Float is the world’s first personal finance related ecosystem of connected devices to track and manage short and long term finances. This project is an ongoing research collaboration between Carnegie Mellon University and PNC Bank and aims to explore how the use of smart ambient products can give physical presence to digital money and transactions in ways that positively influence personal financial decisions among millennials. The devices that we coded and built are being beta tested in 5 households.
User Research, Usability Testing
Interaction Design, Prototyping and Circuitry
Our target users are earning Millennials who are independent and operate their own finances with spending accounts to their name.
They use online banking and primarily transact digitally or at least have credit/debit cards
They are tech-savvy individuals who are comfortable using smartphones
Millennials are the first generation of digital natives, and their affinity for technology shapes how they shop. A study of existing research showed us that they are used to instant access to price comparisons, product information and peer reviews and rely on these to make purchase decisions.
This has led to millennials not wanting to sacrifice the current quality of their life to account for a better future. As a result, they are not preparing financially for their future.
Our secondary research indicated that 28% of millennials rely on financial apps and online financial tools for financial advice. Initial research into existing apps helped us classify them into categories and analyze their usage. The categories that emerged were
Transfers (like Venmo and Splitwise), investments (like Betterment and Robinhood), personal finance (like Mint and Qapital),
Online shopping (TrackIf and Flipp),
Tax related (TurboTax, Expensify, MileIQ) and
Miscellaneous ones (such as Plaid, AllPoint, Credit Sesame, etc.). We looked into these further.
We also found two internet connected devices in development/ market; ‘Porkfolio’, and ‘Pavlok’.
While personal finance apps are very popular and widely used for budgeting and saving, the market for these types of apps is already heavily saturated.
Based on our secondary research, three focus different areas that affect personal finance emerged; impulse purchase, budget tracking, and long-term investment - and we sought to explore these further.
In order to improve overall financial well-being, we realized that each of these focus areas would require device intervention at different points in time, and different decision making time-frames. As a result, we decided to research, build, and test an eco-system of 3 connected devices.
Next, we conducted a survey with 35 millennials who fit the target group. Respondents were asked about their budgeting practices, impulse purchase tendencies, and long-term financial plans.
We collected demographic information to identify interview participants and assessed the need for these devices through our survey questions.
We conducted in-person interviews with 8 millennials found through convenience sampling to understand their thoughts and practices with regard to personal finance.
As an introduction to the devices we plan to develop, we also showed them the framework and flow diagram of the 3 devices (for the 3 focus areas) to gauge their interest and learn about what millennials feel about the use of IoT devices for financial tracking and stability.
All the participants struggled to understand what an ambient device to help with finances might look like. They kept gravitating to the concept of a screen based interface and could not visualize or imagine a physical representation of finances.
Most of the interviewees did not use any tool for explicitly for budgeting, they prefer keeping track of their spending and budget in their mind and always have a general idea of how they are doing currently to knowing exact dollar and cent amounts at every point in time. While four of them did admit to trying Mint for some time, only two of them still use it. However, even the two Mint users claimed that they still track their budget in their mind.
The definition of “impulse purchase” differed greatly depending on the individual. One considered his/her stock purchase as an impulse purchase while another considered a day trip as one. Four interviewees mentioned that their most frequent impulse purchases were food and beverages. On the other hand, another mentioned that she never makes any impulse purchases.
Almost all interviewees felt that they would like to have an ambient device that sat in their environment to inform them about the current state of their finances, since they felt it would be more glanceable, convenient, and easy to access as compared to an app where they have to consciously log in and check every time. However, they had concerns about the security of their data and privacy in the presence of guests.
Overall, all users were concerned about how secure their data would be if they used a device that was constantly interacting with their online banking services. They wanted to know if it would be easy for someone to hack into the device and steal their account details or their financial statements.
Another concern that was raised was that the device which tracks impulse purchases may become a nuisance, or that users could become conditioned to it. At least two people compared it to becoming a snooze button on an alarm clock.
Users being wary of any negative reinforcement that the devices may employ if they are not on track was another common theme that emerged. In some cases, this included the notifications on the impulse spending device.
A speed-dating session was conducted with 32 participants (who are all acquainted with the development of IoT devices) for ideas for personal finance related connected devices. Next, we went through these ideas as a team, refined them and brainstormed more IoT device concepts that would alert users about their current financial status and changes in real-time. We classified all these ideas based on the type of notification or indication of change that they would provide.
Prototypes for all 3 devices were made using the Particle Photon micro-controller and basic electronics. We also paid attention to the form factor so that the participants would be able to visualize such a device in their home or on their person. Initially, we explored devices which looked like practical decorative items, like wind chimes and photo frames, as well as curious objects like a cloud in a bell jar or a light-box. We wanted to understand user’s feedback with respect to appearance, utility, and data privacy and security.
We looked at various form factors - like practical items vs curious objects, and also explored different types of interactions - like light based vs glanceable visual cues.
We built and tested prototypes that rewarded good behavior (image 1- candy dispensing device) and ones that show the status of amount spent on impulse purchases on that day (stress ball, color changing test tubes and thermochromic octopus). We also looked into friendly competition and gamification to limit impulse buys.
These are some of our findings and the resulting decisions:
Participants were excited about curious objects as ambient finance notifiers, rather than decorative items like wind-chimes. They felt that curious objects, like the cloud in a bell jar, would fit into any house with any decor or interior design scheme, and add a feeling of intrigue, unlike photo-frames.
The glanceable nature of the status had positive reactions from users. They did not want to see exact dollar amounts, nor did they want to have to log in to an app to check their status every time.
The curious object form factor of devices 1 and 2 convinced users of data privacy when they had guests over. But, they wondered what would happen if their guests also had the same devices and knew exactly how to interpret the data.
Users were still concerned about data security. We were conscious about data security while choosing services to create the Float ecosystem, we made sure to use reputed services like Plaid for APIs and to provide 128-bit encryption.
The video below demonstrates the initial concept of how Float, as an eco-system, would function. The 3 devices communicate with each other through an online platform to help the user keep track of their finances in real-time.
The devices - Sola, Flora, and Aqua, have been described below.
Sola helps the user track their budget. It is a curious object and is kept at home. The color of the cloud indicates the current budget level and the balloons indicate what the user has spent their money on.
The height of the balloons are controlled by servo motors that wind and unwind wire based on amount spent by users on certain types of items or vendors (like Starbucks or Uber) that they have identified and selected to control.
The user knows how close to their savings goal they are when they place a representative token (can be custom-made in the shape of a house, car etc.) in Flora. The flower petals open and light up based on how much the user has saved towards that goal. The mechanical interface adds an element of wonder.
The images below also show the basic functional online interface for Flora and the resistance values for the tokens.
We finalized the concept for the third device before handing it over to the next batch of student researchers for development, but could not create a polished prototype due to time constraints.
The images show an attachment for a mobile phone case which shows current spending as a colored background for the wave cutout in the front sheet. The level of colored waves indicates how much the user has spent versus how much they’ve saved, or as a competition with their own spending on a randomly chosen previous day.
After finalizing and refining Sola and Flora, 20 of each device were fabricated. These were demonstrated to PNC and prepared for beta testing with users. Representatives from PNC Bank were fascinated by our take on ambient devices for personal finance and excited to see how users might interact with these devices.
Users who fit our target category and lived in Pittsburgh were identified for a trial run of the devices.
The design and fabrication of Aqua was handed over to the new researchers joining the project, along with our fabricated Solas and Floras. They took over the project and proceeded with beta testing and development of Aqua.
Since this project involved a lot of generative research, we did not have any frame of reference or any artifacts to ask users to talk about or compare during our initial interactions with them. They were confused about what ambient finance devices might be, and it caused some frustration among the team about what the next steps might be. We had to change our approach and start prototyping from the very beginning.
I found out that electronics components can be very temperamental, especially when multiple prototypes have to be made. I had to learn to be patient and to take advantage of online resources to help me debug any issues. Wifi problems taught me to always have a backup plan and think out of the box.
Overall, this project opened my eyes to a future where digital products can come out of the confines of a flat 2-dimensional space and move into the real world. Given my engineering background and previous experience with physical products, I look forward to such a future.