The current Insights features in the Cake app will be further expanded and improved in the coming months. Here’s a preview of what we are currently working on.
In our mission to improve the financial well-being of our users, we are offering functionalities to better understand, control and improve your finances. Those features will be further expanded and improved in the coming months, but we want to give you a preview already of what we are currently working on.
The core insights are already available today.
First of all you can get a single overview of all of your transactions from different bank accounts. Most of those are automatically enriched and categorised by our data model. This gives users insight into what they spend, earn and save.
But we also offer the possibility to get monthly overviews of your income, spending and savings. At a single glance, you can see which categories or merchants you spent the most money on.
But there is so much more we can do! Several new insight features have been on our roadmap for quite a while. Some of these are:
- More time controls (e.g. view spending in the year)
- Improved comparison options (e.g. averages, previous period )
- Analysing specific things (e.g. categories, merchant spending)
- See recurring payments and subscriptions
But also other features related to insights are on the planning. Think of:
- Bank account groups
- Peer comparison
- Global search
- Subscription management
These are some mock-ups of what we are working on.
What makes it challenging?
We set out to create a comprehensive and consistent framework that could house all of these powerful features and functionalities. We need to carefully think about how these would work together.
- We need to balance simplicity (make it easy and understandable for all users) with functionality (offer advanced functionalities and detail for power users). This requires us to make some tradeoffs sometimes.
- Ensure consistency: the experience and interaction to consume and explore insights should be consistent across different ‘user journeys’. If we show data about categories in a certain way, we should do it in a similar way for subcategories, merchants or people. This is important because if the user needs to rediscover how something works on every page, they get lost.
- Set priorities: although we need to consider how all pieces fit in the framework, we won’t build and release everything in a single go. That’s why we need to prioritize. E.g. certain advanced proactive insights should only come after delivering general reporting functionalities.
Principles of good insights
So what are our principles to provide good insights?
- Show a complete picture:
For good insights we need to take every expenditure into account. We are working on the implementation of meal vouchers, paypal and others… Especially meal vouchers, since these are currently not shown in the data, although they consist of a significant part of the household budget. What makes this element challenging is the dependency on other parties. But rest assured, we know this is a critical component of insights and work hard on it!
- Make the data accurate:
The data team has already shown it’s powerful abilities to make the data meaningful through enrichment, categorisation, pattern detection and more. However, this needs to improve even more to get to the perfect insights. We see features like manual transaction enrichment, marking undetected transfer, nullifying (linking) transactions and removing outliers as important tools to reach this objective. The key will be in combining forces from the data team with a user’s input and contribution.
- Show additional insights:
By offering more controls and detail, users will be able to understand and explore their data. This is the part we built our insight framework for!
Below are examples of insights we could show for people or merchants. You would now be able to see how much you have spent at a certain store in total or over time. For some people it could be shocking, but it will definitely be insightful!
What did we do?
We know not all users have the same requirements or want the same level of detail. Some users are interested in seeing how much they have spent on electronics at Bol.com in 2020 and how that compares to the spending of their peers. Others just want to know whether or not they are overspending since their last paycheck. We need to offer both in the Cake app.
To do that, we’ll use an approach of multiple layers.
As a first layer we keep the existing Insights tab for a general overview. This is only to consume data and doesn’t require interpretation or interaction from the user. It is very similar to what we currently show in the app.
It’s the second layer that will provide more detail, and allows users to explore, play with their data, answer their questions and make them even better understand their spending.
We plan to introduce 3 important abilities:
- Controlling time (both the period and time frame duration)
E.g. Look at your 2019 spending, or the last 30 days, 12 months, etc.
- Zooming in on specific (sub)categories, merchants or people
E.g. Explore your spending on Food & Drinks, or look specifically for spending at Burger King
- Adding comparison options to put your spending in context
E.g. Compare your current spending to that of last months, or your average, or that of peers
So how did we proceed in concrete terms?
- We listed all the types of insights we could and wanted to show to users.
- We looked for inspiration: sometimes, it’s not necessary to reinvent the wheel. So we can take the best parts of many wheels. This can be from our own previous designs, direct competition, other industries (e.g. health apps), or even sites like Dribble.
- We listed the user objectives: figuring out what a user wants to obtain from insights. We have an existing app with an existing user base, and have quite some direct contact with users through support and the roadmap. That gives us a good feeling of what users want and need.
- Translate all of this into a framework with different ‘layers’ of insight. For this we consider the ‘questions a user wants to answer’, design constraints (e.g. first layer can’t require interpretation or analysis, just consuming), and a long list of things to validate or discuss.
- We start the loop of wireframing, checking with the team, users, our Inspiration Board and of course: Friends, Family and Fools. We have regular sessions with diverse members in the team to discuss, get feedback, point out inconsistencies. We also set up calls or conversations with Cake users to get feedback, let them test out, even if it’s only with a rudimentary wireframe prototype! That’s a nice thing about Cake, no secrecy! It prevents us from heading in the wrong direction early on.
The tools we use allow for rapid prototyping and testing. By making our designs in Sketch and syncing them to Marvel, we can have interactive wireframe prototypes without a single line of code. You can even test it on a phone as if it were a real app. That makes it much easier to get tangible feedback. Throughout the design process of insights, I’ve regularly taken out my phone and did an impromptu user test with friends. Looking at where they get stuck or what actions they take is a perfect way to improve the design.
Some things we needed to keep in mind throughout designing: What is the action a user wants to take? How can we avoid choice or information overload? When are functionalities niche, or for the masses? What functions or interactions are too hidden? Where is the current state unclear? Where is it unclear what the next actions a user can take?
Why am I really enjoying this?
I’ve never done sculpting, but I imagine it feels the same as designing the insights! It’s a lovely feeling! 😁 You start with a rough, big block of marble. Then you start crafting. First the big parts. Then it gets in shape. You refine it. You look from different angles, and refine it some more. You keep refining until you see that things just make sense.
Also, I like how we approach it as a team. It’s not a solo assignment that one person is working on for months in a bunker. We continuously sit together with the team and users to evaluate and make changes. It’s a team effort with close collaboration of the user support, development, data analysts and of course designers. It allows for early pivots and better refining. And it definitely results in a better sculpture than if one person would do it without input from anyone else.
A last thing I really like is that it is so tangible. It’s like a big puzzle that needs to be solved. We have an existing app, with existing data, merchants, categories, existing and future features. Our goal is to make everything fit together into a consistent, easy-to-use, powerful and pretty design.
The screens below tell a bit more about the framework. Whether it’s focussing on a specific category like food & drinks, looking at a general overview or comparing your spending to a year ago. It will all be possible! Of course, we will work in multiple stages. That way we can gradually drop new functionalities to make insights more powerful, useful and complete!
When can we expect this?
We expect to add these new features to the app later this year to give users an even better understanding of their financial behavior. After all, better financial well-being starts with having the right insights!
In the meantime: we need your feedback!
We are in continuous development of these and other features. There is still time to adjust and make changes on the design of insights! That’s why we want to do a warm shoutout to users who want to give their opinion!
We have made a detailed clickable prototype so you can play around with the new framework just as if it were a real app!
If you want to take a sneak peak and give us some feedback while doing so, give me a shout at firstname.lastname@example.org or via our chat function in the app.
After all, the better we can make it, the more you are going to enjoy it when it’s live! You know where to find me!
You might like to read this too