In the last part of this series, we looked at why analytics teams can have a hard time analysing Open Banking data. We discussed why the challenge is not just the technical complexity of the API-driven data, but also the scale.
New rows of data are created whenever customers or third party providers (TPPs) take any action. With over two million customers now using Open Banking in the UK, banks could easily find that they have over a billion rows of Open Banking data to bring into their analysis.
We also talked about some of the potential benefits that await for banks that do decide to tackle the data challenges head-on. Whether the goal is to understand ‘share of wallet’, improve the customer experience, identify vulnerable audiences, or simply minimise the time spent on the OBIE’s monthly regulatory reports, there’s much to be gained from systematic preparation, automation and analysis of Open Banking data.
At Optima, we’ve been working with Open Banking data since 2017. To support other banks on their journeys, we’ve codified our experience of analysing Open Banking data into a five-stage maturity model, with each stage representing a distinct level of data and analytics sophistication. We’ve seen first-hand that the further banks take the analysis – i.e. the more mature their Open Banking analytics function – the more value Open Banking data will deliver for customers and the business.
The purpose of creating this model is both to help banks gauge where they are on the maturity model and to showcase the potential of this unique and valuable data source.
The first stage is mandatory. The CMA9 banks are required to provide the Open Banking Implementation Entity (OBIE) with monthly reports. These reports are less about customer insight and more about technical compliance. The OBIE’s reports must be supplied in a pre-defined Excel format, and the information sought is largely around confirming the availability and accuracy of the banks’ Open Banking APIs. Recently the OBIE has also started requesting basic customer usage statistics in these reports.
Completing these reports can be a time-consuming manual task. Expect considerable repetition of processes around data gathering and completing the reports, and very little true customer insight.
With thoughtful data management and transformation processes underpinning the analysis, banks should be looking to remove manual handling from the end-to-end reporting process. Which leads us to stage 2…
This stage is focused on automating the OBIE’s monthly reports, thereby reducing the time investment from analytics departments from several days per month to next to nothing.
Automating the completion of the OBIE’s monthly reports requires building data pipelines into a central data warehouse or Customer Data Platform. With these pipelines and data transformations in place, it becomes possible to automate the production of reports, allowing the relevant data to flow directly into the Excel template at the touch of a button.
Automation is not the only benefit of proper data engineering, however. With the data nicely formatted, transformed into a usable condition and stored in a central warehouse, banks can also begin analysing Open Banking data for customer and business benefit.
This is where things start to get interesting. Stage 3 of the Open Banking Analytics Maturity Model signifies the start of discretionary reporting and the point at which banks can start to gather valuable customer insight.
The first areas to make sense of are typically around Open Banking adoption and usage. How many of our customers are using Open Banking? Which segments do they fall into? What products are they sharing their data on? How many customers are renewing their data sharing consents when they expire? Answering these questions will highlight customer trends around the bank’s Open Banking users, and will highlight significant changes over time.
More valuable still, banks can infer meaning from how their customers are using Open Banking. Insight at this level will answer questions such as: Which third party providers (TPPs) are our customers using? Which are most popular? How many are using aggregator apps? How many are also using other CMA9 banks’ Open Banking offerings? Much can be learned from this level of insight. The majority of Open Banking apps are incredibly niche, so customers using them are indirectly highlighting a need for a service that they’re not getting – or not getting easily enough – as an existing customer of your bank.
Where banks are themselves providing aggregator apps that use Open Banking, for example, it can be instructive to see whether you have higher penetration than competitor aggregator apps within your own customer base. If not, why not? What can we learn from Open Banking data when we see customers signing up for an app but not reconsenting?
At this stage, we would also expect to see discretionary reports that provide detail around the balance of ‘outbound’ and ‘inbound’ Open Banking usage. By combining outbound and inbound data, analysis can provide a rich and detailed view of a customer across their financial relationships. This picture potentially allows banks to identify customers’ ‘front of wallet’ applications and relative ‘share of wallet’ for customers across financial providers. Of course, as with all activities involving Open Banking data, this type of analysis is only acceptable with the correct data permissions in place.
Data insight is only valuable if it helps people to make good decisions. This stage of the Open Banking Analytics Maturity Model is about unearthing insight that can drive action. Organisations reach this stage when they start using insight from Open Banking data to make strategic choices about where to invest, which improvements to make to their service offerings, how to better serve particular customer groups, and so on.
Open Banking data can support these decisions, in part by helping banks understand how customers’ relationships change once they start using Open Banking. Do customers using particular TPPs start engaging with us less? Does Open Banking usage lead to trends around attrition factors?
Once you have a clear view of which customers are using which TPPs, it allows you to spot trends and take action before the threats become significant. That may include using Open Banking insight to feed into innovation hubs, or to commission research to learn more about why particular apps are proving so popular with your customers. Open Banking data can also tell you whether you should be worried about the impact of challenger banks on your customer portfolio. Do customers who start sharing data with Monzo show any significant drop in value after 12 months? If so, what can we do about it?
It’s at this stage that we see banks starting to visualise all of their Open Banking data using advanced reporting tools such as Tableau or PowerBI, with dashboards allowing for slice and dice analytics to monitor important trends, allow for deep dive analysis, and to support sound strategic decisions, both offensive and defensive.
The final stage in our Open Banking Analytics Maturity Model is where banks transition from insight to action.
The data generated by the Open Banking APIs is now being systematically transformed and stored in a secure, enterprise-level data environment. Advanced dashboards are in place to visualise the data in near real-time, and to allow you to cross-reference Open Banking data with other data sources. You’re aware of trends, of which TPPs are most popular with your customers and why, and you understand the impact of Open Banking on overall customer value.
You are now in a position to take action.
That action can take many forms. It may be improving your existing Open Banking application to provide new features, replicating (or improving upon) those offered by the leading TPPs within your customer base. It may be a new customer contact strategy specifically for vulnerable audiences, introducing a pilot programme to offer tailored support. It may be proactively communicating with certain segments to inform them of existing features that are being under-used.
It’s important to note that Open Banking originated as a means of giving customers more control of their data. As with all data sharing arrangements between customers and organisations, strict legal obligations apply to the nature and purpose of data processing that banks can perform on Open Banking data. In conducting any analysis of Open Banking data, banks therefore need to be hyper-vigilant that they are doing so in good faith and in accordance with customer consents, privacy policies and relevant legislation. Such caution is required throughout all stages of Optima's Open Banking Analytics Maturity Model, and becomes even more critical in Stage 4 (Strategy) and Stage 5 (Action).
Open Banking data offers banks a new and unique lens on customer behaviour. We’ve seen first-hand that this vast, complex data source needs a high level of transformation and preparation by skilled data engineers to make it usable. But once that work is done, the stage is set for analysts to do what they do best. Provide insights that lead to action. And in the case of Open Banking, the data is a veritable gold mine for the banks brave enough to tackle it head-on.
Our new data science report, The Ambitious Marketer's Guide to Customer Retention Analytics, discusses three practical ways marketers can use analytics to retain more customers. Download your free copy now.Get the Free Report