Screen Scraping Vs. Bank APIs: What’s the Difference?

Brandon Dewitt, co-founder and former CTO at MX, was prominently featured in an article from Bill Streeter at The Financial Brand. The article, titled “Open Banking Gets a Big Jump Start from COVID," explores the ways in which open banking (or possibly open finance, depending on which term you prefer)will become more and more standard in the industry. 

Dewitt postulated that the number of financial institutions offering an open banking portal will likely climb from 20 to around 200 by a year from now. Why? Because most of the biggest brands in the industry have already started implementing these portals and because many other financial institutions tend to be fast followers.

“With many smaller community institutions I’ve spoken with this is certainly on their horizon,” Dewitt stated. “I believe that open banking is a major part of the future of serving their community as these communities become technically more proficient.”

This transition will require financial institutions to have a clear understanding of the differences between screen scraping, which is the traditional way data has been shared, and bank APIs, which are central to open banking.

So, what’s the difference between screen scraping and APIs? And how does the evolution from one to other lead the industry to Open Banking?

Screen Scraping

Screen scraping is the process of gathering data from one app by inputting user credentials (such as username and password) and displaying that data somewhere else. Scraping is the foundation of data access today, largely because it allows technology companies to choose which data fields they want to obtain and doesn’t require having an official relationship with the financial institutions they scrape data from. By contrast, sanctioned API channels allow financial institutions to limit the fields they want to share, which can result in consumers losing out on the ability to access the data that’s most useful to them in an aggregated experience.

The biggest downside to screen scraping for fintechs is that the process relies on the scraped institution’s website structure, which is particularly problematic when scraping happens without coordination with financial institutions. In practice, this means that fintech companies that scrape data have to constantly fix connectivity issues resulting from web updates. It also means that website downtime results in loss of connectivity. In addition, screen scraping results in slower connections compared to APIs.

As Brandon Dewitt said, “The only reason that fintechs screen scrape is because it is the only path for them to get at that data. Once you offer a more reliable, more secure and faster path, I think they’ll abandon it overnight.”

The biggest downside to screen scraping for financial institutions (and the reason it has become a dirty term in some circles) is that financial institutions aren’t always aware of who is scraping their data or for what end. Worst of all, these institutions are responsible to regulators for data breaches that occur from scraped data.

Scraping can also be confusing to security teams at financial institutions since it’s really difficult to decipher the difference between that activity and possible malicious activity. Finally, credential based access via screen scraping isn’t ideal since that information does have a chance of being intercepted.

APIs

Broadly speaking, there are open standard APIs and proprietary APIs:

Open Standard APIs

There are two standards worth exploring here: Open Financial Exchange (OFX) and Financial Data Exchange (FDX). OFX access has had some great adoption (more than 7,000 financial institutions, according to their site). However, OFX also has some downsides: It isn’t regularly maintained, past versions vary considerably, it requires relationships with each financial services company fintechs connect to, and data can be incorrect or missing. Sometimes old or incomplete integrations lead to getting data for the wrong field returned in OFX feeds.

FDX is a newer standard that has gained traction in recent years, such that OFX and FDX announced that they’re working together toward a single standard. This is almost certain to be the future of open banking.

Proprietary APIs

Proprietary connections share a lot of the same upsides as OFX and FDX, particularly when it comes to reliability, speed, and consistency. The biggest difference is that the best proprietary connections are generally better maintained because the focus is more niche and there’s typically a bigger market incentive behind them. The downsides are: Proprietary connections aren’t common (though that’s quickly changing), the data can be limited in instances where financial institutions restrict the number of accessible fields, and the connections can be turned off in the event a financial institution decides to pivot.

***

Given that screen scraping is often frustrating to end users and that financial institutions worry about potential security risks, there are many reasons to look forward to widespread APIs in banking. A mix of open standard and proprietary APIs represent the future of the industry — a future defined as open banking.

To learn more about open banking, see our Ultimate Guide to Open Banking.