Modernizing the Right Way with 451 Research

Modernizing the Right Way with 451 Research

Database modernization

451 Research’s Voice of the Enterprise, Data & Analytics, Data Platforms 2022 survey found that only 28% of enterprises reported that nearly all strategic decisions are data-driven. 38% reported that most decisions are data-driven, while 34% reported some or few decisions are data-driven.

If enterprises value data, why are they not using it to make strategic decisions?

Yellowbrick Data’s Umair Waheed talks with 451 Research (S&P Global) analyst James Curtis to explore the core principles of database modernization efforts and the critical factors organizations should consider on their modernization journey.


Jim Curtis (00:04):

All right, welcome everybody to today’s webcast. I’m here with my esteemed colleague, Umair. We’re going to talk today a little bit about modernizing the right way. Just before we do that, I just want to introduce who you’ll be speaking with today. My name is Jim Curtis. I’m a Senior Analyst at 451 Research. We’re also part of S&P Global, so as far as my coverage, I cover all things data platforms and databases, and have been doing that for a little less than a decade now. With that, I’ll turn it over to Umair to introduce himself.

Umair Waheed (00:35):

Thanks, Jim. Umair. I’ve spent the last 20-odd years in data and analytics, working both from a consulting perspective directly with customers, but lately, in the last 10 years, working with vendors, so worked with Microsoft, on their data analytics portfolio, and lately with Yellowbrick as heading up Product Marketing.

Database Modernization Efforts

Jim (00:56):

Great. Thank you very much. With that, we’ll sort of jump into today’s topic. I want to sort of mention, at least from my perspective, I talk to a lot of people, and I’m sure Umair does as well, of industries and organizations that certainly want to sort of deploy sort of digital transformation efforts or modernization efforts. These are really sort of, at least in some ways, catchphrases or catchwords within the industry. But want to talk today, we’ll share, sort of we’ll keep it relatively high level, but we want to talk today about some of the core principles and some of the things that industries might want to sort of think about, or particularly enterprises want to think about.

Jim (01:33):

Today, we’re going to talk a little bit about sort of some of the data strategies that sort of go on. We’ll talk about some of the systems that enterprises need to talk about, as well as some of the places of which those systems might want to be deployed. So today, we’re going to say let’s transition to the right way, or at least transition in the right way. That work for you, Umair?

Umair (01:54):

That sounds great.

The Right Data in the Right Hands

Jim (01:55):

Great, thank you. Okay. Let’s jump into it for a little bit. We recently conducted a survey at 451, and we asked enterprises, we said, “Listen, do you value data?” And not surprisingly, most enterprises do value data. Then we said, “In two years, are you going to continue to value data more than you do now?” And surprisingly, yes, they do. So no surprise there. But what’s interesting, Umair, is that when we asked enterprises a question about like, “Are you actually using data?” It doesn’t quite come back with the same numbers, right? So why do you think that is? Why do you think there’s sort of people value data, but yet in reality, the action isn’t quite sort of matching up there?

Umair (02:42):

That’s a great question, Jim. I think partly, it comes down to kind of history, and where we’ve come from. We’ve spent a lot of time building systems with heavy lift, to service that sort of executive dashboard, that executive-level reporting, but now I think the real big difference in some of these numbers, and as you said, it’s no big surprise that people value data. It’s great because genuinely, I believe that data is, and will be, and can be transformative for a lot of organizations, and for society in general.

Umair (03:14):

But where we’ve kind of missed the mark is on enabling people to go at that data freely, without fear, so that ability to be able to say, “I’m going to be able to ask that question at any given point in time, and I’m going to do it my way. I’m not going to ask an industry expert to go build that query and run it. I just want to use the tool that’s at my fingertips to go ask the question that I want to ask at the time I want to ask it, and I want answers back pretty much in the instant, right?” We’re all addicted to instant gratification now. We don’t want to wait forever.

Jim (03:50):

Yeah, so I mean, it’s interesting. We’ve had these phrases of like, if you have too few individuals accessing it, you’re going to have limited insights, right? And certainly, sort of if you open that up, you’re going to have broader insights. We’re certainly not suggesting give everybody all types of data. There’s certainly sort of a strategy around that, but at least my experience, and I think maybe yours as well, is that the reason people can’t get at data isn’t because… It tends to be bureaucratic reasons or something beyond sort of, “You should normally have this,” okay?

Umair (04:29):


Jim (04:30):

So with that, I mean certainly we want these numbers to rise up, but what we certainly want is as enterprises value data, we expect the use of that data, particularly say from an analytics standpoint, to sort of go up. Before we move on to sort of some of the systems where this data resides, I think it’s important to sort of recap a little bit about what, why data is important. I think as Umair elegantly pointed out, it’s this access thing that’s actually very, very important.

So certainly, if you cap that at sort of fewer access points, or sort of access to individuals, you’re going to sort of limit your insights. But certainly, opening it up to everybody is also maybe not the correct solution as well, right? There certainly has to be restraints. But generally speaking, its access actually becomes a challenge, where people need to have access to data, but simply just simply can’t get it. Is that what you’re finding too, Umair? Is just some of these constraints?

Umair (05:28):

I think so, Jim. I think what’s important is that we don’t know what we don’t know, so until we give the right people access to the data, we don’t know exactly what value they’re going to find from it. That innovation really comes when people start to merge the data and the trends that they’re seeing with the ability to impact their processes, their products, their customer experience through access to it, through data and what they can start to change about those processes from the data they’re seeing, how they can react, how they can deliver enhanced customer experience, and they can do it, and essentially generate more revenue, or more savings, or whatever it is their mission is, through that data.

Data Lakes

Jim (06:12):

That’s great. It’s hard to talk about data and sort of access to data without sort of having sort of a systems, or tools, or environment discussion, and certainly, all these play in place. So we’re going to talk a little bit about some of these systems that, at least from our research, suggests that many enterprises have in play, and Umair certainly has seen these sort of in his experience as well. Let’s first touch on data lakes for a minute. Data lakes are a little more than a decade old, and data lakes…

Jim (06:42):

You know, it’s interesting. They sort of emerged on the market to really address some of the challenges that were sort of there, and one is that if you go back maybe 15, maybe 20 years ago, there’s probably a lot of data that enterprises had, that simply, they didn’t want to collect, or didn’t think were really useful. I mean, is that a fair statement? Just people saying, “I don’t think we need that data,” right?

Umair (07:05):

Yeah, exactly. A lot of data was thrown away. It’s gone forever, so there’s no way to get it back, and I think that one of the useful… The things that have come out of the big data era has been this concept of data lake, the ability for us to centralize and build these repositories of data that are kind of an ounce of history for our organization. If we were historians or organization historians, it would be a great repository of just almost like a big museum or a big library.

Jim (07:34):

Yeah, so this… I mean, it wasn’t bad. I mean, but certainly, we went from throwing data away to then collecting everything, right? And maybe that’s not the solution either. But this idea is that I can certainly collect more than what I was having before, and I can collect that in sort of a low-cost method, was sort a great appeal to data lakes, and it sort of addressed that.

So it also said, “I can take all kinds of data. I can take text, and video, and files, and all this kind of stuff,” right? And maybe there’s something in there, right? But you know, the panacea of that dream was great, but Umair, did that… I don’t know if it certainly delivered on that, so…

Umair (08:17):

Yeah, no. I think you’re spot on. Like, there’s lots of different types of data in an organization, and the ability to incorporate images, video, x-ray images, lidar, all these kind of great things, and scans, geophysical surveys, and satellite imagery. I was speaking to a provider of satellite imagery the other day, and lakes are great for storing that type of data. It’s interesting who needs access to that data, and how many times it’s accessed becomes quite interesting.

Umair (08:48):

And I think the other thing I find when talking to customers about data lakes is that they struggle a little bit with how to organize data in the data lake for maximum effectiveness, and you’re kind of highlighting there on the slide, security and privacy being interesting conversations in the data lake. You know, what is the right security model, and is it right that the school model is baked into the data lake, or is there a higher tier service, a data protection service, that layers on top with technologies like attribute-based access control, for example?

Data Lakehouses

Jim (09:19):

Yeah, absolutely. I mean, these are just some of the things. I mean, we’ve certainly heard data swamp and other things like that, but as enterprises sort of moved on from data lakes, realizing that, “Oh, now we have all this data. We certainly have, now, now we have data management,” right? This thing lives, this data lives, and what are we going to do with it?

So really, when we see sort of the evolution of sort of the data lakehouse, data lakehouse, certainly the term has been in play in the industry for a little bit, but I’m sort of putting it maybe with inside two years, something like maybe give or take, is when it’s sort of come along in the market as sort of a formalized deployment. And the way the data lakehouses are sort of… I’m going to sort of apply it sort of generally, is that it’s, so we’re going to take the best of a data lake and the best of a data warehouse, and we’re going to sort of apply this.

Jim (10:12):

And certainly, it’s addressed sort of some of those. It will certainly continue to say, “Hey, it’s going to be maybe based on commodity hardware, or object storage in the cloud,” or things like that. It becomes sort of a low-cost or very cost-effective way to sort of store data.

But one of the things lakehouses did is it brought sort of a schema and sort of a structure to the data, so you can get consistent data transactions, or data that’s much cleaner, and in some ways maybe accurate in some ways. But you know, it’s still relatively new. I mean, is that what you’re finding, Umair, just in terms of some of these lakehouse –

Umair (10:54):


Jim (10:56):

… deployment?

Umair (10:56):

Absolutely. I think what lakehouses added were things that are difficult to do in file-based storage platforms, like update data. From a European GDPR perspective, you know, the ability to delete data, or correct data, or update data is important, and it’s very difficult to do that in a purely file-based storage. So some of the new technologies around data lakehouse have addressed some of that, and enable you to make some changes to data, and persist those, and actually set, apply schema. I still don’t think they address the enterprise need.

I think where people are still struggling with the lakehouse concept is being able to deliver a level of service to the business. And we talked about these impatient users up front, like people have these expectations on, “I’m going to click a button, and I want the answer back quickly.” And whilst data lakehouses are a step forward in terms of providing a capability for people, data scientists or other, to get access to that data, in terms of a general purpose solution to really help the business exploit the value of that data, and broaden it to a larger audience, that’s where we feel it kind of starts to tail off in terms of its value.

Umair (12:07):

And from a biased perspective, what they’re starting to do is build database services on top of the data lake, so yeah, I mean that’s… it’s an interesting concept of whether you go for a database or you try and retrofit a database concept on top of a data lake.

Modernization Journey: Data Warehouses

Jim (12:25):

Yeah. I mean, certainly there’s going to be sort of the right tool for the right job. I mean, our research suggests that all of these things are in play in the enterprises, and for good reasons, right? I mean, and oftentimes, it comes back to what is the purpose, right? What is the purpose of these things? As we continue on sort of this systems journey, right? Particularly when we talk about sort of using data for analytics-based environments, and decision-making, and things like that, and sort of getting that data out, we’ve got to talk about data warehouses.

Jim (13:00):

Data warehouses have been around for multiple decades, right? And they’ve really actually served well, and their longevity is actually tip of the hat to how they have served enterprises. And they’ve been very highly performing, you know? And as mentioned, there’s schema that need to be applied to sort of codify that data, and because that schema is applied, I mean, Umair, you can certainly… That leads into performance, and accuracy, and all these things in terms of analytics, right? Just in terms of doing that. I mean, is that what you’re seeing? And certainly, we can talk about sort of how those are sort of deployed, I guess if you will, as well.

Umair (13:53):

Yeah, no. So you’re spot on, right? You could have put data warehouse at the top of that list and at the bottom of the list, you know? They’re the precursor, and we’ve kind of come full circle back to data warehouse. But I think what’s changed is that we no longer have this idealistic vision of a single place, a single data warehouse, a single repository where everything has to live.

I think we are becoming more mature in our data conversations, and accepting that data is actually dispersed across an organization, sometimes for good reasons, sometimes because from a compliance or contractual perspective, it needs to be in a particular location, or on-premises, and sometimes because of just M&A type of activity. We might acquire a company that has a data repository in one cloud, and our corporate repository was standardized on a different one.

Umair (14:38):

Do we then go and bring all of that together and force ourselves to go through that same pain that we promoted back a decade ago, where we’re saying, “Everything has to be in one place?” Or do we embrace that, and actually say, “We need to accept that, and work across,” and accept that’s the new normal, is to have data dispersed across an organization, across clouds, across on-premises. Regardless of where it lives, you need to be able to get access to it.

Umair (15:01):

But to your point about data warehouses giving the business that capability to really go at the data in a performant way, with some expectation of performance, some expectation of not just schema, but also model being applied to the data, so all the good things that we do around constraints, and foreign keys, and all that sort of stuff, and building out those structures that help the business understand how that data should be used are still important. None of that has gone away.

Jim (15:35):

Yeah, that was a great discussion. You were getting a little ahead of me there on location, and I think that’s a great transition to sort of jump into that. But before we get quite to environment, and sort of the how I guess, if you will, I certainly come across a lot of individuals that… or at least a lot of vendors and enterprises that sort of are in this vision of sort of this general purpose database, right? And certainly, there are ways in which you see some of that marketing out there in sort of the database sphere I guess, if you will.

But data warehouses probably isn’t going to be sort of this general purpose application. It tends to be highly tuned for specific purposes. And as we mentioned before, you sort of want to use that for sort of your data work. So, I think that leads into that. I don’t know if you have any thoughts on that, Umair, is that these tend to be highly sort of focused systems, I guess if you will, not that they can’t do other workloads, but they’re certainly focused in …

Umair (16:44):

Yeah, focused from the perspective of being the SQL-based query interface to enterprise data, I would say yes. I would say the data lake, it lends itself to doing lots of data engineering type work, things that you do once and kind of move on from. Going back to my satellite example from earlier, those images that you kind of download and process using all the latest and greatest technologies, you’re doing that probably once, or sometimes maybe once or twice, or you discover a new way of analyzing that data, so you run a new algorithm across those images again.

But what you’re really doing is extracting the structure out of those data, out of those images, and turning it into something that’s then used thousands of times or tens of thousands of times, and that’s when it makes sense to have those data in a query engine, so you know, a queryable format like a data warehouse, rather than a data lake I think, where people actually can go and experiment with the structured data that’s come out of that process.

The Right Location for the Right Tool

Jim (17:50):

Yeah, that’s great. Well, but let’s transition for a minute here, to we mentioned earlier… As we’ve sort of gone through sort of all of these sorts of systems, I want to sort of bring it back to more sort of a general awareness, okay? I wanted to give the impression, we certainly didn’t cherry-pick some of these things, that based on our research on a certain survey, is that there’s two things.

One is that enterprises actually still have, and want to have, data lakes as part of their environment, and they’re actually very useful, okay? They’ve been out there. People are finding it useful for them. But some of the challenges that we mentioned earlier are still sort of hanging around, right? And these are just sort of general awareness to enterprises to be aware of, on this sort of this digital journey, as the security, and privacy, and sort of how the data is managed, sort of how it’s playing within sort of other environments.

Jim (18:47):

I mean, certainly, at least I don’t see systems or environments sort of deployed in like pure isolation. It’s really how is X, Y, Z system going to be deployed in sort of an ecosystem of other sort of systems, and sort of playing nicely with that. And that’s certainly what we see. So, we’ll talk more about that, but as we sort of mention that, if it’s okay Umair, let’s jump into sort of this environment question that you sort of jumped in.

To say that cloud is on top of mind or a catch phrase would sort of be an understatement, is that our research certainly suggests that cloud is where people are sort of deploying in that sense. And what you’re seeing here is essentially sort of some research results that we asked hundreds of enterprises, “Where are you deploying your data platforms, your databases, and various systems?” And it turns out that they’re deploying in public cloud and different places.

Jim (19:55):

So just to point out, the way we sort of define this is public cloud and infrastructure would be, “I’m going to take an image. I’m going to sort of probably put it on one of the hyper-scalers, and I’m going to do some work, and they’re going to manage the infrastructure and things like that,” right? And who holds responsibility is sort of where this game is played. And then you certainly have private clouds, where I’m going to take it all in-house. Again, I’m going to do most of that, but it’s the cloud environment, and then the on-prem.

Jim (20:24):

The managed service database is actually kind of interesting, because that’s basically saying, “I’m going to… I want the benefits of sort of these analytic systems databases, but I would prefer that somebody else sort of manage most of that hard work, most of the updating, and most of the infrastructure,” right? I mean, from a cloud perspective, are you seeing this sort of this adoption of cloud, and certainly in some of these iterations as well?

Umair (20:57):

Yeah, and certainly, I’ve had the privilege of, with my time at Microsoft, seeing the cloud go from fringe technology all the way to now probably one of the most important technology advances, certainly of this century, if not of the last 20, 30 years. You know, people are looking for the cloud to help them from an agility perspective, to be able to react to change, but also from a simplification perspective, so not just cloud, but also looking for SaaS-like experiences in the cloud.

Data Strategy: On-Prem and Cloud

Umair (21:30):

So the trend is towards not even managing your own infrastructure on the cloud, so yes, you could take an image and land it in the cloud, but I think what most customers would want from the cloud is more of a managed experience, where they’re still in charge of their data, but they actually are getting the benefits of the cloud, and elasticity, and all those great things.

But I think, as you said, as we spoke earlier, location is important, and we shouldn’t be constrained by location. Because I’m adopting cloud, but I’ve got data on-premises, does that mean all my data goes to the cloud? How does that play with my corporate strategy?

Umair (22:09):

How does that play from a compliance and contractual standpoint? And also, what happens if, for whatever reason, there’s a global outage and I can’t get access to my data for some point in time? Am I able, then, to carry on delivering my service? Do I have to stop delivering my service? It’s particularly pertinent, I guess, in financial services and some of the kind of more regulated sectors, but yeah, I mean not being constrained, being able to run in all those places, I think is important. But as you said, the trend is towards more, I think, and ask for a managed experience, not having to…

Umair (22:47):

When we used to build data warehouses, they used to be a big, heavy lift, right? We used to have to optimize everything from the ground… You know, when we were deploying servers, it would be like which disks, how were the disks laid out, how many disks? We would over-provision disks, just to get the throughput that we wanted to the CPU.

And then we’d do a massive refresh and change because something changed in the architecture or something new came out. But we don’t want to have to do those massive lifts every time there’s a change, and that’s where the cloud can really provide an advantage, because they can come in to take some of that heavy lift away.

Jim (23:21):

Yeah, I think that’s a great point, because I think what cloud gave people, at least initially, is that it gave them options, right? It’s just like, before, the option was only on-premises. Like, that was the option. Now, there’s… And when you said cloud, we have to be careful. There’s cloud locations, right? I guess if you will, or different places. So it really opened up sort of that a lot of it, those discussions.

Jim (23:47):

So continuing on, we said listen, so if you’re going to deploy, where are you going to deploy your systems now, as I showed you, but where are you going to do it in two years? It’s basically a question, it’s like do you think it’s going to change in the future, all right? So not surprisingly, at least all of the cloud iterations on here have some sort of bump, right? Just in terms of percentage bump.

Now, if you’ll notice about the third one down, it’s on-premises, not on cloud. I certainly hear this talk. I certainly hear extremes, saying, “So many years, like no one’s going to be on-premises,” and research just simply doesn’t sort of support that, you know? But back to your point, Umair, is the fact that now that you have choice, you’re going to put the system where it sort of makes sense, okay?

Umair (24:39):

Of course.

Jim (24:40):

You know? And there are certainly reasons why you’d want to say on-prem, you know? We don’t necessarily get into those details, but it’s certainly not as binary as people think, right? Just in terms of doing that, and there certainly needs that. But I think it’s this, if you want to say a SaaS or this cloud experience, is what people are after. They really are.

Umair (25:01):

I think you’ve hit the nail on the head there. You know, cloud is not just a location. It’s also a way of working. It’s an operating model. It’s how we think about delivering infrastructure, and even before the public cloud arrived, we were thinking about converged infrastructures, and all of the sort of hypervisor and virtualization vendors were also delivering and helping us deliver cloud-like services in our own data centers, so that sort of mindset of being able to go and pull from a cloud of infrastructure, not have to stand up something specific for every use case, was interesting.

Umair (25:37):

But from a data warehousing perspective, a lot of organizations struggled, because those general purpose sort of infrastructures, and I think there’s a place in the cloud as well, are not always set up to deliver for a high-performance analytics, so you end up over-provisioning, or delivering inefficient systems to gain the benefit of cloud, but on the flip side, then you’re spending more money to deliver that, and there’s some interesting questions around sort of sustainability and the energy consumption of data, I think that play into some of the discussion, maybe for another time. But yeah, I think whether it’s cloud or on-premises, on-premises is not going away. Data needs to be…

Umair (26:22):

Data’s being produced, often, on-premises, in systems that live in factories and live in offices. They’re in office buildings, which are inside your network, and then you should be able to process that data close to where it’s being produced, and then maybe shipping the data that needs to go into the cloud into the cloud, and having that data available to locally, to be able to process when you need to.

Jim (26:48):

Yeah, there’s two. I think there’s a couple of things at least you brought up, that I think are important. One is, whether you want to call it mobility, but certainly, we see that because cloud is more of a framework, it’s transferable between potentially other clouds, okay? Or at least, that might adhere to some of the same principles. It’s never sort of a clean fit, but there’s still sort of a rebalancing that goes on. “Oh, I’m in the public cloud. Maybe I want to be on private cloud,” or things like that, doing that. That’s one aspect of it that I think just generally, cloud has afforded us to do, and tooling and environments certainly need to sort of enable that, right? From a customer standpoint.

Umair (27:31):

Yeah, 100%, and that’s one of the cool tenets of Yellowbrick, is being able to deploy that same solution across multiple clouds. One of the challenges with some of the cloud solutions that are out there at the moment, they sort of tie you to one cloud, and then what organizations tend to do is to create multiple sort of layers of abstraction around those systems, common service layers, common APIs, to try and harmonize all the different capabilities, and different performance, and different security models that these services offer.

Umair (28:00):

At Yellowbrick, certainly the way that we’re building and delivering our products is the ability to deploy everywhere, the same solution, so deploying a cloud-native solution in the cloud, but also on-premises, so our on-premises offering is also a cloud-native offering. It works in the same way in terms of the agility and the elasticity.

To the point that we were making earlier, it’s about getting into that cloud mindset, and being able to deliver agile data projects, on-premises and in the cloud, and not having to think about multiple different technologies just because you’re changing a location. It takes away a lot of the burden that large enterprises have, of worrying about being locked in, because they can be anywhere. They can be on-premises, in the cloud, and they can move things backwards and forwards as they need to as well.

Jim (28:48):

Yeah, that’s great. As we sort of close in on this, I think… And maybe we’ll sort of do some wrap-up here, Umair, just in terms of as we look back on sort of we’ve sort of covered data strategy, things to sort of consider, certainly around access. Certainly, we’ve covered sort of from a system and environment level, just in terms of some of the systems that are out there, particularly again from an analytics standpoint. And then sort of the environment considerations, to sort of think about.

You know, maybe Umair, I’ll put it over to you. Just, do you have any sort of closing thoughts here, just in terms of realizing that there’s probably people listening to this, talking about this? What are some key takeaways that I can then sort of move forward and act on?

Umair (29:45):

Yeah, I think for most people, you shouldn’t really hear about the data platform. Data platforms are really successful when you don’t hear about them, so when you’re just not thinking about it, and you’re just able to ask and get the answers that you want from your data platform in an easy way, from the tools that you use on a day-to-day basis, whether they be if you’re an analyst using a BI tool, whether you’re a finance analyst in Excel, whether you’re a data scientist or a data engineer writing Python or SaaS, or R-type scripts, it shouldn’t really matter where that data sits. It’s that fluidity of access that’s important.

Data Warehouse Technology

Umair (30:20):

But from an enterprise perspective, doing that without breaking the bank, without going crazy in terms of scale, is very challenging. And it’s not just the cost of that. There’s also the energy impact of deploying inefficient systems. So we’ve got to have that balance around delivering efficient systems on cloud and on-premise, being able to deliver in an agile way, using cloud mentality. So being able to be elastic and move things up and down in terms of reacting to the business, but also doing that in a cost-effective and efficient way. Which I think is where the Yellowbrick data warehouse technology really shines in terms of us really engineering our platform specifically for data analytics scenarios, data analytics use cases, rather than building on top of general purpose technologies. Which I think some of the things around data lake, and some of the early attempts at building data warehousing platforms, where they’ve kind of struggled, and they didn’t necessarily deliver that value, at least not at the enterprise scale that we need now.

Umair (31:24):

You know, now we’re seeing data at the petabyte scale. It’s a different order of magnitude. Now, we put this data in the data lake. Now it’s on us to go get value from that data. Otherwise, we’re just building a big white elephant.

Jim (31:39):

Yeah. Certainly thank you for those comments. I’m going to sort of wrap up some of my own thoughts as well. I’m going to sort of maybe a little more general, just simply because I’m going to sort of put my sort of my analyst on it.

I think sort of piggybacking on some of the things you mentioned, some of the things that I see, just from enterprises, in sort of their digital transformation efforts, and we certainly have research to suggest that there’s certainly a graveyard of failed projects out there, right? And it certainly shows up in our research. But some of the things is that enterprises certainly sort of… Either they have incorrect mandates or things like that, but some of their initial things aren’t necessarily sort of correctly aligned.

Jim (32:26):

And I’ll give you at least one example. One is, is that we’re going to go to the public cloud in the first five years, you know? And some of those can be quite limiting, right? In terms of like was that really the best for the business, right? Or things of this nature, so sort of thinking sort of long term, or I think you used the term agility, or flexibility, or some of these things, really thinking long term, of sort of how this is going to play out.

We mentioned that you may land somewhere, and then want to sort of pull back, or pull a little bit back, or you know? But focus on sort of the ability to move. Give yourself flexibility, right? These are things that we see a great deal in our data, is just customers thrive on or want agility. They want flexibility. They want the ability to sort of have options for themselves, you know?

Jim (33:15):

And then, sort of another point of that, just sort of talking generally, is that if the journey is the goal, you may have… Because it’s got to be operational, and you have to maintain the system, and things of that nature, and if that becomes sort of a burden on the organization, is that at the end of the day, you’re designing, and building, and re-architecting for an optimized state, okay? Or some sort of efficient state. So you know, it’s got to be enjoyable along the way.

At the end of the day, you’ve got to get the data you want. It’s got to serve your purposes, and if we go back, you want more decisions. You want more decisions based on data. You want to have access to that, and be able to do that. So, really sort of aligning some of those things in place up front, I think are really helpful.

Jim (34:06):

With that, Umair, thank you very much. It’s been a pleasure, and we’ll certainly be able to do this again. Thank you very much.

Umair (34:14):

Thanks, Jim.

Get the latest Yellowbrick News & Insights
Why Private Data Cloud?
This blog post sheds light on user experiences with Redshift,...
Data Brew: Redshift Realities & Yellowbrick Capabilities –...
This blog post sheds light on user experiences with Redshift,...
DBAs Face Up To Kubernetes
DBAs face new challenges with Kubernetes, adapting roles in database...
Book a Demo

Learn More About the Only Modern Data Warehouse for Hybrid Cloud

Run analytics 10 to 100x FASTER to achieve analytic insights that have never been possible.

Simpler to Manage
Configure, load and query billions of rows in minutes.

Shrink your data warehouse footprint by as much as 97% and save millions in operational and management costs.

Accessible Anywhere
Achieve high speed analytics in your data center or in any cloud.