SaaS: The Fact, the Fiction, and the Fanatical

One of the hottest technology topics of late is Software as a Service, otherwise known as SaaS. Given the sputtering state of the global economy, it is not surprising. Smaller budgets, bigger workloads and fewer staff have combined to create a perfect storm of sorts for its growth. But the IT industry, made up of analysts, vendors and trade media, is driving the conversation in the wrong direction. Rather than focusing on the practical and proven, debate has centered on the theoretical and academic. As a result, the market is showing signs of confusion which, if industry analyst survey data is an indicator, threatens to turn into genuine dissatisfaction with even the idea of SaaS. Over the next many paragraphs, I am going to explore these subjects and attempt to shed light on which software deployment option may be right for your organization.

What is shaping the discussion?
Rarely in the real world is any issue, be it personal or professional, a matter of the black or white. Instead, it is about shades of gray. The IT industry, however, does not seem to subscribe to this notion. Otherwise, it would not continue to assign value based on how well a vendor offering aligns with a particular definition, conceptual model, acronym, the latest trend or check box criteria used in analyst rankings.

While such mental exercises are interesting, they miss the point. Organizations do not buy definitions or check marks to solve very specific business problems - they purchase software for this reason. Assuming otherwise only serves to do the market a terrible disservice as it does not reflect the gray-tinted reality that decision makers deal with each day.

Understanding this about the IT industry, it should come as no surprise that the SaaS debate is hyper-focused on the strict compliance of vendors with definitions.

So how does the industry define SaaS?
There is no single definition of SaaS. That said, all are somewhat similar and focused on the notion of completely shared architecture. A good place to turn for a widely-referenced definition of SaaS is Gartner, one of the most well-respected analyst firms in the world. The September 12, 2008, Gartner report, “Market Trends: Software as a Service, Worldwide, 2007-2012,” states:

“Gartner defines SaaS as software that is owned, delivered and managed remotely by one or more providers. The provider delivers an application based on a single set of common code and data definitions, which are consumed in a one-to-many model by all contracted customers anytime on a pay-for-use basis, or as a subscription based on use metrics.”

Vendors that strictly adhere to this definition, and whose software is only available by SaaS, are often referred to as “pure plays.”

For some organizations, a pure play SaaS solution may indeed be the right choice. Consider the following scenarios where a by-the-book definition of SaaS might help companies:

• Business X does not now, nor will it likely ever, have a capital budget large enough to cover the upfront investment required of an on-premise solution. Predictable monthly costs and the ability to pay with operating budgets make SaaS an appealing option.
• Company Y does not have an in-house IT person, let alone a team. A SaaS deployment installed and managed by the IT vendor may be the only way to get the software.
• Organization Z wants to deploy a customer relationship management, or CRM, system. It is also interested in E-Learning and Web conferencing. Applications like these involve common business processes and typically require very basic integration.

Bridge over troubled water?
Stating the obvious, not all processes are created equal. Yes, there are scenarios where pure play SaaS is the best deployment option. For example, it may be appropriate when processes are self-contained and similar across many organizations.

But pure play SaaS is absolutely not right for all situations. For instance, transactional content management software, a “flavor” of enterprise content management, or ECM, is at its core an integration system. It is designed to integrate with any number of systems, by SaaS or on-premise, to automate accounts payable (AP) processes, insurance claims, patient charts and the processing of permits and loans. Processes such as these that require a greater degree of workflow configuration and integration with other systems are not appropriate for the pure SaaS model.

The point is that while SaaS should be a choice, it should not be the only choice. The best option for an organization can only be determined after a careful and thorough examination of all such unique factors. Advancing one as anything but a choice is to offer no choice at all. It is also a recipe for disaster.

Consider the May 29, 2009, Gartner report, “Dataquest Insight: SaaS Adoption Trends in the U.S. and U.K.,” in which some dissatisfaction with SaaS was highlighted. For example, according to the study, the top two reasons organizations put SaaS on hold are high cost of services (42 percent of respondents) and difficulty with integration (38 percent of respondents).

In a July 8, 2009, press release concerning the report entitled “Gartner Survey Shows Many Users are Underwhelmed by Their Experiences of SaaS,” Twiggy Lo, principal research analyst at Gartner, was quoted as saying: “At a time when SaaS is becoming more of a consideration for more enterprises, the results of this survey will be somewhat disquieting for SaaS vendors. Underwhelming customer satisfaction scores, hesitation over the true cost of SaaS solutions, and concerns regarding how successfully SaaS applications can be integrated with other applications all point to issues that will need addressing and resolving.”

Does this mean SaaS is flawed?
What this means more than anything else is that the market is evaluating SaaS primarily on false assumptions. One thing you will almost never hear a pure play vendor tell you is that SaaS may not always offer a lower total cost of ownership than on-premise. So when organizations considering SaaS see actual projected long-term costs, they are going to experience something called “sticker shock.” In the case of SaaS, as evidenced by the survey findings, it is leading to disgruntled former prospects.

The issue of integration is a bit different. There is a blanket assumption that no software delivered by SaaS can integrate as well as its on-premise counterpart. As is the case with most blankets, there are holes in this one. Ability to integrate is determined by the capabilities of each individual vendor. Before assuming the proverbial blanket applies, ask vendors, SaaS and on-premise, to provide examples of customers successfully integrating with systems similar to your own.

Let us close this section with another quote, from the same statement noted above, from Ms. Lo of Gartner. “Most importantly, vendors must re-affirm the fundamentals of the SaaS model – that SaaS solutions are lighter, simpler, more intuitive, more agile and more modest.”

Well said. For it seems clear that when pure plays start to promise more than the model is capable of delivering, chinks in the armor start to appear.

What are the choices?
Although pure SaaS has limitations, there are choices that meet the needs of organizations with more complex requirements. Of course, traditional on-premise deployment remains an option for some. But many simply cannot afford it, support it or justify it. Another model worth considering is one I will refer to as the “choice model.”

The focus of the choice model is to deliver software in a manner that aligns with the needs of the customer. These vendors are not committed to only SaaS, but rather provide a choice of on-premise or SaaS. Such providers may be in the best position to meet the requirements of an organization. Choice approaches heed back to the old fashioned customer service credo: “How may I help you?”

Software is a long-term decision. Over time, circumstances may dictate a change from one method to another. Should your requirements change in a way that lessens the value of SaaS, or if your business needs evolve so that on-premise no longer works, you do not want your organization to be locked into a deployment option. Choosing a vendor that does not offer choice is akin to closing the door and throwing away the key.

Does a choice delivery model include a SaaS option?
It depends on who you ask to define SaaS. In most cases, choice delivery models do not adhere to the strict definitions of what a SaaS architecture should look like. In fact, most pure plays and analysts refer to the SaaS portion of the choice model I am advancing as a hosted deployment. To alleviate any confusion, I will refer this model as hosted/SaaS. What distinguishes SaaS from hosted is the degree to which everything is shared. All pure play SaaS customers typically share all parts of the infrastructure, including the hardware, network, database, software application and the client interface. On the other hand, hosted/SaaS customers have infrastructure segments that are dedicated to their individual deployment.

The hosted/SaaS model is not focused on fitting neatly into a definition. Rather, it is designed to meet custom requirements such as tight integrations with core line of business applications. It does all this while offering many of the same advantages of pure play SaaS, including shared costs, redundancy, scalability and Internet access. To the customer, it looks just like the definition, but it comes without the limitations of a pure play solution.

In the end, organizations have to decide if they want to partner with a vendor that adheres to a strict architectural definition or one proven to solve problems either by SaaS, hosted/SaaS or on-premise.

The successful vendors in the long-run will be able to handle this requirement with the fluidity of a choice model. Purist SaaS vendors will not be able to meet customers’ needs in every situation without compromising some of what they believe makes them stand uniquely apart from on premise software vendors.

Before dismissing all offerings that are not by-the-book SaaS, I ask that all purists answer the following questions:

• Is SaaS architectural purity possible or even desirable in all cases?
• Will the economies of scale of complete multi-tenancy hold up in all scenarios?

In conclusion
Market conditions have combined to make SaaS, hosted or whatever you choose to call it, a more attractive, and perhaps the only, option for more organizations than ever before. And increasingly, they do not care how the software that solves those problems is delivered or defined.

Know that your organization has a choice. It is time for the inflexible dogma that declares pure plays’ definition appropriate for every situation to end. Doing so may actually accelerate the growth of the market for the SaaS delivery model.

If you have questions about deployment methodology or ECM, I welcome hearing from you directly at (440) 788.5043 or ron.mcclellan@OnBase.com.

Ron McClellan is the director of OnBase OnLine and Hosting Services for Hyland Software. Based in Hyland’s Cleveland, Ohio, world headquarters, Ron, a former programmer and network engineer, held executive roles at several information technology firms before joining Hyland in 2004. He is frequently invited to speak about SaaS, including at various regional AIIM shows. You can learn more about Hyland Software at www.OnBase.com