Changing Perceptions with Design

How to design products using the power of framing effects

Aaron Otani
The Startup
Published in
7 min readJun 20, 2016


There is no truth. There is only perception.
-Gustave Flaubert

In April 2007, the Washington Post conducted a social experiment. The publication asked Joshua Bell, one of the nation’s most renowned violinists, to pose as a street performer in the subway. Donning a baseball cap and jeans, Bell took out his $3 million Stradivarius and started to play. Would people stop and listen?

Instead of the massive crowds that many expected to gather, only seven people of the 1,097 passersby stopped during the entire 45 minute performance. When questioned after, most of the commuters said that they didn’t even notice the violinist, and were regretful when informed of what they had missed.

This experiment highlights a phenomenon in behavioral science known as framing effects where small changes in messaging or the context of our environment influences our perceptions in dramatic ways. Typically, Bell captivates audiences at concert halls around the world with his virtuosic playing. Yet in a crowded Metro station, commuters perceived Bell as an ordinary street performer and walked right past him.

As product designers, framing effects impact our work more than we may realize.

Instilled within our designs are messages that frame how users perceive our product. We aim to capture people’s attention and leave a positive lasting impression. Failing to do so will undoubtedly lead to the same fate as our famous violinist — with people passing over our product without a second thought.

At Opower, our design team takes this philosophy to heart. We use behavioral science and framing techniques to ensure our products are as impactful as possible. In doing so, we’ve changed people’s habits and helped them save over $1 billion on their energy bills.

Based on these experiences, I’ll describe a three-step process that all designers can use to apply framing tactics that make their own work more compelling, persuasive, and actionable.

Step 1: Learn the science

The first step to applying framing principles in design is learning the psychology and science behind these techniques. The subject of behavioral science is becoming more popular than ever. There are many books (Influence, Nudge, Thinking, Fast and Slow, etc.) and online resources (Zurb University, Product Psychology) that describe the heuristics and biases that drive the seemingly irrational decisions people make everyday.

A small sample of recommended books on behavioral science

In a previous post, I discussed how certain framing methods influence our interpretation of numerical information. These same methods can be applied to much broader design contexts. Here, I’ll highlight a few techniques that our team has found effective in our own product design process.

Social norms

Humans are highly social beings, and we take many of our behavioral cues from social normsthe rules of behavior considered “normal” within a group or society.

Inspired by research conducted by Robert Cialdini, Opower designed communications with social norm messaging, comparing a home’s energy use to similar homes in the area. Knowing that you’re using more energy than your neighbors is a powerful motivator — one that’s encouraged millions of people to change their behavior and use less energy.

In another Cialdini study, hotel rooms were equipped with signs asking guests to reuse their towels. The first sign emphasized the environmental benefits of reusing towels, while the second sign highlighted that a majority of guests participated in the program. The social comparison message increased towel reuse participation from 35% to 44%.

Loss Aversion

Humans have a tendency to strongly prefer avoiding losses over gains — a phenomenon known as loss aversion. When presented with two scenarios of either losing $100 or gaining $100, the psychological pain of losing tends to far outweigh the pleasure of winning. Therefore, loss framing can be particularly effective in motivating people to take action.

As one example, Opower conducted a test to get customers to sign up for a rewards program with their energy utility. By framing our emails in terms of loss, we observed response rates 5x higher than our control.


It’s a human condition to want what we can’t have. Scarcity increases an item’s perceived value and makes it more desirable.

Many new product launches use the scarcity principle to drive up demand through beta invites or long waitlists. Mailbox became known for its notoriously long waitlist when it first launched, which frustrated many but also generated a lot of buzz. Within a few weeks, nearly 1,000,000 people were waiting in line to try the app. Scarcity, combined with social proof, provides strong behavioral cues that a product is worth paying attention to.

People who signed up to try the Mailbox app could see in realtime their position in line.

Step 2: Choose the right message for your audience

Research shows that targeting your message to the right audience is just as important as the framing of the message itself.

A study on charitable fundraising compared how different messages influenced two different user groups — potential donors who expressed interest in a charity but had not yet donated, and regular donors of that charity. These groups received one of two letters asking for donations. One message focused on how much money had already been raised to-date (the “to-date” framing), while the other focused on how much money was left to-go (the “to-go” framing).

Messages sent to two separate donor groups saw drastically different response rates.

The study found that potential donors donated more after receiving the “to-date” message, while regular donors donated more with the “to-go” message, suggesting that different audiences are motivated by different factors. While the potential donor group was more swayed by social proof, regular donors were more persuaded by the appeal to contribute and complete a goal.

The takeaway? Know your target audience and tailor your message specifically to them. Determine which framing methods will resonate with your users for the most effective results.

Step 3: Design, Test, Iterate

Once you’ve learned the behavioral science and determined the types of users who will be most receptive to your message, it’s time to start designing.

Let’s take the example that Opower wants to help people lower their thermostat to an efficient temperature during the winter. We chose this target behavior because heating is one of the largest sources of energy used during this season. Below is an example of a brainstorming worksheet where we’ve taken a handful of behavioral framing techniques and created messages aimed at driving our target behavior.

For ideas on framing methods, check out the great resources at Zurb

Although this list is not comprehensive, we’ve already come up with several promising ideas that we can start applying to our designs. Depending on where we are in the design process, we might expand upon these initial ideas by sketching out design concepts, building out storyboards, creating wireframes, or looking for ways to incorporate new messaging into our existing products.

We also want to think about how our decisions frame the overall product experience. Should people see these communications in an email, a letter, on a specific web page, or as a text reminder? What is the right time for people to receive this message? Channel and timing are important levers that ensure we’re reaching users during the moments they’re most motivated and able to take action.

Examples of products Opower has made to encourage energy efficient thermostat temperatures

After we hone in on a few concepts and start to develop prototypes, we’ll want to test our designs. Qualitative research provides valuable insight into how people understand and interpret our products. But from experience we know that we can’t always rely on what people say they’ll do to predict their actual future behavior. Therefore, we supplement this feedback with product analytics and quantitative testing, which help us determine which messages move the needle and drive our intended outcomes. Learning from our users helps us further refine our designs and make them more effective with each new iteration.

Leveraging framing effects in your designs is just one method from a much larger design toolkit, but one that has a surprisingly strong impact when approached with thoughtfulness and care. Give these techniques a try, and let us know how it goes!



Aaron Otani
The Startup

Designer @Google. Previously @Opower, design fellow @KPCBFellows.