Hyperbolic Discounting: Turning Goldfish into Squirrels

Editor's Notes

• #2: Mick: I’m Michael McKennar, Executive Manager at QSuper: the superannuation fund for Queensland Government Employees. It’s one of Australia’s largest funds with 540,000 members and $54 billion in funds under management. Ash: I’m Ash Donaldson, Principal Consultant and Partner at Tobias & Tobias Australia. We’re a HCD studio that specialises in designing things that intentionally affect behaviours. As a public sector fund, QSuper returns all their profits to their members. Accordingly, they have a simple but powerful vision “Improving retirement outcomes for members.” Of course, as designers, we love projects that genuinely help people, so we jumped at the opportunity to work with QSuper. We gathered our team, including some great contractors: Everybody’s uncle, Jim McCool; Visual thinker, Dharawan Noeller; and Matt Gould, our Kiwi interaction design gun Mick: Today we want to tell you the story of our approach to solving a ‘wicked problem’ or two.
• #3: Ash: But first, let’s talk about goldfish. They don’t really have a reputation for being smart. They basically take each day as it comes and like to live in the moment. If there’s an abundance of food, say your toddler decides that the goldfish is really hungry,
• #4: Ash: then said goldfish will eat everything and die from the toxic waste produced by its excrement fouling the water.
• #5: Ash: Squirrels, on the other hand, plan for the future. They eat insects, seeds, fruit, nuts and fungi. In the warmer months, these things are in abundance. But instead of bingeing themselves into obesity, they eat only what they need, and take the time to collect and store some for later…
• #6: Ash: …because they know, ‘Winter is coming.’
• #7: Mick: When it comes to superannuation, of course, we’d prefer people to behave like squirrels – putting enough away so that the Winter of their life isn’t so bad. It’s the rational thing to do, and so, like most financial institutions, we were built on the assumption that people would act that way. Unfortunately, most of us tend to act more like goldfish.
• #8: Mick: Last year we put out a consultant’s brief to “help us define the strategic opportunity and member experience requirements for a Wealth Aggregator, an Online Advice tool and other associated capabilities within the broader context of an advice led vision.” The project was affectionately referred to as WASA. The deliverables for this brief were: 1. A Member Experience Strategy, with recommendations on implementation an approach based on member engagement, research and outcomes sought 2. A Marketing Strategy built in collaboration with QSuper Marketing and the project team, detailing ways to communicate and deliver valuable outcomes for members. To deliver and communicate the best experience for these tools, we commissioned Tobias & Tobias to work with us to better understand the needs of our members.
• #9: Ash: When we were briefed, we were told that the purpose of the WASA project was to activate ‘disengaged members.’ A little further digging revealed that it was mostly the younger members that were disengaged when it comes to active management of their Super. So, before we responded to the brief, we went out and talked to a few people that would fit the demographic.
• #10: Video of young people answering the question: ‘What investment strategies do you have?’ Universal answer: none
• #11: Ash: These ‘disengaged youth’ weren’t even considering a few years ahead, let alone saving for retirement.
• #12: Ash: They were displaying a natural human bias in planning, known as hyperbolic discounting. Simply stated, it’s preferring smaller payoffs now, over larger payoffs in the future. If something is far enough in the future, the value approaches zero – so to these young people, superannuation was as good as worthless. When we looked at the numbers, we saw that most of the younger members were goldfish. But they suddenly transformed into squirrels in their 50s. When the first snows are starting to fall and they see their retirement looming.
• #13: Mick: Superannuation works on compounding interest. The more you save earlier, the better off you’ll be in retirement. Exponentially better off. Assuming 8.5% interest, the difference between someone adding $1000/month to their super at 20 and adding at 30 is over $1.1million by the time they hit 50
• #14: Ash: To mitigate the financial hurdles caused by simply being human was a wicked problem. We had to figure out ways in which we could either fight, leverage, or work around human nature. QSuper wanted to take the research seriously to figure this out, which was fantastic.
• #15: Ash: We were going to have to go out and do primary research to understand what members faced every day and how they got by, before we could design anything of value for them. A few of these members were also doing the right thing. We needed to know what made them different from the others.
• #16: Mick: We’d asked for a collaborative approach in the brief. Not only were we interested in the outputs, but we wanted to try and create some in-house capability in Human-Centred Design. Ash: Those of you who know us, know that this is our preferred approach. We’re not precious about IP and like to develop design partnerships instead of the typical client / consultancy relationships. Mick: The benefits of taking a design partnership approach are: 1. Gaining a shared understanding of the often complex space by being an active part of research, analysis, synthesis, design and testing. 2. Transfer skills and knowledge of basic techniques that the organisation should be using every day. 3. Getting more research and design bang for buck by distributing effort across the project team Ash: It really is a win-win all around. We get to do proper research and design and our partners reap the rewards Mick mentioned.
• #17: Ash: As we mentioned earlier, QSuper has a very simple but powerful vision: “Improving retirement outcomes for members.” To help focus the research and design, together we agreed upon an initial project mission statement – as we always do, comprised of verb, target population, outcome: Verb: Help Target population: Young members Outcome: Financially savvy
• #18: Ash: The mission was “Helping our young members become financially savvy.” It was put up in our war room, on our documents, and referred to in every meeting. Having a project mission is critical, because it keeps everyone on the same page, working toward a common, pithy, memorable goal. We could always raise the question: “How is this going to help our young members become financially savvy?” As with everything in design, the mission was subject to change based on the available evidence. And change it did.
• #19: Mick: T&T worked with our Member Research Team to define a screener and recruit appropriate members. Ash: Together we defined ‘younger members’ as those in the age range of 18-40 with a heavy skew to the 18-35 range. We wanted an even spread across two of the dimensions we found QSuper had codified: Level of engagement; and Marketing Segment. We wanted to see people across the spectrum.
• #20: Ash: We knew about hyperbolic discounting, but we had to deeply understand how members relate to their finances and what their circumstances were before we could design approaches to mitigate it’s effects. We needed to go into members’ homes and have them teach us how they manage their finances. For this, we created a semi-structured interview to open the discussion. It started broad with “Tell me about your finances” then used laddering to dig into some of the key concerns. The next phase shifted to a master / apprentice relationship to learn how members managed their finances. We wanted to be taught how they received and dealt with bills; the tools and artefacts they used; how they tracked progress with payments, loans and savings. We reviewed this with the project and research team, and once everyone was satisfied of the coverage we’d get, we wrote it up.
• #21: Mick: Before we went out, the QSuper staff needed some training in observation. Jim and Ash taught us what we should be looking for and how we should capture it. We did a couple of pilot runs in the office with other staff and refined the research protocol.
• #22: Mick: We had 3 of T&T’s researchers in the field as we went into the homes of 23 members
• #23: Ash: 10 QSuper staff – from Business Analysts to the Project Sponsor - joined us on the road to observe and document. As we were wrapping up each session, we also swapped roles, giving the QSuper staff an opportunity to follow up on observations and ask their own questions while we took notes. In the car on the way back from each session, we were also able to debrief them to help consolidate the skills they were acquiring.
• #24: Mick: It was 2 weeks and plenty of driving, covering 1400km of the Greater Brisbane Area
• #25: Ash: But we gained important insights into the lives of QSuper members: the types of hurdles they face and the elegant hacks they create to manage their finances. There really is no substitute for going into the field and observing what people do in their own context Mick: <Insert anecdote about the lady who ‘logged into QSuper’.>
• #26: Ash: Of course, when we were finished with the research, we had to comb the notes and audio for insights. Again, it was all hands on deck. After a little training and some reviews, we had QSuper and T&T staff furiously spilling ink onto those yellow squares of super sticky goodness.
• #27: Ash: We organised each member into 4 columns of Post-It notes on the walls around the WASA project team. Because the research was a distributed team effort, we put up a photo of each member to help us tell our stories to each other as we structured the data. In the first instance, we ended up with a total of 1007 observations and insights up on the walls. When doing thematic analyses, it’s critical to be able to walk the walls. Posting up the observations not only makes knowledge tangible, but it helps us locate each chunk of information in space. This is vital when you have to scan, reference and organise large sets of data.
• #28: Ash: The first pass was to surface topics and find commonalities. We saw things we expected: - Having finances spread across products and institutions - The weird relationships people have with credit - freezing credit cards. We didn’t expect: - Similar self-constraining strategies applied to retail finance or offset mortgage accounts. - Almost everyone we saw shared finances which can cause tensions: - Someone not putting in enough for a group dinner; - A flatmate not paying their part of the bills; - One partner trying to save whilst the other blows their earnings. We saw a number of interesting patterns emerge across the people we studied. But all this information is so dense and daunting to consume. After the first couple of passes, we had to take photos of the original Post-Its to save space. There was so much gold in what we’d observed that we wanted to be able to share the stories with the rest of QSuper. We also wanted to generate a buzz for a new way of approaching projects internally.
• #29: Ash: This is why we engaged Dharawan to help us visualise and in so doing, better synthesise our data. Her first task was to visually summarise each member’s story. By using metaphor, these visual summaries surfaced more common perceptions, reactions and circumstances for us – things that we might not have noticed otherwise. Don’t miss her 10 min talk tomorrow on Visual Thinking for Problem Solving. Mick: Not only did these visualisations help during synthesis, they also drew plenty of attention from other parts of the business. People stopping by for other reasons would be attracted to the drawings on the walls. They’d start walking the wall by themselves, seeing what happens with real members and the common issues they face. They’d ask us questions. Then they’d bring their colleagues down. We started to get requests for guided walks of the walls. The visualisations were making the research accessible for everyone.
• #30: Ash: During Synthesis we were able to extract the ways in which members related to their finances – their mental models for this domain. That informed our design principles – 10 simple rules we had to keep in mind when designing solutions
• #31: Ash: The synthesis also surfaced issues that nobody had addressed well: the opportunity areas. Although we found many, there were a few which were quite common. These eventually became our 10 problem statements: our burning issues, ready to be solved. We discovered that there were other hurdles that compounded the effect of hyperbolic discounting. Issues like: Members feeling that Super is out of their control – it’s a black box that they can neither access nor affect; or the Immediate concerns with finances, like tradies or shift workers unable to create a budget due to their irregular income; or as we mentioned earlier, Members dealing with overwhelming credit debt. We designed reference cards to communicate the problem statements in the inverted pyramid style. They had a concise title and supporting illustration to quickly communicate the idea. Above the illustration was a description of what the research found, and below it was the actual problem statement, made up of context, issue and question.
• #32: Mick: We also saw that it wasn’t just the young members who struggled. Some people never become squirrels at all. So we dropped the “young” from the mission statement.
• #33: Ash: Now that we knew what the problems were, we had to generate some solutions. Good design happens when you get the right people working on the right problem with the right process. The right people were those that worked in QSuper and the vendors that were delivering the tools. Together they had the domain knowledge and experience. Put a technologist, lawyer, marketer and financial specialist with a designer, and you’ve got a dream multi-disciplinary team to design solutions for the types of issues members faced. Having the right people at the table also cuts out out months of throwing ideas over the fence, only to be knocked back, or changed beyond recognition. It was time to work on designing some solutions together, so we agreed to employ a full-fat Design Studio method.
• #34: Ash: The heart of any good design approach is to diverge: create as many options as possible; then converge: apply constraints or lenses to refine the thinking. Then iterate. It can be distilled into a simple cycle: variance, selection, heredity – it’s the basis of all design. You start small and… Variance: generate a variety of different approaches to solving a problem Selection: most of these approaches will fail, but with enough variance, some will survive Heredity: take what’s survived, and build on it through iteration This is how evolution works. It’s how the scientific method works. It’s also how Design Studio works
• #35: Mick: Our design studio was run as an all day workshop with 50 people. 10 teams to work on 10 problem statements. All teams were multidisciplinary: 5 people with very different roles or specialisations from across the business, external vendors working with the project team and T&T design staff were sat together at each team table. T&T started the day by defining the problem space. They established empathy with members by exposing participants to primary qualitative research, user insights and problem statements, highlighted by photos and videos that demonstrated the issues.
• #36: Ash: Design studio is a great way of generating lots of ideas, getting team buy-in and letting people take ownership of solutions. The hierarchy is flat. The only tools anyone has is a sharpie and piece of paper. It’s fast paced, interactive, exciting and exhausting. We go through rapid, time constrained iterations, going broad to generate design ideas, then applying constraints like our design principles to converge before going broad again in a typical iterative approach to design. Multiple iterations of Sketch – Pitch – Critique are run. First as individuals pitching to your team, then as teams pitching to other teams. Every step along the way, the participants are encouraged to steal the good ideas that they’d been exposed to. This way, the best ideas start to collide and merge, bringing the room to a shared consensus. Variance, Selection, Heredity
• #37: Ash: In the second half of the day, we introduced some new behaviour design cards I’d put together as a tool to diverge again. I was proud of these cards. Each one had a title, concise description, then deeper explanation, followed by a real-world example of it in use. Each member in a team would have to approach the problem statement with a different behavioural bias to focus their design solution. That meant that 5 very different solutions would be generated in every team. At least, that was the idea. Unfortunately, it caused some confusion. Most people weren’t able to digest the bias they were assigned in a fast-paced workshop setting. Our core team were well versed in these concepts, but we forgot that the other participants weren’t. We tried. It failed. But we learnt that next time, we just need to build in extra time to introduce new or complex concepts. Of course, with some skilful facilitation by Simon, it was just a small hiccup in an overwhelmingly successful Design Studio.
• #38: Mick: Near the end of the day, teams had to role-play their solutions to the entire audience. Acting something out can highlight the absurdity of what seems on the surface to be a pretty good idea. Our bodystorming was able to generate unexpected ideas and insights that couldn’t be realised through discussion or sketching alone.
• #39: Ash: The Design Studio resulted in 6 solid concepts that we captured and went back to our war room to refine. It was at this point that some new models that would form part of the experience framework started to really take shape. With the help of some more senior QSuper stakeholders, we picked the eyes out of everything and tightened up the ideas. We then went straight into testing with members.
• #40: Ash: The first round of member testing was just a set of cleaned up visualisations Dharawan did of the concepts and models in poster form. The protocol called for the members to explain what they thought the concepts meant and how they might be relevant.
• #41: Ash: In the next round, we got more detailed, putting together interface and service concepts. In each round, we made sure to test a number of ways to visualise every idea This approach killed off a bunch of ideas and resulted in a few, more refined ideas arising from the ashes Variance, Selection, Heredity. After 3 rounds of rapid testing and iteration, the final output was 10 solutions with interface designs, service blueprints and a refinement of the models that were to make up the Member Experience Framework
• #42: Mick: The experience framework we finally created was made up of the Mission, the 10 experience principles, a financial interaction frequency model that showed how often someone should be checking in on different types of finances, and how easy it should be to access, a behaviour change model that demonstrated the shortfalls of personal financial managers and what would be required to sustain engagement and affect behaviour change, and finally, a financial maturity model that tied many of the outputs together.
• #43: Ash: The financial maturity model mapped not only the spectrum of different stages we observed people at, but the progression of those stages. One interesting stage was that of creating an emergency buffer. No matter how much income someone was receiving, if they didn’t have the emergency buffer, there was a constant anxiety about the ‘what if’ scenarios. “What if the fridge dies? What if my husband breaks his leg at footy and can’t get out on site? What if I don’t pick up the next contract?” Once people had that buffer set aside – in a place that was hard to access and ‘out of sight, out of mind’ – they were more comfortable with money. They’d be more likely to save, invest and even plan for retirement. It was an anxiety-relieving hack. A key trick that the financially savvy had up their sleeve.
• #44: Ash: The vendors’ tools for WASA had a laundry list of modules, features and functionality that could be turned on. Of course, turning them all on at once may seem to give the best bang for buck when you’re paying licensing fees, but it would just overwhelm those who needed these tools the most. Applying our experience framework, we were able to identify which features should be shown when, and in what order of priority. Alongside the vendor capabilities. we listed features that weren’t yet available in their software, and the supporting content that should be created and deployed. This was all mapped against each stage of the maturity model.
• #45: Mick: We also wanted to communicate what the WASA tools could be. The team had fleshed out and developed a number of interfaces and tools that the vendors didn’t yet support. They refined the interface concepts and applied visual styling that matched our future branding. Ash: We then created a storyboard and script, demonstrating how each of these concepts addresses the typical issues members face. To give context to the interface designs, we went out and took some photos, got a voice over done, and put it all together in a video illustrating a vision of what these tools could evolve to do. The purpose of this was to bring the ideas to life. To demonstrate how our solutions would overcome the very real issues QSuper members faced, and be able to communicate it in a way that would make sense to everyone - from the Executive Sponsors, to Vendors, to the average woman or man on the street.
• #46: Ash: Finally, working with QSuper Marketing, we were able to use the experience framework and primary research to shape the marketing strategy and brief the advertising agency by again walking the walls and showing them the concepts. Together we were able to identify appropriate channels, and create key messages and calls to action. Something we discovered in our concept testing, however, was that although QSuper is a highly trusted brand, providing free tools repeatedly raised the question “What’s in it for QSuper?” For this to have any uptake, we had to explain the reason we want them to aggregate all their financial details in the tools we were providing. So we drew upon the company vision and project mission and reframed the proposition to members as “Helping you with your finances today, so you can invest in tomorrow.” In the next round of testing, using that framing, the penny dropped. Member suspicions disappeared because their question was addressed before they could raise it. “Helping you with your finances today, so you can invest in tomorrow.” What’s in it for QSuper by providing tools and advice? More investment from members. It’s a win-win.
• #47: Mick: So, the project mission and marketing value proposition finally settled on: “Helping members manage their finances today, so they can invest in tomorrow.”
• #48: Ash: After all was said and done in this project, we had involved close to 40 members and more than 50 QSuper staff in a human-centered design process. We generated plenty of insights and concepts which were refined and validated, and have been able to communicate not only what the current tools should be, but what they should aspire to be in the future. We’d also identified some wicked problems that nobody had solved and designed pragmatic solutions to them. Mick: Throughout the process we got to learn firsthand the value of deeply understanding our members’ contexts and behaviours. We have a greater appreciation and awareness of the members and the real hurdles they have to overcome each day. We saw how their biases lead them into traps and so understand what has to be addressed. We have a vision for evolving the WASA tools into the future. We came up with some innovative concepts – things that nobody in the market has created, that can either be designed in house, or can inform our vendor selection into the future.
• #49: Ash: A few months after we completed this project, we received some unsolicited feedback via our website. It’s a nice reminder that frameworks and materials should be able to stand on their own when we’re not there.
• #50: Mick: The Wealth Aggregator, called Money Map, is now live. Online advice has also launched. The solutions we have built include features (such as the ones listed) that will help, guide and advise our members. They won’t completely solve the problem of engagement, but they’ll provide an essential basis for helping our members better manage their finances.
• #51: Mick: This is all linked back to the financial maturity model which, during the course of the project and beyond, has evolved from a linear progression, through a snakes and ladders model, to a circular model – acknowledging that people jump between stages, depending on life events like marriage, children, job changes, sickness or death. The good news is: we now understand the real hurdles our members face. We can address them through content, tools and services we provide them with. We’ll never be able to make everyone a squirrel, but with our new understanding of the issues people face along with a number of pragmatic ways to address them, we should be able to help more of our members improve their retirement outcomes.