The Psychology of Numbers in Design


When it comes to numbers, we’re not as rational as we think we are.

Traditional economic theory has long assumed that humans are logical, unemotional, and make decisions that are in our own self-interest. In recent years, however, the growing field of behavioral economics has revealed that this assumption is flawed — humans are in fact complex beings who often rely on emotion and reflex to make decisions, even if those decisions sometimes defy rationality.

At Opower, our design team thinks deeply about how to combine useful and delightful user experiences with behavioral science to motivate everyone on earth to save energy. We believe that understanding the psychology and science behind how people interpret information, make decisions, and take action enables us to deliver more effective designs that help us achieve our goal.

As the first in a series of posts about how to leverage behavioral science in design, we’ll look at how numbers, seemingly objective units of information, are actually easily susceptible to subjective interpretation. Understanding number psychology is useful to designing a broad range of products —anything from eCommerce sites, to fitness tracking apps, to business analytics software— where numerical information is integral to the product experience.

Glass half full or half empty?

Consider a glass of juice, filled to the midpoint. If asked to describe the contents of the glass, you could respond in a myriad of ways. You could say that the glass is half full, half empty, contains 8 ounces, 110 calories, 20 grams of sugar, or 200% of your daily Vitamin C — these all accurately represent the contents of the glass, but our brains don’t necessarily respond to these descriptions in the same way. This phenomenon, known as framing effects, explains how the same information, presented with slight variations, can drastically impact our perceptions and influence our decisions.

It’s all relative

A 1981 study conducted by Amos Tversky and Daniel Kahneman, pioneers in behavioral economics, demonstrates how framing effects can have a psychological impact on the choices that we make. When participants of the study were asked whether they would drive twenty minutes out of their way to save $5 dollars on a $15 calculator, nearly 70 percent said yes. But when asked if they would drive twenty minutes to save $5 on a $125 jacket, only 29 percent said they would. Why? Even though the $5 savings is rationally identical between the two products, receiving a 33 percent discount feels like a more enticing deal than 4 percent—and we’re willing to work harder for it.

Another example of framing effects in action comes from Dan Ariely’s book Predictably Irrational. In the 1990s, the company Williams-Sonoma introduced a bread maker in their stores for the first time, pricing it at $275. After lackluster sales, the store worked with consultants who recommended introducing a bigger and better model priced at $429. As a result, sales took off — not of the premium model, but the original $275 model. Why? With one product choice, customers with no frame of reference had trouble deciding if the bread machine was worth the money. But when compared against a much-more expensive option, the original machine seemed cheaper and more appealing. This effect, called anchoring, is a strategy used in many retail environments.

Consider Apple’s release of the $10,000+ Apple Watch Edition. Even if the company doesn’t plan to sell millions of its Edition watches, the product’s very existence reinforces the anchoring effect, making the $349 Sport watches seem downright reasonable.

Compared to $12,000, $349 for a watch with the exact same functionality seems downright reasonable.

These techniques can also apply to situations other than pricing. At Opower, we aim to frame home energy data in the most persuasive way possible to encourage people to use less energy. Since most people don’t understand energy units like kilowatts and therms, and actual cost savings are often too small to be motivating, many of our communications feature percentage comparisons to make our messaging more comprehensible and compelling.

As another example, our team has designed experiences aimed at anchoring people toward energy-efficient thermostat temperatures during the summer and winter with energy tips, special seasonal campaigns, and a programmable thermostat app. We’ve been able to measure meaningful energy savings from these efforts, nudging people to choose more-efficient temperature settings than they otherwise would have considered.

Both low tech solutions (e.g. door hangers) and high tech solutions (e.g. thermostat apps) can effectively anchor people toward energy efficient settings.

When tiny details matter

We’re all familiar with the ploy marketers use to make prices seem smaller by reducing the cost by a few pennies (e.g. $50 vs. $49.99). This technique, known as charm pricing, is popular for a good reason — it works.

However, many brands are starting to shift away from this technique, concerned that 99-cent pricing connotes cheapness over quality. Instead, these brands employ other psychological strategies to increase the price appeal of their products and services.

Research suggests that removing decimals and commas in prices can change perceptions and make prices seem more reasonable. For example, the same product presented as $1000 is perceived to cost less than a price displayed as $1,000 or $1,000.00. Airbnb uses this pricing principle throughout their site, which presumably increases the appeal of its listings and increases bookings.

Sites such as Airbnb display prices without commas or decimals

Other studies indicate that removing the dollar symbol ($) from prices altogether reduces the emotional “pain” of paying and therefore increases our propensity to spend. This strategy is frequently used at high end restaurants and luxury retailers. Here, we see how The French Laundry’s wine list displays prices without any symbols or decimals, lessening the immediate sticker shock and visceral connection to our pocketbooks.

A picture is worth 1,000 data points

As our world surges with digital systems, sensors, and intelligent devices, a question persists — how can we harness value from the enormous amounts of data we’re amassing every moment of every day?

Spreadsheets might be efficient when compiling and calculating figures, but as designers, we understand that tables of data may not be the most effective way to tell a story or convey meaningful information. In fact, a recent Cornell study indicates that numbers, when supplemented with charts and visualizations, significantly increase the persuasiveness of that information.

Consider this example from Fitbit, comparing their online dashboard from a few years ago to its more-recent design.

Visualizations support the presentation of numerical data for several reasons. The new Fitbit dashboard enhances the display of a user’s raw activity data with engaging visuals that grab our attention and focus us on the key information being conveyed. Also, charts aid comprehension and make activity trends easier to identify. Lastly, the progress meter takes advantage of the Zeigarnik effect, where a state of incompleteness makes information more memorable and drives us to complete our goals — whether that’s setting a new training record, maintaining a better sleep schedule, or staying active throughout the day.


From framing methods to pricing details to data visualization — these examples illustrate just a few ways designers can make numerical information more meaningful, compelling, and actionable in their own products.

Have other interesting examples of psychology and numbers in design? Comment or send me a note @aaronotani — I’d love to hear!