NOTES AND METHOLOGOY

Our approach to calculating adoption

There are inherent difficulties in tracking and monitoring the adoption of an underlying technology which many consumers and small businesses do not fully understand. We recognise that there are always going to be weaknesses with any approach but believe that the methodology we have adopted is the most robust globally as it is based on data provided by the CMA9.

However, for full transparency we set out here why we believe these numbers are robust, as well as their limitations.

The CMA9 banks provide monthly data on the number of customers who have shared data via an open banking API in the last month or made an open banking-initiated payment. Several of the banks also provide data on the number of online accounts they have, which allows us to estimate the total number of digital accounts.

To create the “1 in n” calculation in the Adoption Section of this report, we divide the total number of open banking active accounts into our estimate of the total number of digital accounts. This could both overstate and understate adoption levels.

There is an inherent challenge of overstating adoption because of duplication in the data, as a single consumer with a bank account at two different CMA9 banks would be recorded twice. To address this, we focus on dividing the total open banking active accounts into the total digital accounts, so we partially account for this potential double-counting.

We also make it clear that our adoption figures are for digital consumers only, as a consumer without digital access to their account could never be an open banking customer.

There is, however, also a risk of undercounting, as we have no data from non-CMA9 banks, which could potentially have higher levels of open banking usage13, but also a high risk of duplication as many consumers use a neo-bank account alongside an account from a traditional bank14.

The definition of “active” used is also quite tight, and there could be consumers who use open banking more periodically, who may be active in some months of the year and not others. A quarterly or annual active definition could see the adoption rate rise significantly, for example.

Many markets report open banking adoption rates through consumer surveys, an approach which is inherently unreliable and tends to result in significant misreporting of adoption.

Notes

Unless otherwise noted, UK data comes from OBL analysis of the CMA9 ASPSP submissions, corroborated against BEIS, ONS, Eurostat and UK Finance.

Penetration data combines CMA9 ASPSP submissions of active user numbers (active defined as active in the last month) and digitally active end-users by the CMA9. It therefore only represents penetration within the CMA9 digital customer bases, although this is a significant share of the UK population. This data has been corroborated against data on the total adults and small businesses currently using internet banking, derived from ONS, Eurostat and UK Finance.

You can see the latest raw data on the OBL website here.

[13] For example, we note that Monzo (which is not a CMA9 bank) reports that 6.1m of its 9.0m customers have used open banking. See here. This is likely to be on a different basis to the definition of open banking active we use in this report, but it indicates that neo-banks could well have higher adoption rates, although many of these could also be active already via another account held with a CMA9 bank.

[14] For example, the FCA noted in 2022 that, “Relative to the major banks, a smaller proportion of the digital challengers’ PCAs are main accounts”. See here.

www.openbanking.org.uk Registered in England. Company Number: 10440081 Registered office: Suite 501, Dawson House, 5 Jewry Street, Aldgate, London EC3N 2EX