After nearly 20 months live with our bank solution we feel confident saying that while Open banking sure has potential, and is a very welcome development, the PSD2 API’s just can’t take the pressure of handling transactions. We have an obligation to our customers to use the integration technique that converts the best, and that is what we will keep doing.

First things first, what is open banking? 

Feel free to skip ahead if you’re already familiar, we won’t be breaking any new ground here. 

Open banking is the practice of sharing financial information electronically, securely, and only under conditions that customers approve of. Banks are expected to develop APIs that give access to financial information efficiently, which promotes development of new apps and services. Ideally, open banking should create a better experience for consumers.
You may already use services that utilize open banking — PFM tools like Minna and Tink use your bank account information to help you track spending and reach other goals.

Sounds good right? Well, as it almost always is with these things: both yes and no. In a perfect world, we’d build our solution entirely on PSD2 connections. Then again, in a perfect world, the conversion for those connections wouldn’t be crap. Which it, more often than not, is.

The Open banking conundrum explained

We’ve built our platform with a focus on conversion and we are testing open banking all the time – and it is certainly interesting. However, for many banks, the PSD2-API’s just don’t deliver the quality that we expect. It works wonders for account aggregation and PFM-services where the connection happens in the background while the end user does other things in the app. But for transactions, something that the end user expects to be immediate and low effort, conversion will take too much of a hit. 


That being said, some API’s are better than others and for any market where PSD2 is a requirement- we are using it. For any market where you have the flexibility, we will always aim to use PSD2 connections first — but we have built a powerful platform on which we continuously test every technical solution to make a deposit with a specific bank and will use the one that converts best. We can see the patterns per bank, which kind of integration works best, and we have a solid fallback structure per technique. 


But enough bragging! Let’s have a look at the worst offenders and why we think a pure Open Banking solution is too risky as it is. Keep in mind that this is on a general level.

No decoupled flow

Some PSD2-API’s don’t support a decoupled flow. So all payments started via open banking will open in one of those cursed pop ups — killing, or at the very least maiming, conversion. We estimate that it affects as much as 10% of deposits.

Slow flows

We have seen PSD2 connections being up to six times slower than our custom solution.

Poor stability

We don’t have any hard numbers on this because we do not dare to test it on a bigger audience. But the preliminary results show that as many as every fifth or sixth call goes unheard.

Limited account information

It’s not always possible to see the name of the account and in some cases not all available accounts get fetched. This affects both conversion and UX. Not knowing if you pay with the salary account or the kids’ college fund could be devastating. 

Don’t fret, it’s not all bad!

That’s not to say that it’s all bad. PSD2-Open banking API in Finland is performing very well on most of the banks, but we still test the alternatives. By comparing techniques per bank, we’ve built a compliant, reliable platform that converts optimally in every given market. We could not do that by only relying on one technique.

Categories: Uncategorized