YYoung and Profiting
← All frameworks
InnovationDr. Maya Shankar

Behavioral Nudge Test

Remove hidden decision friction with one evidence-based change at a time

Difficulty
Moderate
Time to result
~weeks to results
Steps
5
Confidence
95%

A Behavioral Nudge Test redesigns the environment around a decision rather than relying on more persuasion or motivation. First define the behavior people already have reason to want, such as enrolling in a benefit. Diagnose a specific source of friction or an overlooked influence: burdensome forms, a weak default, confusing options, candidate order, wording, or messenger trust. Select one relevant behavioral mechanism and make the smallest practical change. Then compare it with the existing design through an A/B test or another credible before-and-after measure. The White House examples included automatic school-lunch enrollment and replacing “eligible” with “earned” in a veterans' email. The method works by bringing program design into alignment with how people actually decide, while measurement prevents a plausible theory from being mistaken for a real effect.

Origin

Shankar described applying decision science in the Obama White House through defaults, wording changes, randomized designs, and tests of program communications.

Core principles

  • 01Small features of choice architecture can strongly affect behavior
  • 02A good nudge supports goals people already hold
  • 03Testing isolates whether the intervention caused improvement

How to run it

  1. 1

    Define the intended behavior

    Specify what people are trying or expected to do and how success will be measured. Confirm that the action serves their legitimate interests.

    Watch out A nudge should not conceal material information or remove meaningful choice.

  2. 2

    Diagnose the choice environment

    Look for defaults, ordering, effort, wording, complexity, social context, and messenger effects. Identify the factor most likely to block or distort the decision.

    Pro tip Observe the actual decision path instead of assuming low motivation.

  3. 3

    Select one mechanism

    Match the friction to a behavioral insight such as defaults or the endowment effect. State the predicted behavioral change before testing.

    Watch out Do not stack several changes if you need to know what worked.

  4. 4

    Make the smallest change

    Alter one feature while preserving the underlying choice. Keep the intervention simple enough to deploy and reverse.

    Pro tip A single word can be a valid intervention.

  5. 5

    Test and compare

    Compare the modified version with the original using a behavioral outcome. Check for unintended effects as well as the headline metric.

    Pro tip Predefine the outcome so the result cannot be retrofitted.

    Watch out Do not infer success from positive reactions alone.

In the wild

Earned, not merely eligible

The veterans-affairs email changed one word, reminding veterans that they had earned a benefit rather than saying they were eligible. An A/B test connected the endowment effect to a measurable rise in program access.

The revised email produced a nine percent increase in access to the program.

Automatic school-lunch enrollment

The program used existing data to enroll qualifying children by default instead of requiring families to complete a burdensome application. Families retained the option to leave the program.

Millions more children ate lunch at school each day.

Common mistakes

Assuming awareness is the problem

People may already want the outcome but be blocked by effort, defaults, confusing design, or distrust.

Changing everything at once

A bundled intervention obscures which mechanism produced the result.

Is it for you?

Best for

Programs, products, and communications where a measurable behavior can be ethically tested.

Not ideal for

Attempts to manipulate people into choices that conflict with their interests or informed preferences.

From the transcript

if we can better understand human behavior we can in turn design public policies and programs in ways that reflect those surprising features

Maya Shankar · (33:00)

we set up an a b test

Maya Shankar · (35:30)

instead of telling vets that they were eligible for the program we simply reminded them that they had earned it through their years of service

Maya Shankar · (35:30)

From the episode

Dr. Maya Shankar: The Mindset Shift You Need When Life Doesn’t Go as Planned

Dr. Maya Shankar