Introducing Phizzpop!

Stephen Downs, aka Tink, asks how on earth anyone can make money from an RIA. His argument boils down to the observation that it's hard to prevent piracy of RIAs. He concludes that maybe ad-based business models will be the only sustainable model for RIA authors. Although ads are a viable business model, the business case for RIAs goes far beyond ads. In this post, I'll make the business case for RIAs.
We support thousands of ISVs making money on our platform using a variety of business models -- ads and shrinkwrap licenses are not the only alternatives. In fact, the two are somewhat apples-and-oranges.
One way to make money is to establish a payment relationship directly with the people who are going to use your software:
Instead of writing software directly for a consumer, however, you can build a business that connects two different groups of people. These are generally known as "two-sided market platforms", and you can be more creative about getting money:
For an excellent analysis of the various business models used in two-sided markets, you can read "Invisible Engines" from M.I.T. Press.
It doesn't matter if you're doing subscriptions like Netflix, fees like Monster.com, or an ad-funded platform like Facebook -- better user experience means better business. You know this intuitively, but if you need proof, look at this example from Facebook (via Bokardo). When Facebook made a change that reduced the number of pageviews that a user would be forced to do (seemingly suicide for an ad-funded company), their number of users and amount of participation per user rose dramatically.
This is why we invested so heavily in WPF and Silverlight, as well as AJAX. We want ISVs to have more options available to build better user experiences -- no matter what the business model.
The issues of piracy and business model are somewhat orthogonal. The fact that an RIA can be easily copied means that it could be easily hacked to remove ads. Ad-blocker software that attaches to web browsers is already a problem for things like ad-funded search.
Furthermore, traditional desktop software isn't necessarily more resistant to piracy than the average RIA. A WPF-based RIA can take advantage of all of the anti-piracy techniques of a standard Windows application, and it may be possible one day to deploy fully DRM-protected RIAs, but almost nobody does that for traditional desktop applications today anyway. Just as a determined pirate can download and install an ad blocker, he or she can usually find cracked versions of popular software.
This is only a problem when your software is completely self-contained, which is rarely the case these days. Client software is usually coupled with a component that is not easy to pirate, and which provides critical value:
For example, the value to an SAP or Salesforce.com customer is in the information that is stored in the database. If someone pirates the RIA client, but doesn't pay Salesforce.com, she won't be able to see her invoices anymore.
In fact, this is the dominant model on the web. Think about it -- if RIAs are easy to rip off, how much easier is it to do a "view source" and make a "copy" of someone's entire HTML UI? But the Hotmail UI (for example) isn't all that valuable by itself -- before long, you want to get at the actual e-mail messages and photos stored on the server! This is why we say that data is the new platform. If someone ripped off the Facebook client UI, but you couldn't get to your friends list and mini-feed, it wouldn't be all that valuable, would it?
In conclusion, RIAs are simply a new tool that can be used by desktop client and web developers alike to provide better user experiences and augment existing business models. Additionally, RIA developers will use the same anti-piracy techniques used by desktop client and web developers today -- especially by tying the user interface to a service which is critical to the application.