Discovery Research in the world of Open Banking

Nicolas frankcom
Sep 16, 2020 · 7 min read

Imagine you’ve just joined a new company, and your first project involves a whole new proposition, using technology you don’t understand, and following industry standards that had only just been launched.

That was the challenge facing me when I joined CompareTheMarket (CTM) as a fresh-faced researcher and it’s fair to say I was excited, and terrified, in equal measure.

This wasn’t completely new territory for me luckily. I had a background in financial services products, and knew my FCRs from APRs, but as I quickly discovered the world of Open Banking is a whole separate animal, full of potential customer value, but also jam-packed with comprehension and security challenges that could put even the most financially-savvy customers off.

Open Banking 101

So, let’s start at the top. What is Open Banking? Well, there’s lots of aspects to it, but we’ll focus on the most important part for our purposes — giving customers the power to securely share their transaction history without compromising their data.

In layman’s terms its a regulatory requirement for banks to be a bit more open about the information they hold on their customers.

In a pre-open banking world the banks could see all your spending history, which meant it was locked away securely, but also made it very difficult for people who weren’t your bank to access that information and help you make informed decisions using that information. This ‘opening up’ of data is under the user’s control at all times, is done securely, and only allows third parties to see transactions, not do anything with them.

For instance, a third party that has access to your banking information can now tell you about your spending habits and flag up when you’ve gone over-budget, or help you save towards a certain goal, or flag up that regular payment you’ve been making for 6 months but forgot about.

Crucially your bank could already do this, but, sadly, a lot of the time they weren’t, or they weren’t doing a good job of it. Open Banking aims, at least in part, to make the banking industry more competitive, and to make your money — well, the information about your money — work harder.

Sounds amazing……and scary?

When I first got my head around all this it became pretty clear early-doors that we’d have to do a lot of research to make sure we really understood our customers and how they might react to something as big as this, so we started off with a research plan.

There were a few obvious research questions that immediately came to mind, and almost just as many challenges. The research questions were centred around things like:

  • Value: What sort of things are people looking to do with their existing banking data, and what goals are they looking to achieve?
  • Pains: What sort of issues do they hit when trying to achieve their banking needs?
  • Comprehension: Do users understand such a complex topic ?
  • Trust: If they understand it, do they have enough trust in us to open up their banking data when years of media have, rightly, been telling not to share their data with 3rd parties?
  • Brand: What role does the CTM brand play in all this? How will this interact with their broader CTM experience, particularly our Meerkat Meals and Movies rewards propositions?
  • UI/Flow: How will our users react to moving between the CTM ecosystem and their bank? How will they react to on-boarding? How can we make renewal of credentials every 3 months a seamless process?

Research challenges

CTM was already ahead of the game. We’d purchased Bean and had been actively working on using their learnings to develop a proof of concept.

Some earlier discovery work had validated the interest, but the business wanted to flesh it out more, so we put together a research plan.

But there were a couple of clear research challenges that presented themselves as soon as we started thinking about how to look at this issue. For instance:

  • How do we test a proposition that isn’t technically in existence yet?
  • How do we test something as personal as banking data?
  • How can we replicate a UI that goes between ourselves and a huge range of third party providers?
  • How can we determine value for something as broad as, well, your entire financial life?

To tackle these issues we pulled together a research plan, which contained some key requirements that we knew we’d have to use to address some of these challenges.

The plan

We would need to speak directly to a broad range of our customers to understand what people’s existing goals and pains were when it came to managing their money and bills and identify key customer groups:

  • We would have to look at existing solutions that were trying to do something similar, using slightly different technology, and seeing whether we could use any existing services as proxies
  • We would need some high-fidelity prototypes that could accommodate a huge range of scenarios in order to get realistic reactions to security and trust questions, particularly when it came to transitioning between Compare the Market and the banks
  • We would need some detailed lifestyle and spending data from participants we recruited, so we could match the prototype data as closely as possible, and we would need an agency to help us recruit against those criteria.
  • As it was an entirely new proposition, we would need to make sure technical things were working and get some early ‘realtime’ feedback from actual users before we launched it into the wild

What we did

  • 40 open interviews with participants
  • A diary study using these competitors as a proxy, so we could get an understanding of longitudinal usage and interest
  • An extremely high fidelity prototype mocked up by our excellent and highly Axure-proficient in house designer Jason McCarthy
  • Journey mapping exercises to replicate a ‘day in the life’ and propositional testing so we could build up a matrix
  • Quant testing of messaging variants to gauge appeal
  • Internal alpha testing

All of these techniques combined to give us a great holistic vision of what we need to do.

  • The interviews and journey mapping exercises painted a picture of our customers existing behaviours, goals and pains, and gave us an insight into when during the month we should, and shouldn’t, be talking to them
  • The diary study supplemented this with behaviours based on real-life experiences, taking a full 5-week salary and payment cycle into account
  • The prototype allowed us to quickly test multiple variants, and by collecting participant data in advance of research, we were able to personalise the prototypes per session, to make testing more realistic
  • The quant testing then told us which outputs resonated the most, allowing us to prioritise messaging
  • And finally, the alpha testing not only flagged up any bugs, but also brought back some previous comprehension insights

The result

The Open Banking element of the Meerkat app has now been released and is helping our users keep on top of their bills and regular payments, showing them their spending across multiple accounts, and highlighting potential savings they could make or contracts they could switch.

It’s still early days, and we are continuously getting feedback on where we can make improvements. The team is working hard to expand the list of banks that integrate with the feature.

But don’t just take my word for it, take a look for yourself! The Meerkat app is available through the App and Play store, and even if you haven’t purchased through CompareTheMarket you can still connect your accounts and see what you think!

We’re recruiting

We’re currently expanding our UX team and are looking to hire researchers and designers. Get involved below!

Compare the Market

The people behind and the Meerkat App