One of the topics which leading Software-as-a-Service (SaaS) vendors and industry analysts are most vehement about is that software vendors cannot survive and succeed supporting a ‘hybrid’ model.

This issue arises every time an incumbent software vendor–my definition of a “ISV”–rolls out a SaaS solution while also trying to sustain its legacy, on-premise application. There are plenty of impediments to success in this balancing act across the entire lifecycle of a product extending from software development and delivery to sales and support. These technological and organizational challenges are major obstacles to success for ISVs trying to keep pace with the SaaS movement.

However, despite growing interest and adoption of SaaS as well as other ‘cloud’ computing alternatives among organizations of all sizes, many IT and business decision-makers continue to feel that they must make an ‘either/or’ judgement when it comes to on-premise versus on-demand solutions. This often confronts with an unnecessarily polarized set of options rather than giving customers a variety of complementary choices that enable them to locate their applications wherever they like.

I believe that this no longer needs to be the case. Instead, I think SaaS and cloud computing vendors should adopt a different attitude toward the hybrid model to better respond to their customers’ preferences. If vendors adopt this new approach, it could remove one of the last barriers to broad-based acceptance of SaaS and cloud computing among small- and mid-size businesses (SMBs), as well as large-scale enterprises.

As I’ve written, and many others have stated elsewhere, building and selling a traditional software product is fundamentally different than delivering and supporting a SaaS solution. Supporting these two differing models creates internal redundancies and external conflicts which are costly, inefficient and doomed to failure in most cases.

Having said that, I’m becoming convinced that some ISVs can survive and will succeed by offering customers the choice of an on-premise and on-demand solution. In fact, I think it will be necessary to do so in order to satisfy the demands of those customers who are not comfortable with relying on a ‘cloud’-based solution to meet their IT or business needs.

While customer concerns about where a software solution, or even the application data, resides may not be entirely rational at times, it may not be necessary in the future for ISVs to have to convince them to part with their data or depend on an application hosted in an unknown location.

Instead, a variety of players in the SaaS and cloud computing market are leveraging an ‘appliance’ approach which permits customers to deploy the vendor’s on-demand solution behind the firewall where it is regularly updated and upgraded via a synchronization process similar to that which has become acceptable in a variety of other situations, such as managed storage, back-up and security services. It is also becoming possible with Google Docs offline and Adobe Air.

This idea is already being demonstrated by companies like Cast Iron Systems in the data integration arena; NTRglobal in the remote support management services business; and St. Bernard in the security solutions realm.

Although none of these companies are delivering major enterprise applications, they are all offering customers the choice of deploying their equally important solutions in the ‘cloud’ or behind the firewall.

And, if Google, IBM, Microsoft and others can modularize their data center capabilities into ‘pods’ which can be deployed anywhere, what is to prevent Salesforce.com or other enterprise SaaS vendors from doing the same thing with their applications.

(I’ve been hearing rumors for a while that Salesforce.com is already allowing some of its largest customers to host its applications behind the firewall.)

Now, it is important to note that this approach still requires an ISV to evolve its software design to sit on a single multi-tenant style architecture and code base in order to be operationally feasible and cost-effective.

But, the enabling technologies are quickly evolving to satisfy these requirements. And, customer demand definitely exists to make this approach readily acceptable and profitable.

Let me know if you think I’m crazy or if you know of other examples which support my argument.