Do You Really Want Software As A Service?

CIOs know one thing for sure: They are fed up with current licensing models for software. A recent IDC study (IDC is owned by IDG, which also owns CIO magazine) of 257 IT executives (unfortunately, I can’t link to the survey) found that they are so frustrated with software licensing that deep discounts on the software–as much as 100 percent!–don’t lift their moods. They are convinced that the software companies will rip them off somehow. That somehow is usually via maintenance or subscription fees (check out this article we wrote on the subject).

The IDC survey found that companies believe they use just 16 percent of the software they buy. The rest is just there to pump up the vendors’ fees. Analysts I’ve spoken to over the years have estimated that maintenance fees are about 70 percent margin, of which some portion (guesstimates are at about 30 percent), goes to fund future development. With enterprise software such a tough sell these days, vendors are discounting the stuff up front and building their profits into the back end.

It’s unreasonable to expect vendors to discount up front and then cut maintenance fees on the back end–they have to get their money somehow. Customers bombard them with requests for new features and functions they say they want in the next version. There has to be some kind of ongoing payment arrangement to keep that cycle going. But the lack of transparency has killed the model. Customers don’t know how much money the vendors are really making and when. When you’re paying for software endlessly without knowing whether you can afford to upgrade the software when a new version comes out, or how much you will be charged for new software that the vendor develops, or how much of a say you will have in that development, you are tempted to simply reject the entire model out of hand. I think we’re there. When CIOs say they don’t trust the model even if they get the software practically for free up front, then something is clearly wrong here.

But the question is, how to escape this co-dependency of mistrust? Software as a service, at least as it is currently being offered, is not the answer, according to the survey. It was interesting to read between the lines of the results. For every major category of enterprise software, IT executives in small and large companies (we used to think only small companies wanted software as a service, right?) said they wanted software delivered as a service. Twice as many wanted to buy by the drink rather than owning it outright. Until you get to the one category where software as a service is really being tested: CRM. We all know about the success of Salesforce.com. No need to belabor that. But when the IT executives were asked whether they wanted CRM as a service or wanted to own it, the numbers reversed. Twice as many wanted to own. That seems to be a direct reaction to Salesforce’s success in the market. Business users can go around IT and set up their own Salesforce.com accounts. IT can’t justify the investments it has already made in CRM any longer. And probably the biggest resistance point of all: you can’t integrate the internal, usually customized CRM stuff with the service (though Salesforce has some customization services–see Chris Lindquist’s blog for more details).

Makes me think of Winston Churchill’s famous quote about democracy being the worst form of government except for all the others that have been tried. I think that we will get to software as a service someday, but not with the architectures that we have today. Integration is holding it all up. An independent integration layer in the architecture could make it easier for CIOs to want what they say they want. What do you think?