Planet of the Apps – A Case for Wholesale Bank APIs

Let’s play a little game of palm reading. As a consumer, you downloaded your first mobile app somewhere between 2008 and 2010. You got it from a trusted online app store. It was dirt cheap and it didn’t require you to have any IT skills. At that time, the app store made you comfortable that the app was secure enough to enter, browse or retrieve fairly personal or professional data.  

The app was pre-configured to exchange data with its server somewhere on the internet so that you could consume the service immediately. The app was very good at a limited number of things (maybe as a video player, an online game, or a document viewer). Then an increasing number of apps became available to choose from. Your behavior shifted in a matter of weeks. You became a picky buyer, downloading the ones that were free of charge, with excellent reviews.

Last psychic reading – after getting your first banking app, you never used your personal card reader to generate a pin code to access your bank account over a desktop computer.

Corporates are moving treasury functions into the cloud

Let’s move back to the world of wholesale banking and focus on corporate treasury. Accounts payables and receivables business processes have not changed much over the last five or six years. However, there has been a major shift in the way those departments rely on technology to manage liquidity, decrease working capital and keep cash flowing where they need it most.

Treasurers are moving away from in-house TMS or enterprise resource planning modules in favor of new tools. These are like the “apps” consumers use for mobile banking; however they are definitely on steroids and enterprise-grade. Virtual accounts, cash pooling, supply chain and trade finance are examples of “cloud app” services a typical Fortune 5000 organisation would leverage from a vendor.

Banks are failing to compete within this market. Their apps or services are usually dedicated to their own brand and do not overlay other banks. Those bank initiatives never gain critical mass. It is far more opportunistic for them to pursue finding a vendor to facilitate, deliver and promote the whole service or app and to derive revenues from it rather than to deliver it on their own.

Financial Institutions fail to publish their services over that same cloud

Banks cannot continue to build bespoke integration setups for each and every single new intermediary they agree to work with: payroll service suppliers, service bureaus, SEPA mandate and e-mandate services, commercial cash management cloud platforms, electronic invoicing networks, e-commerce infrastructures – the list of non-FIs and non-corporates goes on forever. Intermediaries always fit in two categories. It is either a shared revenue business or partnership with the bank, or a successful supplier to the client as a result of the banks it can claim to integrate with. The dynamic is no longer one-sided, where the client used to impose the specific integration and service model to the bank. Gaining scale and keeping costs down means that banks have to stop bending over backwards. It is time to consider publishing access to services in one vanilla manner that is flexible enough to comply with specific business and technical requirements. The answer is something called service-oriented architecture (SOA), commonly known as the world of application programming interfaces (APIs) or web services.

Market adoption and compelling events

Since 2010, there might not be a single enterprise software package that comes without an API capability. Ask any IT department of a reasonable size and the answer is that every single business application exchanges information with other business applications over an API. Any TMS, ERP, middleware or managed file transfer software release supports APIs. There are even industry best practices around APIs, something called SOA governance.

The main reason a bank’s key transaction banking revenues flow over file-based channels is simple: this is the way they’ve been doing it since the magnetic tapes disappeared. Banks stopped collecting magnetic tapes because the file channel enabled clients to transact faster without hours of lag, therefore optimising their treasury operations. The compelling event for a global market adoption of APIs is exactly the same, more than a decade since the last tape was read in a banking mainframe.

A great number of transaction banking executives are taking steps towards rolling out this service model. Most bank mobile apps actually run thanks to those APIs. It is only a matter of time before this becomes a wholesale banking standard. Magnetic tapes and online banking personal card readers belong in a museum. A final psychic reading? The file channel will probably join them within five to seven years.


Related reading