There are three metrics jostling for the position of most important when considering SaaS implementations: SAS 70 Type II compliance, business continuity and resiliency, and application performance. Each of these measures is of substantial importance to firms; determining which is the most important is nearly impossible.
Firms seeking to optimize their business processes utilizing SaaS must insist on their providers being SAS 70 Type II compliant. This certification provides all of the necessary assurances that your data will be secured, access tightly controlled, all changes to the environment as well as the applications will be documented, as well as a lengthy list of other benefits. If you are going to put your critical business applications in the cloud, make sure your provider has obtained this compliance.
Business continuity and resiliency is a generic argument used by SaaS providers as a benefit to implementing their services. When a firm is conducting work in the cloud, access is available any time, from any web browser, from most any device. However, the discussion point of working in the cloud does not ensure continuity and resiliency for your business application. That assurance comes with geographically distributed data centers as a part of the SaaS provider business model. Recall an earlier post regarding pure SaaS providers not discussing infrastructure. That definition has not changed. A pure SaaS provider partners for mirrored data centers. Mirrored data centers are part of the SAS 70 Type II compliance certification, another strong reason to ensure your SaaS provider has obtained this certification.
Application performance measures are critical when using a SaaS provider. Applications have different characteristics and need to be implemented in a manner that optimizes the application. For example, if you are using a customer care application in the cloud, the voice over IP component is inversely related to the data component, e.g. instant messaging. Voice is highly sensitive to delay and jitter, impacting the quality of the transmission. Data is not. Customers won't notice if an interactive data transmission experiences a second or two of delay.
The Aberdeen Group released a report in December, 2008 titled, Virtual Vigilance: Managing Application Performance in Virtual Environments. The lead analyst on the report was Bojan Simic. I gave him a call to see if his work included the SaaS environment and what insights he could share. He was a wealth of information, but a few items really stuck out.
Mr. Simic’s team interviewed 137 organizations for its report and a few nuggets he shared with me specific to enterprise firms using a SaaS provider included:
1. The top two reasons enterprises use SaaS were to reduce operating costs and to increase flexibility of the enterprise infrastructure. The metrics the enterprises used to measure these two benefits substantiated the SaaS decision and reinforced it as positive for the company.
2. It is not enough to collect performance data; that is a reactive transaction. Those firms that were using SaaS were able to collect actionable performance data. This is data that reflects performance degradation and enabled modifications to be made before the users become aware. Actionable performance data substantially reduces the number of help desk calls and trouble tickets.
3. Firms that used SaaS are able to successfully identify and solve problems better and faster than any other sourcing model including ASPs and Managed Service Providers.
4. SaaS users measured their mean time to repair at a 57% improvement over previous in-house application support.
5. As a percent of total IT spend, SaaS users identified a significant decrease in labor costs specific to application performance management.
Recognizing that this is one analyst firm's research efforts, the findings are consistent with other information we are seeing in the market that using a pure SaaS provider has significant, measurable positive impacts for a company. As you build your business case to move toward a SaaS model, check out the Aberdeen Group's website for additional research it has conducted to support your discussions.
You can find me at: