Making the web into a banking platform (whether they like it or not)

Last week, my company, Wesabe (which makes a personal finance community site), launched a REST API that allows anyone to get their bank or credit card data in XML, Excel, CSV, or a bunch of standard financial formats. Tonight, we launched an open source Firefox extension that allows anyone to automatically extract data from their bank every night, and upload it to Wesabe, regardless of whether their bank provides automatic download or not. Both of these features work for any bank in any country, as long as they support one of our export formats (OFX, QFX, QIF, OFC, Quicken, Money, and a few others coming soon). Data in, data out, free and easy.

There's a basic Web 2.0 story here, which is simply that opening up APIs and embracing the web as a platform is a great way to empower the people using your service. It's been amazing to me to see one developer after another approach us about using the API, even in its early form. But, while obviously I'm totally biased, I think think there's a deeper story here, too, and I thought it would be worth calling out some of the things that make the Wesabe API and Firefox extension releases different and interesting.

• First, we've gotten a lot of interest in the Wesabe API in large part because the bank and credit card industries are so tight-fisted with their data. This is why I called my blog post on the API release, "Your bank has a REST API now (shhh! — don’t tell them)." Other companies have done well with their APIs by owning a large set of data and letting people at it through the web; we're doing well by liberating data from the Phantom Zone of the bank web sites, and making it available to the people who already own it -- the banks' customers. It's easy for us to offer value to our API users since the companies that currently store financial data do such a fantastic job of putting up barbed wire around it, in the form of archive access fees, download fees, obsolete data formats, and just plain bad programming. Making all of that data available, consistent, and free is value enough.


• Second, we've taken a data model from the financial services world and flipped it on its head -- fighting fire with fire, you could say. In the world of credit reporting, creditors voluntarily report on their experiences with consumers to centralized credit bureaus, and these credit bureaus providing ratings of each consumer, to let other creditors know whether that consumer is a good or bad credit risk for their money. We do exactly the same thing, but in reverse. Consumers voluntarily report on their experiences (using the Firefox extension and other tools) with merchants to us, and then we publish ratings of each merchant to let other consumers know whether that merchant is a good or bad value risk for their money. People like to talk about the web providing individuals with more power through collective knowledge, but I think this is one of the best, concrete examples of how that story could really work.


• Third, it matters a lot less to us that you use our web site than that you use our API. We know that there are plenty of people who prefer to manage their money in Excel than any other way, which is why Excel is one of the formats our API provides. We think the best way for us to provide value is not to reimplement Excel, but instead to collect the most transaction data and put that data to work for consumers, as described above. Instead of just thinking of this as software above the level of a single device, we're thinking of the Wesabe database as valuable to all of our members whether they ever visit our web site or not.


• Finally, by using Firefox as an automation engine, we're avoiding the trap of paying for basic data aggregation, and instead building community tools to make data accumulation a free outgrowth of joining the site. Where Web 1.0 companies would pay some enterprise service for each sync of data, we're using a community-driven approach to make data download fast and easy for everyone -- for free, which lets us provide open APIs and other features, unencumbered.

None of these approaches are unique, but taken together, they offer a lot. The financial services industry has relied on data being hard to get, to bring you back to them for information, and to make it easier to change the fees that make up so much of their revenue. Their web sites, while popular for bill pay and balance checking, certainly reflect a deep reluctance to have anything to do with the web and its users. Exploiting that reluctance is a great model for startups to pursue.

It's fun for me to see other startups going down a similar path in other industries. For instance, Get Satisfaction seems to be taking a related approach with customer service, another industry with a SuperMax approach to data. Free the data and flip the model, and you can make even the stodgiest industry into a web platform participant -- whether they like it or not.