I've been working on an open sourced tool called Bank Rank to track bank account fees, rates, and balance requirements in the United States. For a quick primer, there are 4 broad categories of insured deposit accounts in the U.S.: Checking, Savings, Money Market, and Time Deposit. Banks oftentimes offer multiple tiers of each account, and with over 9,200 banks and credit unions, some quick back-of-the-napkin math means that there are over 70,000 accounts competing for consumers, with rates and fees that can change daily.
I've personally grown frustrated with a number of problems[0] that existing tools have in trying to make finding the "best" deposit account easier for consumers, such as black-box methodologies, SEO/affiliate spam, and incomplete data, which is what inspired me to start this project.
While banks may change rates and fees often, the overall structure of their product sites remains much more consistent, providing an opportunity for a scraping engine to monitor and track this data on behalf of users.
Bank Rank is open-sourced because I can't build all 70k scraping templates by myself, and I would love to see a community come together to expand the reach of this project and make the data more accessible to consumers so they can truly choose the best product for their personal situation without affiliate links, SEO spam, or other biases.
As far as the tech stack goes, the core engine runs on top of playwright and the entire project is built with Typescript. If you're interested in contributing, there's a contribution guide[1] I've written up, and I'm currently working on video guides to make it easier to get started.
I originally started working on this project about a year ago, but wrote the scripts with puppeteer. I have around 700 product scripts which I'm currently migrating to Playwright because I'm more optimistic about the broader support for languages and browsers. Currently, you can pull down the repo and get started tracking products locally[2]. There are only 3 products tracked at the moment, but once I finish migrating those 700 others, I'll be about 1% of the way to total market coverage. I hope that with some additional contributors, we can get close to 100% coverage much sooner than if I keep working on this alone.
I'd love to answer any questions and would gladly appreciate any and all critiques, feedback, or other suggestions. Feel free to drop a note in HN, Github, or send an email to project.bankrank@Gmail.com
Cheers
Links:
[0] https://github.com/project-bankrank/bankrank/tree/main?tab=r...
[1] https://github.com/project-bankrank/bankrank/blob/main/docs/...
[2] https://github.com/project-bankrank/bankrank/tree/main?tab=r...
