With traditional business intelligence initiatives sometimes taking up to 18 months to complete at costs that can run into the hundreds of thousands of dollars -- if they don’t end up failing altogether -- it’s no surprise that a still-small but growing number of organizations are eyeing cloud BI tools as a more cost-effective, fast-track approach.
However, Software as a Service (SaaS) BI technology also has potential downsides or complicating factors that would-be users should factor into their thinking before committing to a deployment.
First, the plus side: BI in the cloud, like other applications delivered via the SaaS model, promises a number of inherent advantages over on-premises systems. For starters, the cost of entry for SaaS business intelligence typically is much less expensive because companies trade the need to purchase software licenses and stock their data centers with new hardware for a pay-as-you-go monthly subscription model.
In much the same vein, SaaS BI tools also hold out the promise of a lower total cost of ownership (TCO) over time because there are no systems to set up and manage internally, thus reducing demands on IT resources. Furthermore, industry analysts say that organizations usually can get up and running with cloud-based BI applications much more quickly than they can with on-premises systems -- often within three to four months, or even less time for small workgroup applications.
The economic downturn and continued tight IT budgets have also opened doors to SaaS BI that otherwise might have remained shut. “The fact that there are lower up-front costs, that it’s simpler to get started and that it’s all operational expenses, so there’s no need to capitalize anything, are big advantages right now and are not to be overlooked when evaluating the cloud,” said William McKnight, president of McKnight Consulting Group LLC in Plano, Texas.
Potential red flags for cloud BI tools
On the flip side, cloud BI tools do have some drawbacks, particularly for organizations in industries that are highly regulated. For example, businesses in the health care or financial services industries, where there are strict rules governing privacy and how and where data is stored, might not be good candidates for SaaS BI, according to McKnight and other analysts.
In addition, analysts caution that costs aren’t always lower with SaaS BI. Factors such as hidden fees charged by vendors and the number of data sources that need to be integrated to feed the BI system can have a big impact on the final bill for a SaaS deployment, they said.
“There’s a misnomer put out there that SaaS is always cheaper, and it’s not,” said Shawn Rogers, an analyst at Enterprise Management Associates Inc., a consulting and market research firm in Boulder, Colo. “Companies have to do a TCO analysis to ensure that their particular project makes sense given the number of users and data integration issues.”
PSA Insurance & Financial Services Inc. didn’t start with SaaS BI, but business intelligence in the cloud ended up being a better fit for its reporting needs than on-premises software was, according to Andrew Bartels, IT director at the Hunt Valley, Md.-based company. PSA initially deployed a traditional BI application but found it was expensive and difficult to manage, particularly when trying to integrate data across different information silos. Moreover, business users had to wait for IT to design and generate reports, creating a lag time in delivery that reduced the usefulness of the reports.
“From my perspective, BI is slicing and dicing data on demand to better understand how the business is performing and to look at how it’s performing in different ways,” Bartels said. “The time between reports was too long to be effective. By the time a business person got the report, the data was old, it could be wrong or it was missing fields.”
Cloud business intelligence to the rescue
The cloud BI tools, which PSA switched to two years ago, mitigated that by enabling business users to build their own reports. In addition, the SaaS application provided increased flexibility compared with the on-premises software: “It can absorb any data I throw at it, and it allows me to create relationships between data so it’s meaningful to the business,” Bartels said.
While data security was an initial hurdle given that PSA is in the financial industry, Bartels said a check on the SaaS BI vendor’s security practices and consultation with PSA’s compliance group eased his concerns. “Security needs to be there, but the important thing is to do due diligence,” he advised. “When something is physically stored in the data center behind your firewalls and monitored by your own people, you have a sense of security. Whether it’s a false sense, that’s another question.”
The Association for Manufacturing Technology (AMT) is another SaaS BI user. The technology enables the AMT to deliver BI data and reporting services to its member organizations in a timely and scalable fashion, said Pat McGibbon, vice president of strategic information and research for the McLean, Va.-based trade group.
Prior to its SaaS BI implementation earlier this year, the AMT served up monthly reports to members via the Web in PDFs or a comma-delimited file format. But now, McGibbon said, end users at member companies can create their own reports and dashboards, giving them much quicker access to data for analysis purposes.
In addition, user upgrades at the member level are automatically handled without the need for any intervention by the AMT’s IT workers, and the SaaS BI system can scale up to support more users without requiring the trade group to invest in additional hardware or staffing. “The ability to have a scalable product was a really positive thing,” McGibbon said of the decision to go with SaaS BI. “We could build this out for one person or a million people without making a $6 million investment in a data center.”
Beth Stackpole is a freelance writer who has been covering the intersection of technology and business for 25-plus years for a variety of trade and business publications and websites.