Vance F. Brown
What is SaaS? The term “SaaS” is an acronym for “Software as a Service.” Forrester Research reports that 54 percent of all organizations are interested in SaaS deployment. And according to Gartner, even though only approximately 5 percent of the ITSM software market today has embraced SaaS, it is estimated that 10 percent of this industry will do so by 2012.
So while it certainly appears that many organizations are demanding the benefits offered by SaaS, evidence indicates many organizations are confused about what SaaS truly entails, and whether or not it is the best choice for their software needs. Consider the following question recently posted on a SaaS blog: “Please tell me, just what exactly is SaaS? Is it a ‘business model?’ Is it a ‘delivery/hosting model’? Or is it something else?”
What a great question! This person is not alone in his or her confusion. Unless we can define the term and understand the components of SaaS, it is difficult to determine whether or not SaaS is the best option for an organization’s specific needs. When I ask this question of others, the typical response is something like the following: “SaaS is a term that describes software that is delivered over the Internet through a browser and is paid for on a subscription basis.” Wikipedia suggests that SaaS, “sometimes referred to as ‘software on demand,’ is software that is deployed over the Internet….With SaaS, a provider licenses an application to customers as a service on demand, through a subscription or a ‘pay-as-you-go’ model.”
The purpose of this article is to help define and eliminate some of the confusion about SaaS. After looking at a brief history of SaaS, we’ll explore SaaS from its three major components: (1) a financial model; (2) a deployment model; and (3) a user-interface model. Some believe that SaaS is limited to subscription pricing, off-site hosting, and browser-only access. I believe SaaS is about customer choice. We’ll examine the “once size fits all” SaaS approach versus SaaS that offers choices in its core components. Finally, I believe so strongly in implementing SaaS with user-optimized choices that I propose a new concept that might more accurately describe what most mid-to enterprise-sized companies are demanding: Software as a Choice (SaaC).
The SaaS Evolution
In “On-Demand/SaaS Reality,” author Rick Sklarin does a great job of describing the three generations of SaaS evolution. The first generation, which he calls “SaaS 1.0,” became popular in the 1990s and involved the early application service provider (ASP) model. With SaaS 1.0, an ASP would purchase a limited license from the software vendor and then offer the software functionality to companies on a subscription-pricing basis. This first generation of subscription-based pricing was typically offered via client-server, was mostly accepted by the SMB market, and had very limited ability to customize the applications and integrate with other systems.
The second generation of SaaS, which Sklarin calls SaaS 2.0 or “pure SaaS,” was made popular by the success of Salesforce.com. In the ITSM space, this “pure SaaS” approach was first successfully marketed by Service-now.com. This second generation of SaaS currently defines the SaaS marketplace. It typically requires: (1) subscription pricing; (2) deployment and hosting of the application and the data off the customer’s premises (i.e., “off-premises”); and (3) delivery over the Internet through an HTML browser (such as Internet Explorer, Firefox, Safari, etc.). In short, “pure” SaaS (or “modern” SaaS, as some vendors label it) offers few choices. This model has also been criticized as having very limited customization and integration capabilities.
As a result of these deficiencies, Sklarin predicts the evolution to SaaS 3.0, which is a hybrid SaaS alternative, offering many more choices for end users. As he describes it, “larger companies are demanding an evolution from the pure-play SaaS model.” He goes on to say that larger customers are requiring integration with on-premises applications and sophisticated configurability to meet specific business requirements, together with flexible delivery models. In other words, enterprise companies will ultimately demand more options than “pure SaaS” can provide.
Software as a Choice: The Third Generation of SaaS
The Financial Model: Lease or Own?
It is quite odd, and unfortunate, that the financial model for “pure SaaS” has been inherently tied to the delivery and hosting models, which are technical issues. The financial model has legal and accounting implications, but no technical connection. From a legal perspective, software is generally licensed from software vendors. However, these licenses can be either purchased or leased. By purchasing the software license from the vendor, the customer generally has the right to use the software in perpetuity (i.e., forever). When the software is leased, however, the customer only has the right to use the software for as long as the subscription fee is paid. If the vendor goes out of business or becomes insolvent, a leasing model could be problematic. Leasing software, commonly referred to as a “subscription” model, is not much different than leasing a car: It may be cheaper in the short term (i.e., the first two to three years), but it is typically much more expensive over the long term.
Notably, innovative software vendors are beginning to offer a “lease to own” financial option. They understand that some organizations may prefer initially to lease the software because of short-term cash constraints, and then purchase the licenses when operating conditions change. Some companies offer various discounts on the purchase price based on how long the software has been leased.
From an accounting perspective, when software is leased it is paid out of the company’s “operations budget.” In other words, all of the current subscription costs can be expensed on the income statement in the current period. Alternatively, purchased software over a certain dollar amount must be capitalized. The software shows up on the balance sheet as an asset that must be depreciated over time (e.g., three to seven years).
So which financial approach is preferable? It depends. Some companies that are constrained from a cash-flow perspective may be more short-term oriented and might choose to lease the software. Other companies might be concerned about utilizing a mission-critical application that they do not own. These are questions primarily for the lawyers and accountants, with some input from the IT professionals. There is no “right answer.” It will vary based on the business’s circumstances. With respect to the financial model, choice matters. But there is no technical reason that a software product could not be deployed on site, but paid for using a subscription model.
Deployment Model: On-Premises or Off-Premises Hosting?
Should the application and data be hosted “on premises” or remotely, “off premises”? Traditional “pure” SaaS typically requires that the software application and the company’s data be hosted off the company’s premises, which has some real advantages. For example, the software vendor or service provider supplies and administers the server and the operating system, and deploys all software updates from a remote location. Accordingly, all such costs and services are bundled in with the leasing or subscription price. Such a model can offer great efficiencies and economies of scale.
However, this approach is hardly a “no-brainer.” Some organizations, such as government and financial institutions, have greater security requirements for their data. The concept of hosting the data somewhere else and trusting an outside organization to protect their confidential data can be a show-stopper. Hosting data off site also has implications for organizations that are subject to government requirements, such as the EU Data Privacy Directive, Canada’s PIPEDA, the US’s HIPAA, and other governmental regulations. Additionally, most SaaS 2.0 hosted environments require that every client has its software and data upgraded at the same time, offering no choice regarding when that occurs. That may be great for “early adopters,” but for companies that would rather wait until all the kinks and bugs are worked out, it is not the preferred approach. Another consideration is that integrating with existing on-premises systems is inherently easier when the application and data are hosted in the same environment and behind the same firewall as other integrated on-premises applications.
From a technical perspective, customers should be able to switch back and forth between hosting on-premises and hosting off-premises as their business needs change. This choice should be driven by the customer and should not matter to the software vendor, with the obvious exception that the software vendor will charge additional fees for off-premises hosting costs. The bottom line is that the hosting model should be a choice. Once again, there is no “correct answer.” It depends on the business needs.
User-Interface Model: Rich-Client or Browser?
Transporting data via the Internet has become the de facto standard for delivery, and for most companies, this approach is much cheaper companies than deploying large area networks on a global basis. The Internet has been revolutionary, in that it provides an inexpensive way to share data from multiple geographic locations. However, should the data and application be accessed via a browser or a smart-client? From a technical perspective, the hosting choices discussed above (“on-premises” versus “off-premises”) should support either method of access. However, many “pure SaaS” vendors favor remote hosting models that require the application and data are accessed using a common browser, such as Internet Explorer, Firefox, or Safari. This is not the case with more modern, flexible software applications.
Regardless of whether a smart-client or browser-client is utilized, software applications today can deliver data over the Internet in exactly the same manner, utilizing protocols such as SOAP over HTTP. In other words, today’s smart-client applications can still be web-enabled. These two types of user interfaces can access the data in the same manner. Similar to the financial choice and the hosting choice, there are advantages and disadvantages to each. With a smart-client, less bandwidth is typically needed and the user experience can be faster and more robust, because the screens and other software overhead do not need to travel over Internet. As I wrote this article, I found the user experience of Microsoft Word’s rich-client application to be far superior to the browser-based Word application. Similarly, the user experience for an iPhone app (also a smart-client) is far better than accessing an application on the iPhone from its Safari Internet browser. Normally, when someone is working in an application two to eight hours a day, the rich-client user experience is greatly preferred over the browser-based application.
However, browser-based applications also have their advantages. For example, all upgrades to the software are done remotely, at the server level. This approach is no different than the mainframe computer in the 1980s that was accessed using a “dumb client.” With a smart-client, updates to the client must be downloaded or “autodeployed.” However, this automated installation typically requires very little effort on the part of the end-user. Apple’s iTunes application is a common example of a client update that must be autodeployed. Windows Update is another example. Probably the greatest advantage of a browser-only application is that it can be accessed from different computer platforms, such as Windows, Mac, or UNIX. Such crossplatform access is very convenient, and is often a business requirement. Browser-based applications are also ideal for occasional-access applications, such as Amazon.com, or for users that only occasionally access the application.
Once again, the choice of interface depends on the circumstances. There are good business reasons for each. It seems like the ideal solution would be for the user to be able to choose a rich-client interface as the standard, yet have the capability to switch to a browser interface when required (e.g., when accessing the data from a UNIX system). Once again, the data should be blind to how it is exposed. With both user-interface options, the data can be delivered over the Internet.
SaaS 2.0 (aka “pure” or “modern” SaaS) has been narrowly defined and deployed. Because of this, its days may be numbered. Today’s enterprise-savvy customers are demanding more choice in the financial model, the deployment model, and the user-interface model. An organization should be able to choose, for example, to host the application and data on-premises, yet pay for the software using a subscription model, all while accessing the data using either a rich-client or a browser. SaaS is here to stay, though it will continue to evolve; and if any model is really going to offer true “service,” it must provide the customer with choices. Good-bye “pure” SaaS…hello SaaC, “Software as a Choice”!
Vance F. Brown is president and CEO of Cherwell Software, a leading developer of IT service management software, offered “as a choice.” Formerly, Vance was president and CEO of GoldMine Software Corporation (currently FrontRange Solutions, the makers of HEAT and GoldMine). Vance graduated summa cum laude from Wake Forest University, with concentrations in economics and computer science, and received his law degree with honors from the University of North Carolina.