Small Change: Why Behavioral Science Needs to Go Beyond ‘Nudges’ to Make a Real Impact
Behavioral science has become a popular approach across multiple industries. The practice of using human tendencies to produce desired outcomes is often cited as the tonic for practically everything that ails us. From weight loss to work performance to financial health, the easy-to-swallow message that tiny tweaks will lead to impressive payoffs has found a receptive audience.
As behavioral scientists we should be enjoying this ride, helping identify small lifestyle and product adjustments that will lead to life-changing results. But something has happened on the way to the promised land. Could the critics of behavioral science have a point? Are we creating unrealistic expectations about the effect of a “nudge”?
Common Cents Lab is a behavioral science group helping financial service providers design products that improve financial decision-making for low-to-middle income consumers. In stepping back to assess our work over the last three years, we found something interesting.
We run various projects with partners, involving randomized controlled trials of changes ranging from big product modifications to small language tweaks. About half of these projects have produced meaningful positive change or impact. While this is better than the 5% success rate of corporate innovations, we wanted to figure out how to increase our likelihood of success.
Accounting for product type, experiment methodology and other variables across all 75 of these experiments, it became apparent that the reason for the clear difference in success was directly tied to a project’s scope. Our largest gains have been made through transformative re-visioning of products or services, not by small nudges or tweaks.
Small Nudges = Small Change
Take for example our collaboration with Self-Help Credit Union. To help more new members commit to saving for retirement, we created an entirely new product that incentivized and automated regular savings deposits. This free, introductory retirement savings account was bundled with a checking account and pre-seeded with a $100 deposit from Self-Help, which customers would lose if they closed the account or made a withdrawal in the first year. This created a sense of loss aversion. We then reduced the pain of saving into the account by automating a transfer of a fixed percent of each checking account deposit.
Over the last year, more than 20% of the members in the pilot have kept their account open and continued automatically saving. In an industry accustomed to much more incremental improvement, this is a significant win. Extrapolating wider, if financial service providers could convert 20% of the estimated 18 million working households in the U.S. who are not currently putting away anything for retirement into savers – even if they’re only saving small dollar amounts – it would make an important dent in America’s growing retirement crisis.
This sizable impact stands in stark contrast to projects like an experiment we ran with Lake Trust Credit Union. We targeted members who had previously opened an Aspire Certificate of Deposit (CD), a goal-based product that earns 1% interest with a $50 minimum deposit. In the experiment, we emailed members with different messages encouraging them to save. While we found that, across the board, 42% of people who opened the email made a small, one-time deposit, the different messages inside the email made no sizable difference. So while it’s important to expand the number of people saving more than planned, a small reminder like a nudge leads only to small increases in savings. And subtle differences in language made no meaningful improvement at all.
Moving Toward Bolder Behavioral Science
Ultimately, though they’re cheap and relatively easy, small tweaks tend to have small, marginal changes in behavior that are likely not enough to reverse America’s financial health problems. And as the popular saying goes, you can’t put lipstick on a pig: A clunky product that requires considerable effort to use cannot be saved with a tweak. What’s more, besides being largely ineffective, these half-measures are leaving most of the potential of behavioral science untapped.
As behavioral scientists, if we want big effects then we need to make bolder changes. Incremental nudges and improvements are fine, but they should point to large underlying design choices informed by behavioral insights that can lead to significant gains.
Behavioral science must become a discipline core to the product design process, from writing terms and conditions to establishing delivery channels to designing user experience – and even to gathering the data we use to measure the success of these efforts. In this way, it can become a key consideration of marketing, accounting, technology and more.
To make this scenario a reality, we challenge everyone – including ourselves – to move beyond our focus on tweaks and simple email nudges on their own. Instead, be bolder in your undertaking. Let behavioral science in, seek to disrupt products in fundamental ways, and then test and measure the impact of those substantive changes.
This year, we’re putting this goal into practice by focusing on two key areas.
We are completely re-imagining how to design automatic savings products in a way that encourages greater uptake. Rather than saving a static amount on a specific date, we plan to add flexibility, control and safe-guards – while still capitalizing on the frictionless process. This includes smarter savings rules and timely prompts to allow a saver to easily cancel or change an impending transfer.
Rather than nudge people to look at a personal finance manager, which categorizes their spending over the last month, to see where their money is going, we are exploring ways to move accounts from multi-line spreadsheets and pie charts into various physical account structures, such as separate checking accounts for fixed versus variable expenses. We are testing whether providing timely rules of thumb for spending – like texting someone on Friday to only eat out for one meal that weekend – can decrease spending in a more meaningful way than complex monthly budgets.
So join us in thinking big. We must move behavioral science to center stage and boldly rethink products for financial health. Low-cost experiments and small tweaks can hold more meaning when they drive people to something larger. The true power and potential of behavioral science is not in the nudge, but rather in the design.
Mariel Beasley is a co-founder of Duke University’s Common Cents Lab, which is supported by MetLife Foundation and the Blackrock Emergency Savings Initiative, and she is a principal at the Center for Advanced Hindsight.
Photo courtesy of Vlad Chețan.