I was driving to work the other day when a radio ad caused me to do a double-take. The product pitchman was pushing an HD radio service. Having just updated the televisions in my house, I know that “HD” is a term typically used to describe picture – not audio – quality. Sure enough, a quick Internet search clued me into the fact that HD radio stands for “hybrid digital” radio, which has nothing at all to do with high-definition television.
On the one hand, I give credit to the HD radio product vendor for its innovative marketing technique. Certainly there is logic to leveraging a hot industry acronym as a means to drawing-in customers. On the other hand, I wonder what impact the tactic has on customers, who are persuaded to buy based on a misleading and misbranded pitch.
The experience reminds me of the on-demand software model. Legacy software vendors everywhere are telling anyone who will listen about their new “on-demand” strategy. Clearly, the term means different things to different people. Unfortunately, only one definition of on-demand software – one that includes a multi-tenant architecture – returns the value that customers expect when they choose to go on-demand.
Legacy software vendors prey on customer confusion regarding on-demand software. Like the HD radio vendor, traditional software companies have a lot to gain by tying themselves to a hot industry catch-phrase such as on-demand. Rather than continue to debate this issue, I think it’s more valuable to address the questions I hear most frequently from customers: “Why do I care whether your application is on-demand?” and “Why is a multi-tenant architecture good for me?”
Believe it or not, the answers are surprisingly straightforward. Customers should care whether or not a software application is on-demand because what is good for me, the vendor, is ultimately good for you. If we can support a single line of code and maintain only a single version of the software, we can focus our engineering talent on bringing new functionality and improvements to you. If we employ a true multi-tenant architecture, we can deploy seamless upgrades on a large scale whether we have 100 customers or 10,000 customers.
By contrast, legacy software vendors whose “on-demand” applications are lacking a multi-tenant architecture simply can not offer the same value to a customer. Just imagine what happens when the vendor has to upgrade 10,000 customers, all with unique application schemas. How many engineers does the vendor have to employ to manage that process? What do you, the customer, have to give up since these staff resources can not work on new features and enhancements? It only gets worse when the vendor has to make a data schema change in a future release.
Clearly there is an important distinction between on-demand software that is multi-tenant and that which is not, just as there is a difference between HD radio and HD television. That’s why my guidance to customers is, ask a vendor before you buy whether or not they have a multi-tenant application architecture.