Creating a saving app which humanizes finance.

Sketch, Invision, Zeplin,
My role
Product Lead/


I was the Product Lead/Manager for Boss, an internally funded product that my consultancy, Cacpo, aimed to build in partnership with a Big 5 bank or credit union. Capco typically does digital transformation work for existing platforms, but in recent years has started investing in some of its own products. I pitched the product to the company's partners after being personally frustrated with the lack of user-friendly saving products available. After it became sponsored, I led the project to completion of a front-end MVP.

Jump to
The problem ↓
The process ↓
The outcome ↓

The Problem

Canadians under 35 typically save only a fraction of their income, leading to stress and pessimism


average annual household saving rate - Statistics Canada

At the same time, there is growing trust gap between customers and big banks.


of people feel that their bank knows them. - Celent


see their bank as a source of credible financial advice.


Existing saving apps typically aren’t informed by research in psychology about how to motivate users without relying on guilt and hand-slapping

Apps like Mint tend to be only minimally gamified, and prompt users to create strict budgets, stressing them out if they go into the red.

The Process

To understand our target users and the current market, we did discovery research in the following areas:

Competitive analysis

From heuristic evaluations of apps like Mint and new players on the market, like Joy, I established where Boss could sit.


I co-ran a workshop with the core team and stakeholders to ideate on distinctiveness, features and benefits.


We surveyed 80 people to gather quantitive and qualitative data investigating habits and emotions related to saving.

Can saving be as fun as playing Mario?

We explored various psychological models to make saving easier for users. One that had a lot of data behind it was the video game model of gamification, popularised by Nintendo.  

It suggested that we should give users bite-sized challenges, the easier the better. We needed to get users to voluntarily accept challenges, and frequently provide rewards. The aim was that these would be AI-driven, personalised challenges.

From pen and 
paper to pixels

We initially fleshed out a Learn section but later decided we didn't need it for the MVP. We returned to the educational piece later on with Insights.

We wanted to create an experience that felt fun and unique to give it that gamey feel. But as we didn't want to reinvent the wheel, we decided to primarily use treatments native to Material Design after early exploration in a few different directions.


We were moving away from certain conventions, but we needed to avoid confusing people

Goals vs. challenges

We knew from guerilla feedback sessions that there was confusion between goals (for instance, paying off your credit card)  and challenges (for instance, spend $50 less on coffee this month).

Lack of onboarding

The original flow first prompted the user to set up the goal (thing they are saving for), and then once they were in the app, they could then add challenges towards that goal. There was potentially not enough of a walkthrough.

Mental model shift

People were used to setting budgets and tracking how much money they had spent in each category. In our model, we challenged them only to spend less than previous months, making tracking less straightforward.

A/B Testing

From this we created hypothesis that an alternative flow: An end-to-end onboarding which included both setting up goals and adding challenges (little ways to save money towards the goal), would make things clearer. We created the flow and did A/B testing to determine which option users preferred.

Version 1: Setting up challenges separately from onboarding.
Version 2: Challenges incorporated into onboarding as one flow.

A/B testing findings


A visual walkthrough

In our initial pitch, we used the metaphor of saving for a house as climbing a massive mountain. We brought back the mountain metaphor in an onboarding walkthrough. The onboarding explained that goals are your destination (the top of the mountain), and challenges are how you get there.

Challenge categories

To improve challenges, we came up with several rules to make sure they resonated with people. The first rule was: “Don't be preachy”. To do this, we gave people a variety of challenges across different categories. Having a range of challenges would prevent them from getting bored from week to week and allow them to make spending cuts in areas that suit them.


Values can seem abstract; they often don't stick in people's heads. If I ask you to imagine five apples, you can easily do it. But if I invite you to imagine 5000 apples, you will struggle to do so.

So we included a humorous graphical breakdown to help users understand the value of money.

Reflection: What I would do differently

I recruited a group of 10+ volunteers who largely contributed to Boss on top of their core projects. We completed the work after-hours and during hackathons, but we had one full-time developer for two months, which greatly assisted with successfully delivering the MVP.

Looking back on this project, there are things I would do differently. Because we were pitching early on and wanted to create visually impressive assets, we prioritized aesthetics over accessibility at times, which led to rework.

We should have introduced a cadence of user testing earlier, even if it was just guerilla testing with people in the building.

Now that we are all working remotely, I wish that I had organized more fun social events with the team to reward everyone for their hard work.