In a post last week about Google as a Platform Partner I listed some of the things that make Google's platform in general, and Google Calendar in particular, attractive to me as a third-party developer. But I left something out. At the risk of stating the obvious, without a large and growing number of users already on a platform, apps built on that platform have little chance of success.
I was reminded of this simple fact this morning as I was reading Paul Buchheit's blog. Paul was the original developer of Gmail, and is now at FriendFeed. (He's also famously responsible for Google's "Don't Be Evil" motto.) In a recent post, Paul quotes Marc Andreessen's Guide to Startups:
I'll assert that market is the most important factor in a startup's success or failure. Why? In a great market—a market with lots of real potential customers—the market pulls product out of the startup.
So how do you find a great platform? (I think it's fair to substitute "platform" for "market" when talking about horizontal applcations.) Using Marc's definition, you look for one with lots of real potential customers. The "lots" part is easy: you're looking for a large number of users. But Lotus Notes has lots of customers and, with apologies to Ed Brill, I wouldn't suggest a startup develop for it. It has few "real potential customers", which to me is indicated by growth.
Before we decided to create Spanning Sync, I had been building various services on the Salesforce.com platform. I was intrigued by Google's new Calendar application and its even newer API, and decided to bet that it would provide fertile ground for application development. I started working on a service that would sync Salesforce.com and Google Calendar. It was a technically interesting project, and was proving to be a popular idea. In fact, Salesforce asked me to speak about GData at their annual DreamForce event.
I didn't know how many users Google Calendar had, but I knew it was a large and growing number. But at some point it occurred to me that Salesforce had fewer than a million users (and perversely, only a fraction of them are even allowed access to the Salesforce API), and while their growth was impressive, that absolute number of users was still relatively low.
The other platform I used on a daily basis—Mac OS X—had roughly 25 million users with growth that CNET called "nothing short of spectacular". So I put the Salesforce sync project on hold (and later passed it off to a consulting firm) and shifted my focus to syncing Apple iCal and Google Calendar. That turned out to be a good decision.
I attribute much of the success of Spanning Sync to the fact that the two platforms on which it's built—Apple's and Google's—are both large and growing. But there's another platform factor that I think was absolutely critical to ours success. I'll write about that next time. Right now I have to fix a bug relating to canceled detached events and refresh syncs.