Concept/Behavioral Economics/No. 0144

Choice Architecture

Choice architecture is the way options and the steps for choosing are arranged. In behavioral economics, Richard Thaler and Cass Sunstein popularized it in 2008. Defaults, order, wording and ease of action can shape what people choose, even when the available options stay the same.

a concept: name it

01You've seen this when…

  1. in life

    A subscription page puts the annual plan first and displays its price per month. You select it before noticing the smaller link to monthly billing.

  2. at work

    Your employer’s retirement portal opens with a contribution rate already filled in. You accept the settings and return to your inbox.

  3. out in the world

    At the library kiosk, a large button offers to print a receipt. The paperless option sits beneath it in smaller text. You tap the large button.

02The idea

Every decision comes with a setup. Someone chooses which options appear, what information accompanies them, where they sit and how many steps each one takes. That setup is choice architecture.

A restaurant menu, a pension form and a ballot all have it. Even an alphabetical list gives some options the first position. Leaving a box empty creates a starting point just as surely as checking it does. There is no arrangement that removes every influence of presentation.

Several features do much of the work:

  • Defaults give people a starting point. Keeping a preselected option saves effort and can feel like accepting a recommendation. The resulting tendency to stay with it is the default effect.
  • Visibility directs attention. Size, position and emphasis affect which information people notice. A prominent monthly price can overshadow the total commitment through salience.
  • Wording supplies an interpretation. Describing an outcome as a gain or a loss can change its appeal. This is the framing effect.
  • Steps impose a cost. A password reset, extra form or required phone call adds behavioral friction. Small obstacles can matter when people are busy or uncertain.

A nudge is a particular intervention within this broader setup: it steers choices while preserving options and avoiding large changes in financial incentives.

03Why it matters

People often make decisions while answering messages, managing children or trying to finish a task before a meeting. A form designed around unlimited attention can turn an ordinary choice into an obstacle course. Clear comparisons and sensible starting points help people carry out intentions they already have.

The same influence gives designers power. A company can make enrollment effortless and cancellation exhausting. An agency can simplify applications for eligible residents or require repeated documents that discourage them. Dark patterns use interface design to steer people toward outcomes that serve the provider at the user’s expense.

When reviewing a decision process, start with four checks:

  • Name the outcome being served. Specify whether success means enrollment, informed understanding, sustained use or something else.
  • Trace each route. Count the steps required to accept, decline, compare and reverse a choice.
  • Check who bears the effort. Include people using phones, assistive technology or a second language.
  • Test the consequences. Use an A/B test when appropriate, and measure understanding and later outcomes alongside immediate selections.

04A worked example

In a 2003 online experiment, Eric Johnson and Daniel Goldstein asked participants to imagine moving to a state and deciding whether to become organ donors. One version began with donor status as the default. Another required people to opt in. A third required an explicit choice without a preset answer.

What it looks like A simple expression of willingness to donate. Each person can choose either answer, and changing the default requires little effort.

What’s actually going on The starting arrangement changes the responses substantially. About 42% chose donation under the opt-in arrangement, compared with 82% under opt-out and 79% when asked to make an explicit choice. The available answers stayed the same. The default and the required decision step changed.

These were hypothetical choices. The experiment measured stated willingness, so the percentages cannot be read as increases in organs available for transplant. Family decisions, donor eligibility and hospital procedures also affect actual donation.

What would have helped A designer concerned with informed consent would explain both options clearly, make changing the answer simple and test whether people understand what they selected. The active-choice condition also shows why an explicit decision deserves consideration alongside competing defaults.

05Where people trip up

  • Treating acceptance as informed agreement. People may retain a default because they understand it, trust the designer, overlook it or postpone dealing with it. Those explanations have different implications for consent. Check comprehension when the stakes justify it.
  • Calling every increase a success. More sign-ups can coexist with unsuitable purchases, unwanted messages or contributions people cannot afford. Define the benefit for the person choosing before optimizing the selection rate.
  • Assuming fewer options always help. A 2010 meta-analysis found an average choice-overload effect close to zero, with substantial variation across studies. The popular paradox of choice depends on context. Clear categories and useful comparisons may help more than deleting options.
  • Adding friction to protect a preferred outcome. A confirmation step can prevent a costly mistake; repeated warnings can also wear someone down. Judge the step by the error it prevents and the burden it creates.
  • Stopping measurement at the first click. Check what happens later: whether people use the service, reverse the decision, complain or encounter costs they missed. An arrangement that raises conversion today may create problems next month.

06Where it doesn’t determine the outcome

Choice architecture influences behavior without fixing it. People with clear preferences, relevant experience and time to compare may readily override a preset option. Prices, eligibility rules and available services can dominate presentation.

Design also has limits. A simpler application cannot create appointments in a fully booked clinic. A clearer savings form cannot supply disposable income. Improving the decision process helps most when people have a workable option to choose and the arrangement is part of what keeps them from choosing it.

07Roots

Richard Thaler and Cass Sunstein opened their 2008 book Nudge with an imagined school-cafeteria director arranging food displays. Fruit could appear early in the line; dessert could sit at eye level. The director had to choose a layout, and each layout would influence what students selected. Their story made a practical problem visible: arranging choices carries responsibility even when every option remains available.

Behavioral researchers had been studying pieces of that problem for decades. Amos Tversky and Daniel Kahneman’s 1981 work examined different descriptions of the same public-health decision, including lives saved and lives lost. Research on defaults later examined what happens when one answer becomes the starting point. Thaler and Sunstein brought these strands together under a name that made the designer’s role explicit.

The term traveled into public policy, employee benefits and digital products. It gave people a way to examine the menu, form or screen itself, alongside the preferences of the person using it. The question became how to arrange a decision responsibly when presentation will influence it either way.

08How solid is this?

ContestedMixedUsefulEstablished

Experiments support several mechanisms within choice architecture, including defaults and framing. Their effects vary by setting, and some popular claims, such as reliably benefiting people by reducing the number of options, have mixed support. A change in selections alone does not establish a benefit to the chooser.

09Connections

countersincludesincludesincludesincludesincludesincludesincludesincludesChoiceArchitectureEgo DepletionBehavioralFrictionNot written yetActive ChoiceNot written yetDark PatternsNot written yetNudge ConceptDefault EffectSalience BiasFraming EffectCommitmentDeviceBoundedRationality

+ 2 more in the list

10Origin and sources

Richard Thaler and Cass Sunstein popularized the term in Nudge (2008), bringing together earlier research on framing, defaults and the design of decision environments.

  1. [1]Thaler, R. H., & Sunstein, C. R. (2008). Nudge: Improving Decisions about Health, Wealth, and Happiness. Yale University Press.
  2. [2]Johnson, E. J., & Goldstein, D. (2003). Do Defaults Save Lives? Science, 302(5649), 1338–1339.
  3. [3]Tversky, A., & Kahneman, D. (1981). The framing of decisions and the psychology of choice. Science, 211(4481), 453–458.
  4. [4]Scheibehenne, B., Greifeneder, R., & Todd, P. M. (2010). Can There Ever Be Too Many Options? A Meta-Analytic Review of Choice Overload. Journal of Consumer Research, 37(3), 409–425.

Suggest an edit· Updated 2026-10-02