most simple, the most personalised way, in the most accessible way.
So people can really take action on our data. And I think the best example of any data product that I know of is Google.com. Every organisation out there wants to be able to build products, just like we built Google.com.
And so I think that’s where we are fairly unique. because we relate to the issue that the enterprise CIO has today, how do I take this massive amount of data complex, fast growing, often disconnected into a simple interface, where my people can just get the service that they want, so they can move on to the task they’re trying to accomplish?
Alex Zaharov-Reutt 5:15
Well, you just mentioned the CIO. iTWire is a business oriented publication, although we do cover consumer gadgets too, but what are the challenges, in more detail, that CIOs face today, when making sense of all this data?
I mean, for years, people have spoken about data overload, but that was just on a personal level.
For organisations data is coming from everywhere, their storage requirements going ballistic, Google itself, you must have the biggest storage on the planet, you’re at least competing for it with Amazon’s and all the rest. But there’s a lot of players out there. What do you say sets Google apart? And what are those challenges that CIOs face?
Bruno Aziza 5:50
I’d say probably the biggest challenge that organisations have today is that the approach of the past just doesn’t match the opportunity they have in the future – most organisations have had to witness on-prem constructs.
And what I mean by that is this idea of having to think about the capacity you need to load and having to think about the compute, you need to provision and so forth. And a lot of the cloud vendors today, it kind of take this concept of on-prem thinking and make it available in the cloud.
I think our value proposition is really around where the most modern, most complete data analytics platform that you can think of, because we built it for the future.
And right, so in our world, we think about simplicity, we think about a product like BigQuery, for instance, people think about it as a cloud data warehouse, it’s way more than that. It’s think about an analytic system, that you can throw any data at structured, unstructured, semi structured, and it just works.
And so I think that’s where we’re uniquely positioned. Because we have been where these customers are today, we’ve had to build these products that have to scale for billions of people that have to scale for unpredictable workloads. And so that’s a really unique position, we can truly relate with the issue that customers have, and for having been here now, over two years, it’s really quite amazing the type of innovation that we see customers take by thinking about the future way differently.
You don’t have to think about VMs, you don’t have to think about shrinking the data, you don’t have to think about the world of the specialists, you have to think about, look, I’ve got data and need to put my hands on as much data as I can, I need to be able to trust it at very large scale, and they need to give access to people in a way that they can consume it in a very personal manner.
And when I think about BI (business intelligence), for instance, often people think about the BI workloads is hundreds and 1000s of people, we think about millions of people. And so that’s really kind of the differentiation is today, the sad truth in the data analytics world is that 30% or so of data analytics solutions are adopted by employee.
So you’re kind of missing 70% of the organisation. And you’re missing that engagement that is critical to making you a valuable organization. And we’re really focused on closing that gap.
Alex Zaharov-Reutt 7:58
Yes, well, I’m thinking of a couple of sayings, and I’m probably mangling them, but one of them was that yesterday’s solutions are not going to fix tomorrow’s problems. And the thinking that got us here is not necessarily the thinking that’s going to get us there.
I mean, sure, there’s the fundamentals that will stay in place. But you got to expand your mind and expand what you think is possible. And often solutions are turning things on their head.
Galileo thought that the Earth revolved around the sun. And he was right, it turned orthodoxy on its head. But look, we’ve just had this multi year pandemic. And while we did have great technological acceleration, sporadic lockdowns still happen in China.
Today, we hear about it affecting tech, manufacturing and logistics. So what have been the biggest shifts in the big data landscape since the pandemic? And what learnings can we take from away from that to inform business strategy in 2023 and beyond?
Bruno Aziza 8:48
I think the first change is it has accelerated the level of awareness for all organisations, that data is really the way for them to innovate and think about 68% of organisations can get value out of their data today.
And so while most of the industry was aware of it, what we found is that this crisis really accelerated the need, okay, if I want to maintain a relationship with my customers, I can’t do it through a physical location anymore, right. So if your retail financial services, any of those industries, it really just puts this complexity right in front of you.
And it also has changed, I think, how IT leaders think about their role inside an organization. Previously, they might have thought about, their role as technology as a cost centre. And now they’re really seeing themselves as now I’m creating solutions that are going to accelerate the transformation of my organization.
So the biggest switch that we see is how CIOs organise their teams. So they can provide data products to their business counterparts. And it could be internal stakeholders, it could be external stakeholders.
And so we’re really seeing this trend of people moving from building a data ocean for their data, access to any data, internal, external, any type of data in real time to data mash, which has the ability to work unite yourself and provide innovation at the, at the edge, if you will, for your innovation, all the way to building data products that are solution business purpose solutions, so people can activate the data, and then make decisions faster.
And so in a way, it has accelerated this transformation for us, I mean, we have the opportunity to work with some of the most innovative organisations in the world. So, we’ve learned a lot from what they’re doing. And so it’s both a technology change, but also a mindset change. And I think it’s great news, if you’re a CIO or chief data officer, today, you’re playing a very, very important role to help your transition get to the next phase of its evolution.
Alex Zaharov-Reutt 10:36
I’ve come across a couple of companies that turn everything on their head by saying that they don’t want data lakes or data oceans, where they copy everything into the data lake, and things get lost.
And what they’re able to do is find data wherever it is, and it can be anywhere, and then they can structure that and put it into a much more easily accessible business tool. So, is that an approach that you’re also taking?
Bruno Aziza 11:05
I think what’s interesting about the term that data lake is, and we hear this from customers, rather than the terms that I used it ocean and national data product, or not Google terms, there are terms that our customers use to describe their environment.
And I’ll go back to the CIO of Vodafone. He said the data lake is comfortable, because it’s landlocked, you see the end of the data. But typically, it’s not very representative of the actual picture of your data – we designed for instance, our products to be multi-cloud by default.
So we assume that the data you’re going to need to use is going to be of course in the GCP cloud, but it’s also going to be in the Azure cloud and the Amazon cloud. And so we want to provide the most open data clouds, so you as an executive, you can reach to where the data is. So you can make the right decision.
So we have this great product called the query Omni, that allows us to send the processing, where the data is very differentiated to competitors, who might want you to move the data to their cloud, or might want to replicate the same environment across multiple clouds.
And so trying to make that as simple as possible is really important for us, this idea of simplicity, in a very complex world, is what I think we’re pretty good at.
Also the fact that we have very good capabilities around artificial intelligence, which is a great way for organisations to take their data and turn them into something differentiated, think about fraud, analytics, anomaly detection, propensity to buy. Those are great outputs and outcomes that organisations are trying to look for, when they partner with us.
And so any industry is now looking to see all the data, trust all the data and activate the data in as real time as they can. And that’s what’s going to make them different from their competition. So many of the leaders come to partner with us to co-innovate in that space.
Alex Zaharov-Reutt 12:51
So with all of the experiences that you’ve had across multiple companies, as you described before, and with all that you do now, what do businesses need to consider when building their modern data stack in 2023, even though that we know, of course, you’d prefer to use businesses to use Google, but whether they they’re using Google’s solutions or not?
Bruno Aziza 13:11
I think there’s a few key constructs, right, I think the first one, and we talked a little bit about it, this idea of of mindset, and right, so you do not want to dumb down your data stack for answering the complex questions your people have.
So you want to have a sophisticated stack to answer the complex questions. But it doesn’t mean that you have to dumb down the environment and simplify it down. So you really want to partner with an organization that has really great depth, and the ability to handle data at scale, and the ability to use artificial intelligence to kind of power this data.
And that’s probably the first one because we see a lot of customers saying, Well, maybe I’m just gonna start with something really simple. And then quickly, what they realise is they’re limiting the potential of their employees, right? So you want to be able to create this environment that’s flexible, that’s limitless. So you can engage any data, any workload, and any people really across their existence. That’s the thing. That’s a really important construct.
The second one is how you organise your team, and what is the percentage of people that work in data inside your organization. So if you take the organization like Wayfare, for instance, in the US 18% of Wayfarer employees are data people.
And so that says a lot about the culture of the organization where they think about data as a central component of the growth of the organization. So that’s really the second concept. And then I think probably the third one is to realise that the majority of the data that you need, is actually outside your walls.
And so being able to tap into solutions, like what we have with analytics hub, or solutions around spark that allow you to take advantage of the the innovation in the ecosystem is a really important one because you don’t want to just live in your own bubble that what’s happening in the data world today is that it’s the result of a large ecosystem.
So for us, when it comes to cataloging solutions for us together great partners like Collibra, when it comes to business intelligence, we have great partners like Tableau, right?
So we want to embrace the ecosystem so everybody can grow inside in the organisation, or it’s an industry that is 25 – 30 years old. And so you’re not getting to build a data stack from scratch, you’re building it in conjunction to the investments you’ve made in the past. So it’s important to embrace what’s in existence in the market.
Alex Zaharov-Reutt 15:17
Now, one question that’s been running in my head, whilst I’ve been listening to explaining all the the answers to all the different questions, you talked about how 70% of the people in companies are still not with the program, not using it.
And one of the things I love to do is show people all the cool advanced features of their smartphone, even simple stuff that they don’t know.
I mean, it’s incredible how many people don’t use simple voice dictation, which is now offline on devices, it doesn’t even have to go into the cloud anymore to to get that response.
So what are you doing to help companies to encourage the other 70% of users, the bulk of their users, to get on board with the new program?
Bruno Aziza 15:53
Well, there are a few areas where you really can take advantage of of the Google Data analytics platform to really kind of scale to the next level. The first one is this idea of convergence of workloads.
So today, the world thinks about doing analytics, primarily using SQL, but what we’re working on is, and we’ve announced that next is its ability to bring in new workloads within BigQuery.
A great example is Spark, for instance. So the ability to run Spark workloads, right from within BigQuery is a big differentiator, because setting up a spark environment today can be intense, you have to set up new infrastructure, you gotta move data, and so forth.
And you got to tap into a different skill set, we’re doing the same thing from machine learning. So BigQuery machine learning is SQL we’ve created to tap into machine learning models, very adopted by our customers, and really kind of enabling any business analysts to start building models, right?
I mean, the game here is that if you can reduce the cost of experimentation, you can now get a lot of innovation across your employee base. And so that’s probably the first areas being able to work with data, any data from just sequel or using Spark or using machine learning. Also, the data types that we support have increased, right, you think about analytics, typically with a structured datasets, but we just announced support for unstructured datasets.
So now you can run machine learning on images. So you can imagine doing prediction on just images, which we think is really a game changer for a lot of organisations. So that’s on the how I work with data, and how I activate the data.
You might not know this, but today, we have over 10 million monthly active users on our product called Looker studio, which is our business intelligence self service solution. And so we’re really trying to make sure that these capabilities are effectively available to as many people as we can, because the the idea here is that, for us would be a partner of your innovation, we have to enable as many people as we can on as much data as available, and on as many workloads as they’re comfortable with.
And so I think it’s a combination of the three things, right, limitless data, limitless workload, limitless reach, that’s going to enable organisations to kind of break through some of the limitations they’ve had in the past.
If you look back in history, we’ve been pretty restrictive and did access and did activation, typically you rely on a small team of data scientists or small team of business analysts, we want to make that relationship with data a lot more liquid. So you can innovate at a much faster pace.
Alex Zaharov-Reutt 18:16
Do you also run like customer workshops, where you take them through all these capabilities and show them and say, right, you need to get your staff to be using all these cool tools that we have? It’s so easy. Do you actively work with companies to show them so that there’s no excuses?
Bruno Aziza 18:33
Absolutely. So there’s two ways that customers kind of interact with us.
The first one is – what you might call product lead growth, where they’re discovering about the product, and they’ve got access to the product.
In fact recently, at the Next event, we announced the availability of connected sheets, which is, essentially think about the future of spreadsheets – connected.
It’s a web experience, it’s connected to BigQuery. In live mode, billions of rows of data can be accessed by financial analysts, that functionality is available.
Now, in Workspace, it used to be available in lonely workspace enterprise, we’re making it very easy for people to get access to our tooling.
So if you have a personal Gmail account today, you should try connected sheets, and you can try it on BigQuery sandbox, which is you don’t need a credit card, and you can do massive queries that you couldn’t do before.
So I think that’s probably the first level of engagement, you’re really massive globally, people are really working with the product, and we let the products talk for itself – we don’t need to do a lot of marketing for that, because people know about our innovation.
Alex Zaharov-Reutt:
Word of mouth marketing?
Bruno Aziza:
And then, I mean, that’s the best type of marketing, right? It’s just that the product should talk for itself.
And then of course, we have a strong enterprise engagement and where we really try to think about innovation with customers. You look at some of the the large organisations that we work with, we’re really trying to get them to the next level.
What is the product that you’re trying to build what what is what you couldn’t do before? Is that our approach our DNA, our engineering mindset can help you really break through problems that you couldn’t do before.
So really, I want to say that the art of possible, but it’s probably the science of what’s possible if I could say it that way.
Because there’s a huge potential, especially when you change your mindset of starting to think about anything’s possible, you think about what we just announced: Analytics hub is a fantastic data sharing platform.
You and I can collaborate on building data analytics products that we can put available on an exchange and a highly secured manner, and we can secure it, and so we can turn it on, turn it off, we’re really bringing to market capabilities that just have not been available previously, in that way.
And so I’m excited about the next five and 10 years, because I think we’ll see a lot of great innovation for customers. These customers have been waiting for someone that understands the potential that they have, and they have a lot of great ideas that just have been constrained by the old thinking of legacy of purchase, not just legacy technology, but also legacy thinking, I think, of what can be done with data.
Alex Zaharov-Reutt 21:08
I think a lot of times when things are turned on its head, it’s like we go from a mentality of lack, and things in short supply, and we’ve moved to a mentality of abundance.
Think of cellphone plans, they used to be by the minute, now it’s unlimited. Same with data, which used to be limited. And of course, with mobiles, they have the artificial cap, and they slow you down. But nevertheless, it’s like even in Star Trek, I mean, they’re in a post scarcity economy.
And you’ve got replicated great things you can make whatever you want, there’s just – we have to have that shift, we need the technological shift to – well, you can wish for it, but you need the technology to make it happen.
But we’ve never been more sophisticated and mature in that area. Now, you mentioned a couple of things from Google. Next, which you held last month. Is there anything else you want to mention? That was launched at Google Next?
And also, I can see right in front of you, or behind you, it says Google Cloud Next, leaders connect Sydney. So tell us a bit about that event, as well. But is there anything else you wanted to mention about the Next event that was on last month?
Bruno Aziza 22:00
Absolutely. So we announced lots and lots of products, of course, it’s been interesting, since I joined, originally, we announced most of our products at Google Next. But over the last few years, we’ve created multiple moments for the community to engage with us.
So we created the Data Cloud Summit, we created the data engineering Summit. And now we’re doing next not just the global next, but we’re doing local events. And the whole idea there is, to one communicate clearly the context around these launches, but also connect with the community so we can get their feedback.
What’s important to us is that we’re building these products and these solutions for you. And so we want to hear globally as much as we can. So we can kind of turn around and then provide great capabilities for you.
And a great example is big lake, which is a product, a lake house product, if you will, that brings kind of workloads together. So data warehousing and like, workloads together, we announced in Preview about three months ago, and we got such a great response for the market, and great feedback that we’re able to turn that from preview into GA in just three months.
So this idea of communicating and getting the feedback directly from customers and very large scale also allows us to focus on what’s important to them. So we can deliver as much innovation as fast as possible. The other areas around data plaques, which often it’s an area of governance that people don’t talk about a lot.
But it’s really critical to make sure that the data that you have is data that your people can trust. And so we’re announcing a lot of functionality there around data quality, around lineage run, bringing Catalog under the umbrella of data Plex, so I’m really excited about that one.
And then finally, business intelligence is a space that I’ve been in for a very long time and where we had acquired Looker about a few years ago.
And now we are bringing Looker and Data Studio together in the suite called Looker studio for self service. And then of course, the great Looker capabilities around look ml because we see that the way the world is evolving is the CIOs are building these data products where they can interface my goal from a dashboard.
And my goal from a call into an API and my goal from an embedded custom application they’ve built so we want to be their partner along the way here where they can get access to gigantic amount of data that is relevant to their people that they can trust. So they can activate an any interface that they choose. Right.
And I think one of the constraints we’ve seen the past in this industry is really kind of getting people stuck in small data, and dashboards.
And the opportunity turns out as in all the data across any interface, including the dashboards, and so those are some of the key innovations I think about.
Alex Zaharov-Reutt 24:53
We are running out of time. And as it turns out, I’m in San Francisco, you’re in Sydney, but you’ll be back here and also we would love to see to continue this should continue, like do a part two of this one, because I know you have a hard stop very soon. But tell us about your Data Journeys podcast, some of the highlights and where people can subscribe to it.
Bruno Aziza 25:12
Absolutely. So like I said at the beginning – we do all this for customers. And so we’re trying to find ways to scale the connection between the customers, typically the way enterprise software vendors, connect customers, you have to go through them.
And so when I joined, I really thought that because we have so many customers that are innovating in such an amazing scale, I couldn’t be in the way of that connection. So we created this program called Data journey.
It’s every Tuesday or every other Tuesday, we publish a video that’s an interview just like this. So it’s not scripted.
Alex Zaharov-Reutt 25:50
A fireside chat!
Bruno Aziza 25:52
That’s exactly right. So imagine that I’m just curious about how you as a chief data officer, or, or head of data analytics, how did you mature and typically,
I ask them three questions. Tell me about your company. Tell me about your use cases. And tell me about your do’s and don’ts. What did you learn.
And by doing that, I’m achieving two things.
One, I’m frankly, I’m learning myself from these customers, I am giving a platform for my customers to make sure that there are visible in the community and people know about their work.
But what’s been amazing is that now other customers reach out to the folks that have been on the program.
And so it just keeps on giving because now we’re building this community at a very large scale, you’ve got people like the head of data at Mercado Libre, in Argentina, talking to the chief data officer at WPP.
In England, they would have never been able to connect otherwise. And now I’m able to connect these folks and don’t have to go through me anymore. And so anyone that’s listening to us first, I mean, should definitely follow this channel, but if they want to be on it, they should reach out to us, because it’s a very lightweight process to get on the program. But the value and the impact that it has to the community has been really, really great for us.
Alex Zaharov-Reutt 27:03
Yeah, wonderful. Well Bruno Aziza, head of data and analytics at Google Cloud, thank you very much for your time and best of luck and whilst you’re in Sydney!
Bruno Aziza 27:13
Thank you so much, Alex. Thank you. Bye bye.
Here’s the “Big Lake in a Minute” video from Google Cloud Tech, more information here.
[ad_2]
Source link