By Phil Wainewright on July 17, 2009 3:24 PM

Many of the calculations routinely used to compare the cost of SaaS to conventional software are based on completely bogus assumptions. This isn't an act of deliberate deception, it's simply a case of people shoehorning SaaS into the same criteria framework they'd use to compare two competing licensed software products.

The most egregious consequence of this sort of false calculation is to grossly understate the long-term costs of owning and running conventionally licensed on-premise software. All too often, I see people taking the cost to acquire a conventional application system, dividing it by three, four or five years and then simply comparing that figure to the annual subscription cost of a SaaS equivalent. Such calculations seem reasonable enough until one considers that they miss out one of the SaaS model's most significant benefits of all: the elimination of the upgrade disruption that occurs whenever a conventional software product goes through a major release. SaaS products have a completely different approach, with frequent, non-disruptive upgrades, both to the underlying infrastructure and to the top-level application functionality. To end the lifetime cost-of-ownership comparison at precisely the point when the benefits of that different approach really kicks in is to stack the comparison heavily in favor of conventional licensed software — and that's why I use the strongly loaded term 'bogus' to describe such cost-of-ownership calculations.

A blog post this week by Dan Druker, SVP of SaaS vendor Intacct (disclosure: a recent client) discusses a number of ways in which TCO comparisons should be structured to ensure that SaaS is compared equitably against conventionally licensed alternatives. He sets out a checklist of components that are often overlooked when totting up the true cost of implementing and operating on-premise applications. I've selected below the ones that I feel are most often overlooked:

• Infrastructure stack software "... Client Access Licenses, Windows Servers, Application and Database Servers, Middleware, SharePoint, Citrix servers and VPN's for remote access."


• Development and test systems "... most buyers will need two sets of systems for development / test and production."


• Hardware replacement costs "... Servers, storage and other hardware components wear out and need to be replaced ..."


• Faster problem resolution "... The customer doesn't have to describe the problem to the support rep anymore — the both can look at it at the same time. The ongoing support savings are significant." I would add that total system downtime is generally far lower for SaaS, both planned and unplanned.


• Re-implementation costs in years 3 to 5 (As mentioned above) "Most TCO models miss the need in the on-premises software world to re-implement and upgrade to new versions of either the applications or the underlying infrastructure stack ..."


• Higher worker productivity "... Because workers can access the system anytime and from anywhere, they typically get more done than if they can only use the system in the office ..."

The other benefit I would add that I think Dan has missed from his list is the ability to modify functionality and processes in SaaS applications much more quickly and in a more iterative way than with conventional software, which means the business can operate more efficiently and seize more competitive advantage. These are huge knock-on effects of adopting SaaS, but they only become evident after a business has switched to SaaS so they're difficult to build into a return-on-investment projection.

One day all this will have been measured and quantified, but by then the only people still moving to SaaS will be doing it to catch up with everyone else. If you want to get ahead of the field, you'll have to adopt SaaS before all the benefits have become plain for all to see.