Blog: Wayne Eckerson Subscribe to this blog's RSS feed!

Wayne Eckerson

Welcome to Wayne's World, my blog that illuminates the latest thinking about how to deliver insights from business data and celebrates out-of-the-box thinkers and doers in the business intelligence (BI), performance management and data warehousing (DW) fields. Tune in here if you want to keep abreast of the latest trends, techniques, and technologies in this dynamic industry.

About the author >

Wayne has been a thought leader in the business intelligence field since the early 1990s. He has conducted numerous research studies and is a noted speaker, blogger, and consultant. He is the author of two widely read books: Performance Dashboards: Measuring, Monitoring, and Managing Your Business (2005, 2010) and The Secrets of Analytical Leaders: Insights from Information Insiders (2012).

Wayne is currently director of BI Leadership Research, an education and research service run by TechTarget that provides objective, vendor neutral content to business intelligence (BI) professionals worldwide. Wayne’s consulting company, BI Leader Consulting, provides strategic planning, architectural reviews, internal workshops, and long-term mentoring to both user and vendor organizations. For many years, Wayne served as director of education and research at The Data Warehousing Institute (TDWI) where he oversaw the company’s content and training programs and chaired its BI Executive Summit. He can be reached by email at weckerson@techtarget.com.

This is the fourth in a four-part series on cloud computing for business intelligence (BI) professionals.

Business intelligence in the Cloud is inevitable. In fact, it's already happening. Although Cloud BI hasn't slowed the growth of on-premises BI software, an increasing number of organizations are using Cloud-based BI services and many more are waiting until the time is right.

By Cloud BI, I primarily mean reports and dashboards that run in a multi-tenant, hosted environment and which users access via a Web browser. The reports and applications can be packaged (i.e., Software-as-a-Service or SaaS) or custom-built by the service provider or customer using Platform-as-a-Service (PaaS) or other tools. (See Part II of this series for more detailed explanations of SaaS and PaaS.)

Benefits. The Cloud offers numerous benefits for organizations that want to run reports and dashboards. There is no hardware and software to buy, install, tune, and upgrade. Consequently, there are no IT people to hire, pay, and manage. Applications upgrade automagically and can be scaled seamlessly. Customers pay a monthly subscription based on usage rather than an upfront licensing fee. Essentially, the Cloud speeds delivery and drives down costs. What's not to like?

Impediments. But there are concerns. One of the biggest impediments to Cloud BI today is security, or at least the perception that data in the Cloud is less secure than data housed in a corporate data center. In reality, data is actually safer in the Cloud than in many corporate data centers. The real issue is not security, it's "control." Executives today simply feel safer if their data is housed in a corporate data center. (Ironically, most companies have already outsourced sensitive data, like payroll and sales, to third party providers.) To be fair, some companies, especially those in financial services, must comply with regulations that currently require them to keep data on premise.

E-Commerce. Interestingly, many industry experts raised the same security bogeyman in regards to e-commerce. In the late 1990s, many experts said, "Consumers will never type their credit card number into a Web browser and ship it off to an unknown destination via the public internet because it could be stolen." Of course, we know how this story played out. In 2009, more than 154 million people in the U.S. bought something online, and online sales are growing four times faster than retail sales in general, according to Forrester Research. When it comes to security, convenience trumps fear, especially when fear isn't grounded in reality. The same appears to be happening with the Cloud.

The biggest challenge with running BI in the Cloud involves packaging custom BI development into a cost-effective, online service. By its nature, BI involves creating custom applications that integrate data from unique combinations of data sources. Cloud BI vendors are still figuring out how to deliver BI services without losing their shirts or turning into a custom development shop. (See Part III of this series.)

Finally, some Cloud BI vendors only deliver interactive reports and dashboards (e.g. SAP, Indicee, and GoodData), while only a few offer more in-depth analysis using on-line analytical processing (OLAP) (e.g., Birst) or pivot table functionality against big data (e.g, PivotLink). However, for most organizations getting started with BI, reporting and dashboarding functionality is more than sufficient to satisfy their information appetites.

Gaining Traction

Despite these obstacles, Cloud BI is gaining ground, according to a recent survey of BI Leadership Forum, a global network of BI Directors and other BI professionals. (See www.bileadership.com.) More than one-third of organizations are currently using the Cloud for some part of their BI program, according to the survey. (See figure 1.)

Figure 1. Are you using the Cloud for any part of your BI program?
Cloud- Figure 1.jpg
Source: BI Leadership Forum, June, 2011. Based on 112 responses. www.bileadership.com.

Organizations that have embraced the Cloud point to "speed of deployment" (30%) and "reduced maintenance" (30%), followed by "flexibility" (19%) and "cost" (11%). (See figure 2.)

Figure 2. Motivating Factors
CLoud Figure 2.jpg

Momentum. So far, Cloud BI users are happy campers. Almost two-thirds (65%) said they plan to increase their usage of Cloud BI in the next 12 months. Only 3% said they would decrease usage while another 16% will keep their implementation the same and 16% weren't sure. (See figure 3.)

Among respondents who are not using Cloud BI, 16% said they plan to implement Cloud BI in the next 12 months and 32% were not sure. So Cloud BI has momentum. However, it may take a five to 10 years for Cloud BI to reach the tipping point where it becomes a mainstream component of every BI program. Given Cloud BI's benefits, this trajectory is inevitable.

Figure 3. Future Usage
Cloud Figure 3.jpg

Small Companies Lead the Way

A closer look at the data confirms what many pundits have said about the target market for Cloud-BI software: that it's currently ideal for small companies with few IT resources, limited capital to spend on servers and software, and minimal to no BI expertise. Almost half of small companies under $100M in annual revenues (46%) use Cloud BI in some shape or form. In contrast, large companies with over $1B in annual revenues are almost less than half as likely to adopt the cloud (29%), while medium-sized companies with between $100M and $1B in annual revenues lag further behind with less than one-fifth using BI in the Cloud (18%). (See figure 4.)

Figure 4. Cloud BI Deployment by Company Size
Cloud Figure 4.jpg

Small Companies. For small businesses without legacy BI applications, Cloud BI services are a godsend. The economics and convenience are compelling. Instead of passing around spreadsheets, small companies can implement a Cloud BI service to standardize reports and dashboards and make them available to all employees anywhere via a Web browser.

"What's refreshing for me is that I can go in at any time of day and [run a] report on any metric in our organization, such as item received delivered, inspected at the category, personnel, or employee level and track it by any time period," says Wayne Deer, vice president of operations, at Gazelle, an electronics recycler, which uses GoodData's Cloud BI service.

Large Companies. Interestingly, large companies are the next most prevalent users of Cloud BI services. Often, it's a department head who wants to build a BI application quickly without getting corporate IT involved. Like small companies, departments at larger companies often have limited budgets and BI expertise and most don't want the headaches and expense of having to maintain servers and software.

But enterprise BI managers have assessed the potential of Cloud BI and like what they see. Unfortunately, many are hamstrung by legacy BI implementations. "I see us moving very slowly with adoption because of installed base and switching costs," wrote Darrell Piatt, Director and Chief Architect at a large professional services firm based in Virginia, in a BI Leadership Forum discussion thread. "When and if we decide to replace our BI infrastructure, Cloud BI offerings will be seriously considered."

Mid-size Companies. According to figure 2, medium-sized companies are least likely to adopt Cloud BI services. The reason is that most have already implemented an enterprise IT platform, usually Microsoft SQL Server, which bundles BI tools and applications for free. If the organization has assigned an analyst or IT administrator to build and maintain enterprise reports and dashboards using the platform, it likely has little bandwidth, incentive, or capital to change courses and introduce an alternative BI stack, unless it is having difficulty meeting user requirements.

Cloud BI Vendor Perspective

Given that they provide a service that makes them an extension of an organization's IT team, Cloud BI vendors have a good handle on who their customers are and what they are doing with their services.

For example, Sam Boonin, vice president of products and marketing at GoodData, says that his company's customers fall into three camps: 1) fast-growing technology companies that run all their applications in the Cloud, 2) departments in larger companies, many of which have implemented Salesforce.com and are comfortable with the SaaS model, and 3) SaaS vendors who OEM their product.

Boonin also said that 90% of GoodData's customers start by using one of its packaged applications, which generate reports against a single SaaS-based, front-office application, such as Salesforce.com, ZenDesk, and Google Analytics, or an on-premise package, such as Microsoft Dynamics CRM. (GoodData currently offers 20 packaged applications.) These applications, which deploy in hours, start at $1,000 a month and are often bundled into third party, SaaS products.

Many customers then extend their packaged SaaS BI application by customizing data models and adding other data sources. GoodData configures or customizes the application to the customer's specifications. The process takes roughly six weeks and typically raises the monthly subscription price to between $3,000 and $10,000 a month. "For companies used to spending $500,000 on BI and getting virtually nothing, they see us as a godsend," says Boonin.

Today, almost half (43%) of GoodData's customers generate reports that run against multiple applications. These customers generate the majority of GoodData's revenues. Currently, GoodData has about 100 direct customers and 3,000 indirect customers through OEMs. This is comparable to other pureplay Cloud BI vendors. "Business is good," says Boonin. "We have a $25 billion failed market to disrupt."

Conclusion

Some experts see dark clouds in the Cloud BI market. While ramp up of Cloud BI services hasn't been as fast as some anticipated, it's clearly catching on. The BI market poses unique challenges compared to other SaaS market segments that automate operational business processes. That's because BI applications are generally custom built and require companies to integrate data from multiple sources. Cloud BI vendors have taken different approaches to "servitize" a custom application, which is a logical contradiction. Not all have succeeded, but those still in the market are making headway. The value of Cloud computing is high, and the BI industry will eventually find a way to succeed with it.


Posted September 9, 2011 9:34 AM
Permalink | 3 Comments |

3 Comments

Thanks Wayne - this has been a very interesting series.

I agree completely that a move to cloud-based BI is inevitable, and the benefits will far outweigh the limitations. I'm in financial services so there are regulatory limitations we're bound by, but I still firmly believe the transition is a 'when' not an 'if'.

My main questions over cloud adoption in the BI space are ones I haven't seen addressed or discussed elsewhere.

-Once the cloud BI lifecycle moves beyond early adopter to maturity, what will the implications be for mergers and acquisitions? Merging 2 cloud-based BI environments adds a level of complexity, particularly since one BI provider will probably be losing business because of it

-In a similar vein, SaaS and PaaS will likely make changing to a different vendor more complex. 3rd party tools to migrate semantic data may no longer work, and in a worst case you may be faced with starting again from scratch

-Finally, the 'service' aspect of SaaS and PaaS would need to be closely watched from a global point of view. An update scheduled for 10pm on a Sunday in NY would be 2pm on a Monday here in New Zealand.

None of these are show stoppers of course, but they will present challenges in the medium term, and potential opportunities to the vendors and organisations which can stay ahead of the game.

Richard Jarvis @ManBagBlog

These are valid concerns, but I think vendors and service providers will soon offer utilities for handling such issues. We're still in an early adopter phase and administrative and management utilities always come much later. But ultimately, it's still about the data; it's just now that it's stored in a different place. In many ways, cloud-based services will ease migration tasks in an M&A because there is no hardware involved and fewer staff. 24x7 issues don't change when you move data off premise either.

Wayne,

Thanks for the 360 discussion on cloud computing, I find your discussion thoughtful and provoking.
One area I think you underestimate, perhaps in other blogs, is the value the cloud brings to those "fixed costs" involved with operating a BI solution. I have deployed cloud computing solutions and have actually found this a major plus. For example: if a client wants hot or cold backups, the environment is defined to provide it, with cost considerations provided. Also, the cloud computing infrastructure bears the responsibility of finding usage for unused infrastructure capability (mips, bandwidth and storage, etc - a responsibility the business carries when in house. For example, when we operate an etl application that runs 8 hours weekly, we bring up the server, run the application(s) and then bring down the server, and the cost of server operation only applies when the server is up!
Of course, you are spot on to say that BI applications are loaded with data integration challenges that do not go away just because the application is on the cloud. And that has been true here. But, imagine the plus for integration services that use the same technical infrastructure, tools, design practices, etc, from one consulting or managed services engagement to another. The consultancy does not have to learn a new way to use the etl tool, the configuration management practices, audit, balance and control, DB administration, or at least not have these (they feel almost a custom call out every engagement) affect the time and effort with a BI solution. Instead, they can leverage the same infrastructure and design practices from one service to another across different companies. Last, when the project infrastructure needs some more "oomph", the performance in either dev, or test, or production begins to lag, I have found that moving the integration solution components, one or all, onto preconfigured infrastructure options, is an exercise that takes minutes, not weeks or months, with the associated costs practical, easy to communicate and get business support for. My 2 cents, thanks for the great discussion!

Leave a comment

    
Search this blog
Categories ›
Archives ›
Recent Entries ›