Echo: An Extensive Look at Our Research 

 

During the spring semester, we created several core questions to help us understand the Portuguese banking space. We focused on understanding how our users thought about and interacted with their money, what their needs were, and which goals they had. As we uncovered the answers to these questions, we found many different problems users face when they interact with banks. We used a variety of methods to answer the questions we had: interviews, literature reviews, a competitive analysis, photo diary studies, and surveys.

  • What is context? 
    Context Ecosystem Mapping
  • What is already known about the banking market? 
    Literature Review
  • What do users already have and use?
    Survey
  • What are our competitors doing?
    Competitive Analysis
  • How do users think about money?
    Interviews
  • How do our users currently live?
    Photo-Diary Study

We conducted a variety of research methods as we investigated the Portuguese banking ecosystem. Through our methods, we founds many key insights.

Context Ecosystem Mapping

Before we started any of the research, it was important for us to understand what exactly context means for us. We wanted to see how context is connected between people and concepts. 

We defined context as  “The set of circumstances or facts that surround a particular event, situation, or mental state”. These circumstances include using banking applications in various settings and include factors like a person’s mood, intentions, the setting, or the person’s goals. We created lenses which help us frame the context users find themselves in. These were:

  • Time 
  • Availability
  • Age
  • Location
  • Mental State
  • Culture
  • Work
  • Recreation

We concluded that when thinking about what we should build in the future, we should keep in mind that context is very broad and can spread to many different aspects.

16-03-10-elementary-8.jpg

Visualization of the context mapping

Field Observations

There's no better way to know more about people than to go out in the field and observe them. Especially for us as we were in a different country surrounded by a different culture and lifestyle, it was important for us to get to know their behaviors. We would go out into the city center and observe and interact with people. 

SAM_1519.jpg
SAM_1513.jpg
SAM_1515.jpg
SAM_1517.jpg
SAM_1523.jpg

Literature Review 

We examined existing academic and business research to provide a foundation to our research, as well as giving us information we could not gather ourselves. We gathered and analyzed information on spending habits, technology use, creation of financial services, and contextually-aware systems.

Reviewing technology trends and usage habits gave us insights into the context of how people interact with both older technologies, and current technology. Some key insights include:

  • People create their own financial services when the industry doesn’t (US Financial Diaries, 2014; Interview with Cotten, 2016). 
  • In the US, low-income households prioritized “financial stability” over “moving up the income ladder.” (US Financial Diaries, 2013)
  • Trust in technology providers for financial services differs widely per age category (55+ years old being the least trustful) (TransferWise, 2016)
  • Mobile is first (Pew Research, 2012, 2013)

Competitive Analysis 

To help us understand the financial technology business we performed an extensive competitive analysis in 4 different sectors of industry: 3 relating to banking and financial services, with a fourth sector examining contextually-aware services and systems. The competitive analysis gave us a structured look at the market that providing us a better understanding of the banking atmosphere and customer preferences.

We analyzed 34 companies and products examining metrics such as unique features, design strengths and weaknesses, types of users, size of customer base, Apple App Store rating, requirements of use, and our estimated user experience rating. We split the companies and products up into four groups.


TRADITIONAL BANKS

Screen Shot 2017-10-12 at 10.56.06 AM.png

We defined these as banks that have been around for a long time with various traditional banking offers (e.g. personal, business, investment, loans).

  • Slower to innovate
  • Only surface level user interface changes
  • Layer other services over their core

Example: Santander
Banks like Santander offer many products including personal and business banking, investing, credit cards, and insurance.


LAYERS OVER BANKS

Screen Shot 2017-10-12 at 10.57.40 AM.png

These look like banks to the consumer, but rely on partner banks (with core banking infrastructure) to support their services

  • Unique features over existing banking structures
  • Strong focus on user experience
  • Usually younger companies
  • Designs are mobile-first

Example: Simple
To its customers, Simple looks like a bank with helpful, intuitive features like saving in goals and showing a “Safe to Spend” amount. Behind the scenes, Simple uses a partner bank (BBVA) to handle the core-banking backend.


DISTANT-BANKING COMPETITORS

Screen Shot 2017-10-12 at 11.00.40 AM.png

These are services that relate to money, but do not operate as a bank

  • Startups focused on one problem/solution
  • Push the boundaries of  fintech
  • Many use peer-to-peer or social aspects
  • Fewer, but more passionate users
  • pinionated in ways that help the user spend smarter and save more

Example: Square
Setting up payment processing can be hard for small businesses. Square isn’t a bank, but they provide well-designed tools for businesses to accept card payments and manage their stores. A bank is required, but the user mostly interacts through Square.


CONTEXTUALLY-AWARE SERVICES

Screen Shot 2017-10-12 at 11.02.21 AM.png

Companies and products that work with contextual-awareness, but do not primarily offer  financial services

  • Bots and AIs often use meta-data from message or voice-based interactions to gain context
  • Applications that take into consideration of the user’s present location, or other factors that have an effect what the application may offer the user when they first open the application.
  • Open-ended input makes interfacing easy, but it isn’t always clear what is possible

Example: M
Facebook is developing “M”, an artificial intelligence for their Messenger product. M can make purchases and reservations, look up information for you, and stay on hold on your behalf. While M isn’t a direct banking competitor, it might change how we interact with money.


Some Insights

We saw that the user customer size tended to be higher for traditional banks, because they tended to have existed for a longer period of time so they have time to capture the market. We see that Paypal is the most prevalent third party banking application due to its seniority at 17 years old and its early integration to many different services. The next two largest, venmo and mint, are owned by Paypal and intuit, a tax software company, which help them grow their customer base.

 

This graph shows the user size vs. the satisfaction of users vs. apparent ux perceptions. User size is based on how big the circle is: the bigger the circle, the greater the user base size. There are some banks who are starting to learn that user experience is a factor that influences their income, while others are lagging behind. Most of the startups and smaller fintech companies tend to have higher user experience scores.

 

Unbundled Banking

Here is a diagram that I created for a visually better understanding of how banks are represented. Traditional banking tries to represent every type of banking opportunities at once. This tends to leave them having large enterprise software systems with clunky information architecture and poor user experiences. However, many small financial technical startup companies only focus on only one or two of the features that core banking softwares offer (represented around the circles). By focusing on just one type of feature, this allows these companies to create better user experiences for their users.

Survey

Whereas interviews are good at capturing qualitative data, surveys are better suited for quantitative analyses. They include simple-to-answer questions that are formed for an easy analysis after the data is collected. Due to their form, surveys are easy to send to many people at a time.

We constructed a questionnaire that asked participants about several different factors of their life, from technology usage, to banking habits, to online activities. We used Amazon’s Mechanical Turk to administer the survey to Portuguese citizens. Mechanical Turk is a service where people complete “human intelligence tasks” (HITs), small tasks like categorizing a photo or taking a survey. We had 22 valid survey responses.

However, we were unable to use the surveys to test for statistical significance. There were two factors that led us to not use the surveys. First, there was a small sample size. Second, there was a population bias: looking at the demographics, most were college students, who can speak English and also have easy access to technology. Since we were looking at the whole population of Portugal bank users, we decided to not continue the surveys (as we also ran out of time).

Photo-Diary Study

Our camera study allowed us to see into our participants lives based on given prompts. This study let us ask questions and further helped us to gain another perspective into understanding of how money is used and perceived. This research method was primarily to help us build empathy with our users. This study was particularly helpful for us as we got to see how Portuguese people lived and more about their lifestyle.

We asked each participant to take 29 photos based on given prompts (29 in total). We had 8 participants from Portugal complete the study. Some of the prompts were:

  • Take a picture of something that is a big investment for me.
  • Take a picture of something (object or place) that my friends and I do for entertainment.
  • Take a picture of something that you buy everyday (or almost everyday). 
  • Take a picture of something that you use everyday (or almost everyday).
  • Take a picture of how you see the news.
  • Take a picture of how you purchase items.
  • Take a picture of how you get to work (or school). 
     
Screen Shot 2017-10-12 at 1.21.32 PM.png
Screen Shot 2017-10-12 at 1.21.25 PM.png

Once all the pictures were collected by the participants, we gathered them to see how they differ from each other. By having these prompts, we could see more into people's lifestyles and where it involves any monetary actions. 

Interviews

Interviews were the most time-consuming yet informative method for our user research. One-on-one conversations with consumer banking users helped us understand their thoughts on banking: their desires, goals, frustrations, worries, and how they use money. These interviews gave us an in-depth look at qualitative aspects of information, in contrast to the surveys which focus on quantitative aspects.

We completed 11 interviews with participants of varying ages, backgrounds, and occupations. Asking about how their daily activities – involving money, jobs, and social interactions – we were able to extract their perceptions and patterns.

SAM_1539.jpg

We focused on five categories: life events, spendings, savings, family, friends, and investments. For each category, we had extensive questions that we asked eac

Screen Shot 2017-10-12 at 1.28.46 PM.png

After each interview, we went through an interpretation process where we wrote key insights and created models (sequence, flow, and cultural) using an industry-standard practice. 

Flow models helped show the flow of information and data between people and artifacts. 
Cultural models depicted entities that share the same values to provide a comprehensive picture of the overall environment.
Sequence model provided a chronological list of events that depicts what our stakeholders need to go through in order to achieve their goal.


We then consolidated these models by bringing the individual ones together to build one consolidated model that would help represent the whole customer population. 


We also consolidated our notes using affinity diagramming: a method by which we group interview notes based on their conceptual nearness to other notes. Grouping allowed us to see overall patterns, extracted directly from our data.

 

Walking the Wall

After categorizing all the notes on to the affinity diagram, we revisited the diagram and the consolidated models. Using sticky notes, we wrote down any questions, design ideas, and breakdown points that we identified and stuck them next to the notes. 

 

Moving On

Whew, and that's the end of our research process. As a team of 4, we were all heavily involved with all the stages of our research methods. This extensive research process was important for us to be able to shape our ideas while fueling new ideas. We were able to learn about people's behaviors towards banks and money, understand their culture, define context, and set our focus on what we wanted to achieve for our final product.