a lot of work, Goldstein says. In addition to the usual focus on scalability and security, he says, Pageonce had to solve three other more unconventional challenges.
The first, which I mentioned above, was the problem of building smart, adaptable connectors to scores of websites for different financial institutions. (Pageonce never used Yodlee, opting to build its own data interchange software.)
The second challenge was knowing how to handle and plan around the rules and exceptions imposed by different financial institutions. Bank A might only allow Pageonce to pull account data between 1 pm and 3 pm, for example, while Bank B might limit Pageonce’s queries to no more than 3,000 per hour. “There are all kinds of constraints,” Goldstein says. “The main difficulty around payments is not actually doing the payments—it’s managing the exceptions.”
Third and finally, there’s the task of monitoring the platform’s performance and responding to alerts. “The system isolates problems that need human interference and people go fix the system on a daily basis,” Goldstein says. “There’s a whole monitoring team, and that’s what they do all day.”
Pageonce was able to marshall its team’s experience in the quality assurance and testing sphere to build a robust financial-data aggregation application that stacks up strongly against Mint.com. But Goldstein doubts that anyone will come along from behind to try it a third time. “Most startups would not want to get into building a complete payment platform—it’s too expensive from a compliance perspective,” he says. “It takes a lot of deep thinking to make it very simple.”
I’m glad they’ve done that thinking. Nobody wants to spend their time worrying where their money is or when their bills are due. Pageonce has both the design sense and the back-end engineering chops to solve those problems.
