Business Intelligence in the Cloud - A Spotlight Q&A with Dwight deVera of arcplan
by Ron Powell
Originally published October 10, 2011
BeyeNETWORK Spotlights focus on news, events and products in the business intelligence ecosystem that are poised to have a significant impact on the industry as a whole; on the enterprises that rely on business intelligence, analytics, performance management, data warehousing and/or data governance products to understand and act on the vital information that can be gleaned from their data; or on the providers of these mission-critical products.
Presented as a Q&A-style article, these interviews with leading voices in the industry including software vendors, end users and independent consultants are conducted by the BeyeNETWORK and present the behind-the-scene view that you won’t read in press releases.
This BeyeNETWORK spotlight features Ron Powell's interview with Dwight deVera, Senior Vice President at arcplan, Inc. Ron and Dwight talk about cloud-based business intelligence (BI) and how to determine what BI initiatives are suitable for the cloud.
Well Dwight, lately “the cloud” is everywhere – on billboards at the airport, on television commercials, on radio talk shows and, of course, in online and print technology media. We find that a lot of readers of the BeyeNETWORK and SearchDataManagement.com are looking for more information about the cloud and how it can help their organizations. The cloud has already gained broad adoption in CRM and document collaboration, and CIOs and IT managers are evaluating the benefits as well as the risks of moving other applications to the cloud. It's also obvious that cloud-based solutions are here to stay, but there are some inherent risks that need to be addressed especially for business intelligence. But before we get to the specifics for business intelligence, let's start by looking at the market for cloud-based solutions. Are the organizations that you're working with really going to be investing in cloud-based solutions and is there a real future for the cloud?
Dwight deVera: arcplan, as well as our customers and partners, are absolutely on board with cloud business intelligence (BI) and agree with some of the hype that's going on out there. The cloud has had great marketing from what we've seen. But there are real benefits that make this kind of a no-brainer from a technology perspective. These include the ability to truly access grid-based processing and the ability to have guaranteed availability of systems – systems that are maintained centrally. Another benefit is having fewer moving parts, less physical infrastructure, and less boxes and wires to worry about.
One of the simple facts is that even though there's a considerable amount of unemployment going on in the economy, the IT knowledge management workers or information systems workers are folks that really aren’t out of work. There truly is a shortage of technical people. By being able to leverage the cloud and having less infrastructure to worry about – not everyone needs their own business intelligence or infrastructure server – we can certainly reallocate some of those scarce resources a little better as well. So we're all on board with things moving into the cloud.
Well Dwight, we're seeing a lot of movement to the cloud as well, but from our perspective we're seeing a lot of private clouds, not so much the public cloud. But every time I turn around I hear about Amazon's public cloud. Based on that, where have cloud-based solutions seen the broadest adoption to date. Do you see more interest in private clouds or the public cloud?
Dwight deVera: We're absolutely seeing a level of mistrust of the Internet, but with BI in particular though, you're going to see things forced into the "public cloud" due to the proliferation of mobile business intelligence, mobile devices or what have you. If you're on the AT&T or Verizon network, you're in the public cloud. If you're dealing with information in the public domain, like Google Maps, you're in the public cloud.
Now, getting back to your question of what solutions have the broadest adoptions, let's look at the first movers into the cloud and their applications. We've seen CRM systems and document management systems. We've seen HR systems. Every time I talk to one of our customers, they're talking about other ancillary systems moving into the cloud. Those are those "first mover systems." But one common thread that we've seen relates to where the data comes from – end users, not transactional systems, unlike traditional BI. If you're working with a CRM system for a deployed sales force, the marketing and sales folks are actually creating that content. They're creating new customers, opportunities and prospects. They're downloading data from a cloud service or an internal service, and they're keying their data into the CRM system to manage opportunities. From a data volume perspective, that stuff is tiny compared to what we do in the BI and data warehousing world, and it's also managed, maintained, and created by these end users. If you look at document management solutions – Word documents, spreadsheets, things like that – it’s the same. That data is not created by transactional systems. We're seeing right now that systems that are not necessarily core to an organization, such as HR payroll, Salesforce, Sugar CRM, and document management, are very popular in the cloud. One of the common themes is really the size of the data and then where that data is coming from.
Now, the other systems that we see are systems that already are in the cloud that are providing data from the cloud. Basically, that's one of the jumping off points for getting started with cloud BI – pulling data from the cloud into your own internal infrastructure for analysis (like Google maps, which I mentioned earlier).
You mentioned a number of systems that are in the cloud and, obviously, they’ve been very successful. What do you see as the outlook for moving business intelligence to the cloud? In addition to the fact that most BI implementations are very large and very complicated, many companies have spent years developing and fine-tuning their BI systems and have invested heavily in their internal infrastructure to support their BI efforts. Why should they consider moving to the cloud?
Dwight deVera: Okay. Let me answer that in a few parts. The first part is why they should consider it. As a business intelligence and data warehousing practitioner, I’ve spent years in the data trenches, the ETL trenches or what have you. We're in the business of taking data and making more data out of it by taking transactions and building cubes and analysis applications. But one of the things that's absolutely true, even though we have terabyte-sized and now even exabyte-sized data warehouses, is that we don’t have all the information we need to get a 360-degree view of our customers. If my BI deployment has my sales transactions or sales history data, that's one part of it. But what about customers’ experiences with customer service or customer support? And what if that data for customer support and customer service is from an outsourced cloud-based provider? Additionally, what about competitive research on my customers and competitors? Many of those other data sources, like Hoovers or even something such asSalary.com, are in the cloud. So as we work to create that 360-degree view of customers and opportunities in sales or whatever we happen to be measuring, we can't help but see that there are pieces of that view missing. Many of those pieces that are missing are actually available now and are becoming more and more available from cloud-based services. They're coming from external parties.
So why would business intelligence move into the cloud? It's because today there is the requirement for more data than what is physically stored, maintained and managed with an organization’s internal business intelligence and data warehousing initiatives.
As business intelligence architects, providers, and software companies, one of the reasons we haven’t fully moved to the cloud yet is because we have lots of “stuff.” We take transactional data and actually turn it into more data. We're data multipliers. Then we start warehousing, storing, cleansing, doing all these things, and to take something like that and push it to the Internet is absolutely impractical right now.
One of the things that we've been seeing is the leveraging of cloud-based services – query web services, or data services – to act as what I call "cloud-based ETL." Replicating everything you have internally within your own firewall and pushing it to the cloud is not the way to be successful in the cloud. The way to be successful is to leverage cloud-based technology to securely access your data within your corporate firewall and then combine that data with the other external data that you need that's already in the cloud – CRM data, or customer support data, or even transactional data from, for example, an HR system – to create an integrated and world-class cloud-based BI solution.
There are many that look at the cloud and say security is a big issue. Do you see that concern starting to lessen as more and more people move to the cloud?
Dwight deVera: I think security is a natural knee-jerk first objection for many things, including the cloud. Is it secure? The press and the media don’t make it any better by talking about data breaches, or WikiLeaks, but security fears will subside because to do world-class cloud-based business intelligence, we use the same encryption technology and the same kind of point-to-point transmission technology that, for example, keeps online banking transactions secure.
When we actually deploy cloud-based business intelligence, we're using those same security protocols and those same encryption protocols. We're using a strategy that we've been talking about lately with a lot of our customers, which is a secure DMZ where we control the devices that are accessing the internal data sources. We're controlling the security access; we're controlling content management on mobile devices and in the cloud just as if users were actually sitting on the network and in the office. A lot of the security concerns we've heard have been related to the mobile business intelligence push. Users are coming in over their BlackBerrys, their iPhones, and iPads. And a lot of times those devices are not even owned by the corporation, but we need to securely deliver business intelligence to these devices through the cloud, and the way we do that is the same way we've been encrypting data with external trading partners through secure extranets.
Many times when you talk about security breaches, they're very low-tech things – someone leaves their laptop in the back of a cab or a disgruntled salesperson prints off my potential pipeline for the next two quarters or what have you. We're not really going to blame the cab driver and we're not going to blame the printer manufacturer. We do everything that we can from a technology perspective to keep everyone in a situation where they can sleep at night. We can encrypt everything. We can control everything. We can guarantee that if that salesperson gets terminated because they're kind of a bad egg that they'll never be able to access our systems again because we've followed all of these currently accepted security protocols that have been vetted and are currently being used for incredibly secure things that are much more valuable to people than, say, your sales report.
That's a good point, a very good answer on security. The other question is whether moving BI to the cloud increases costs? Is that a big expenditure or in the long run does it decrease costs? What's your take on costs?
Dwight deVera: Let's not talk about traditional costs from a technology perspective like software, hardware, infrastructure. Let's talk about why business intelligence is really in the mix and why it's important to many CIOs. What's the cost of a bad decision? What's the cost of making a decision if you actually didn't have the information or the proper level of information you needed to make that decision? One of the points that I've been talking about with a lot of our customers is that there is absolutely a shortage of competent business intelligence architects and business intelligence professionals. I joke around all the time with my team that there's zero unemployment in BI because there just are not enough of us. There's an incredible demand across any organization.
By having systems that are more centrally managed and controlled in the cloud, business intelligence professionals are able to work more closely with the business side of the analytics. They can really focus on the math and true business intelligence rather than the infrastructure of BI.
Now, I do want to make a point because we've actually analyzed this. Is moving to a cloud-based BI provider cheaper than an on-premises BI provider? The funny thing is when we've analyzed the data, the answer is no. If you graph the cost lines, they actually cross each other about 2½ to three years into your deployment. It's really the difference of choosing to have an on-premises solution where you're buying software, paying maintenance, installing servers and things like that and upfront costs versus paying monthly to a cloud services provider. But from a traditional cost perspective, it's not any cheaper or more expensive; it's really just how you're paying for things – up front or as you go.
Now one of the things is a massive benefit of using cloud-based business intelligence is that I can guarantee you that your time-to-market will be faster when you leverage a cloud-based service. It comes back to my initial point that making a bad decision is terrible and making a decision without the proper information is terrible. Those are the real lost opportunity costs of not doing this well. But making a better decision faster is absolutely more valuable, and that's really where there's a benefit in true BI based on ROI analysis of BI deployments.
So if I look at what you're saying, it really comes down to BI agility. I mean for those who are looking at implementing business intelligence, it can be done faster by deploying in the cloud, but in the long run, it's not a cost-based move. The benefit is really the agility and not having to concern yourself with a lot of the infrastructure. Is that what you're saying?
Dwight deVera: Absolutely.
One last question. Can you talk about the benefits that make cloud-based BI so compelling and the steps a company would have to take today to move their business intelligence to the cloud?
Dwight deVera: Well, let's talk about it. It’s not a black and white kind of thing. Cloud-based business intelligence is a component of business intelligence. There's a reason. There's often an initial first project that you're going to move to a cloud, and you're going to do that for a reason. Mergers, acquisitions and divestitures are very common reasons. For example, consider you happen to be working with an organization that buys another organization. You're instantly taking an organization that possibly has two different data warehouses possibly on two different technologies – almost always on two different technologies for some reason – and you need to bring information together very quickly from those two organizations because you're doing something like reporting to the Street.
Now, there's a very traditional way to do that – hiring a thousand interns that are highly educated and fairly well paid, and they can crank out these reports with Excel and glue them together every month until somehow these data warehouses merge sometime in the next three years.
Or you can make this the initiative for your initial business intelligence in the cloud deployment. You can literally make it a first moving component of how you would actually leverage this emerging technology and say, okay, let's take the acquired company and let's grab their data as a secure web service. Take that data as a secure web service just as if you were buying it from an external cloud provider like Salesforce.com or Hoovers.com. Just merge that data with business intelligence data from the traditional system. So it's not that everything will be in the cloud. The organizations that will be 100% totally in the cloud have all their data is in the cloud. They're using all cloud-based services, and they don’t even buy copies of Microsoft Office. They use Google Docs. For them, it’s a no-brainer. All of their BI is going to be in the cloud. But if you're a company that has been doing data warehousing and BI and have existing infrastructure to support that, the cloud will just be one of the tools in your BI toolkit.
Now from a benefits perspective, I think I touched on some of those things earlier. The biggest benefit is really this agility concept – total solutions time-to-market. And total solution time-to-market has a lot of corollary benefits. Making decisions faster absolutely provides bigger benefits than waiting three years or waiting for a spreadsheet to come via email. So faster is better. The solution and the data will be less stale. You hear tragic stories all the time in BI where there is a requirement that’s needed in two weeks. It’s given to the BI team, but they never produced it or they produced it two months too late and missed the boat. It was a stale solution. The total cost from the initial deployment may be lower, but every time you touch data and move a piece of data, you're creating cost, you're creating reconciliation points, and you're creating “that job didn’t run last night” kind of problems. If you just grab it from where it is securely through cloud-based technologies, you deliver faster and with lower costs.
And then finally, it's really just the simple fact of reduced maintenance expense. So when you're using a cloud-based provider, you literally have less infrastructure. BI shouldn’t be moved to the cloud because the analysts say we should move there. BI should move to the cloud because there's a legitimate business reason to leverage cloud-based technologies for a particular BI initiative.
Well said, Dwight. I really appreciate the insight that you provided for our readers in this spotlight.
Dwight deVera: Thanks for inviting us.
Recent articles by Ron Powell
Copyright 2004 — 2019. Powell Media, LLC. All rights reserved.
BeyeNETWORK™ is a trademark of Powell Media, LLC