UX case study: Smart transactions and spending breakdowns using location data
At Monese, we want people to bank freely wherever they are without all of the annoying, sometimes impossible, bureaucracy. That’s the reason we devoted a great deal of engineering power to build a unique KYC process that allows us to automatically onboard users that typically struggle to verify their identity with banks after moving to a new country. We are providing answers for some really fundamental needs: the ability to receive a salary, pay rent or use a debit card. That’s the main reason our accounts are so popular.
We want those users to stay with us even after they have settled down, so our second goal is to make the best banking service out there. It has to be simple to use and provide all the basic banking features. It must be available everywhere, affordable and transparent. It should also provide additional value, which is why we are always trying to understand our users better by reading every single review and conducting interviews, surveys and user testing sessions.
The one thing that we hear repeatedly is that customers want more from their Monese account than just the ability to use their card or send and receive payments. They want us to help them secure their financial future — they need to build their credit score (we are on it), open savings accounts (it’s on our road map too) and provide spending management features with insightful information.
Mostly, users would say things like “I want to know more about changes in my bills” or “it’s hard for me to understand old transactions because I don’t remember them”. Sometimes, users will tell stories that seem like a cry for help — like that one user that told us he downloads the statement in a CSV format, loads it into excel and analyses the data on his own; or another user that told us she was using a different service for her TFL rides only, just so she could easily see how much she spends on transport each month. It was obvious that our users are in need of a tool that will breakdown their spendings.
Introducing: smart transactions
For the spending management features to be understandable, we had to get the clearest data that we can extract out of the raw, original transaction data. Using location databases and online resources, we are able to match the name of the merchant, the address and a lot of other details. In addition, we built a merchant service in-house, so we can show the right logo for most merchants and treat chains and stand alone businesses differently. The result was a drastically improved look and feel, with richer, smarter insights.
Smart transactions are more than just a visual improvement. They are also solving a real issue that our users keep bringing up: in many cases, they just don’t recognise their own transactions. One of the reasons for this issue is that most of our users are living and working in a country that’s new them, so naturally, a lot of the locations are not familiar. Remembering names is hard enough, especially when the name is not in your native language and when it makes no real sense, like “TescoPfs 5608”.
The new smart transaction can show the location’s map, a logo, some basic stats, a spending category and a clean name that can be recognised. We took an extra step and included additional information that is available with a single tap.
To be honest, the location information screen wasn’t validated in the research we did prior to building the feature, but we had a hunch that this could be really useful — I mean, it turns your transaction list into an automated timeline that provides access to the places you visit. We think it’s pretty awesome.
When we user-tested it, customers were keen on it and even brought up ways that this can be useful for them, without us asking anything about it. “If I forgot my coat in a bar, I can see exactly what time the bar will be open again”, said one user, while another said “That’s nice, I can easily find a place I liked and share it with my boyfriend”.
Obviously, user testing isn’t the holy grail when it comes to identifying if a specific feature will be used and liked. As it is better to have a bigger sample group, we added extra usage tracking on this section, to learn how people use it and how we can improve it in the future.
Sneak peak: Spending
Now that the spending information is cleaner, it’s much easier to use the data in a broader way, break it down and present valuable information. Since we are not the first company to offer spending features, we didn’t want to just do what others are doing. We wanted to create the best solution, one that will fix the pain points of other services.
Available spending management tools can be roughly divided into two groups:
- Third party services that can be connected to your account
- Lists and graphs that are provided within a specific bank app or website
Some of the third party tools are actually pretty great. They can be insightful and well crafted. However, they don’t appeal to most of our users. For a start, the users need to be aware of the service; Then, they should care and be tech-savvy enough to download a third party app. After passing these two obstacles, they need to allow permissions access to their bank account to a company they know almost nothing about. That final bit seemed to be a huge holdback for our users.
Current spending management features within banking apps have a different batch of issues. Yes, they are accessible and the user doesn’t need to do anything to get them, but they tend to be underwhelming and limited. Most of them are just dry lists of spending categories, while some have graphs with cool interactions that don’t provide any real additional value.
According to the interviews we conducted with users of some of these banks, they hardly ever look at their spending breakdowns. Most of them said that it’s really nice to have this information, but they either forget it’s there or don’t think that viewing it more than once every couple of months is useful. Even the users that said that they love their bank’s spending breakdowns, often admitted that it has been a few months since they actually opened it.
In addition, people kept complaining that the data they receive from their bank is inaccurate — categorisation is a very difficult task, and a lot of times it’s just wrong. Take M&S as an example — they sell so many different things. Should the category for it be “Shopping”, “Groceries” or “Eating out”?
Most of the users we interviewed or tested loved the idea of categorisation and thought that it could be incredibly constructive, but they imagined that the categorisation will happen automatically without too many errors. Since most of the information that can be extracted is about the actual merchant and not about what the user has payed for, accurate categorisation requires some manual work from the users. Unfortunately, most of our users are busy.
That’s why the first thing we tried to tackle was the categorisation problem. We realised that we will have the same issues and even though we will try to get the data to be as accurate as possible and continuously improve our database, it will never be perfect. That why categorisation, as essential as it may be, shouldn’t be the only approach.
Looking at our usage data made it apparent that most people don’t often shop for groceries at multiple stores and that in general, they tend to use the same merchants. By this logic, breaking down spending via merchant can be a much more accurate metric for helping the users make sense of their patterns.
Solving the feature stickiness was also important. For users to actually be on top of their financial game, they need to constantly be aware of their spendings and earnings. Showing clean information isn’t enough, we need to add some fun into the mix and make the whole thing engaging and interactive.
We took a lot of inspiration from music charts. Not just because “top lists” are amazing but also because they are built to keep you engaged and excited. A music chart, just like our merchant chart, is always updating and can point out changes and trends.
You can actually learn about your behaviour by seeing that this month, Pret A Manger dropped two slots and made room for your new Adidas shoes, while your Deliveroo monthly total is almost half of your monthly rent. If there is a new entry on your list, we will highlight it with a little cute chick, just so you’ll notice how your new Playstation obsession might have gone too far.
There is also a new interactive graph that sits on top of the chart and shows how much money is coming in versus how much money is going out. You can view it next to your spending breakdown, to better understand how much money you can spend and what exactly you are spending it on.
We still believe in the classic categories breakdown. It may have some issues but we found that it can be really helpful — when it’s accurate, it’s the simplest way to get a quick overview, and if some categories are wrong, it’s super easy to fix by choosing a different category.
Users that would like to take control over their funds can set a budget for each category and for the entire account within seconds, see their progress in a simplified graph, while getting notified every time they are nearing their budget or going over it.
We tested all of these spending features with some of our current users (and a lot of Monese employees) and the responses we got were the best we’ve seen so far in our testing sessions. It’s not just that everything was clear and people “got” the interface in a fluent and intuitive way — the part that made us thrilled was how every one of the users showed excitement and made it clear that they can’t wait to see all of these features in the app already. Obviously, we feel the same.
Smart transactions are now live for all our users on Android and iOS. The rest of the spending features will be rolling out during 2018 and will likely evolve after further user testing.
This article was written by Alon Uziel, Lead UX Architect @Monese. This is the forth post in a weekly recurring series from our Product & Engineering team.
