esc
  • WorkflowsPages
  • Case studiesPages
  • Machine Learning FoundationsLearn
  • AI EngineeringLearn
  • AI Guardrails & Red TeamingLearn
  • UI/UXLearn
  • Design SystemsLearn
  • Production System DesignLearn
  • Book a callContact
  • Download ResumeContact
Let’s talk

I’ll build the agent for it.

Elicit Implementation Intentions: Ask Users When, Where and How

Good intentions often fade. Learn how to ask users for a simple if-then plan inside your product, so they know when and how they will act.

In short

An implementation intention is a simple if-then plan saying when, where and how someone will act, and a product can ask users for one so good intentions do not fade.

11 min readFree

What You Will Learn

Many people mean to do something and then never do it. They download a fitness app, plan to walk every morning, and forget by Tuesday. A design method can help close that gap. It is called eliciting implementation intentions. To elicit means to draw out something from a person, usually by asking a good question. In this method, the product asks the user to name the moment, the place and the first step they will take.

In this lesson you will learn why plans work, how to write the question, where to place it in a product, and how to check that it helps rather than annoys. Read the key words first. Each one will come up again.

Key words

Implementation intention
A simple plan in the form “if this happens, then I will do that.” It says when and how an action will happen.
Goal intention
A wish about an outcome, such as “I want to save more money.” It does not say when or how.
Intention-action gap
The distance between what a person plans to do and what they actually do.
Cue
A moment or sign in the world that tells a person it is time to act, such as a set time, a place or a reminder.
Onboarding
The first screens a new user sees after they sign up or install an app.
Prompt
A question, message or button that asks the user to do or decide something.

Why do good intentions fail?

Designers often say the problem is not that people lack will. The problem is that the bridge from wanting to doing is missing. A person leaves a workshop wanting to give more feedback. Then the week fills up, and the wish fades. Irrational Labs calls this the intention-action gap, the common habit of not following through on things we mean to do.

Three reasons explain most of this gap. Each one points to a design fix.

  • Goals stay vague

    “Get healthier” does not say when, where or how. Leaving those open sounds like freedom, but it hands the choice to a later moment. That later moment is often busy and tired.

  • Decisions happen under pressure

    When the moment finally comes, fatigue, distraction and other tasks compete for attention. The person has to decide again, from scratch, and often picks the easy option.

  • Old routines are strong

    Current habits already have clear cues, like coffee in the morning or the sofa after work. A new behavior has no cue of its own, so it loses the race.

Elicitation attacks the first reason most directly. It turns a vague goal into a specific plan, while the user is still inside your product.

What is the if-then plan?

The idea comes from the psychologist Peter Gollwitzer, who introduced implementation intentions in 1999. His plan has a set shape: if a certain situation comes up, then I will do a certain action. Researchers describe the link as a way to start the action when the cue appears, without a new round of thinking.

  • The if part

    This is the cue, a situation the person will really meet. It should be specific. “If it is seven in the morning on a weekday” is clearer than “when I feel like it.”

  • The then part

    This is one concrete action. It should be something the person can start now, such as “I put on my running shoes.” A big goal split into many steps is not one action.

  • The link

    The plan ties the cue to the action in advance. When the cue shows up, the action is ready to go, and the person does not have to weigh it again.

Compare two sentences. “I want to exercise more” is a goal intention. “If it is seven on a weekday, then I put on my running shoes and step outside” is an implementation intention. The second one names a time and a first step.

Tip

Ask users for a cue that already happens in their day, such as getting home, finishing breakfast or opening a work tool. A plan tied to a real moment is easier to keep than one tied to a hope.

What does the research show?

A large review by Gollwitzer and Paschal Sheeran looked at many experiments. It found that forming an implementation intention improved goal achievement. The authors reported a medium-to-large effect, with a Cohen’s d of about 0.61. Cohen’s d is a way to measure how big a difference is.

The same research is careful about limits. Strong goal intentions alone do not guarantee that people reach their goals. A plan helps, but it is one tool among several. Later reviews also show that the size of the effect changes with the goal, the person and the way the plan is written.

Two real-world examples show what the effect can look like in practice.

A flu shot example. Sue Behavioural Design reports on a field study by Katherine Milkman and colleagues. Employees all received the same information about flu shots. One group was also asked to write down the day, the time and how they would get there. Sue reports a vaccination rate of 33 percent for the control group and 72 percent for the group that wrote a plan. Check the original paper before you quote these figures in your own work.

A voting example. Irrational Labs describes an experiment by David Nickerson and Todd Rogers. One group of possible voters heard the standard election details. A second group was also asked when they would vote, where they would come from and how they would get to the polls. The second group had 4.1 percent higher turnout. The page suggests that thinking through details in advance works like a planning tool.

Tip

When you read a study for a design decision, look for the size of the change, the group tested and how the plan was written. A single small test does not prove a design will work for your users.

Where should you ask the question?

Elicitation means asking the user to write or pick a plan inside the product. The question should appear at a moment when the user is already thinking about the goal. Three common places work well: during onboarding, right after a goal is set, and close to the moment the action is due.

Each place has a different job. Onboarding sets the plan from the start. A goal screen can ask for a plan at the point of commitment. A reminder can bring the plan back when the cue arrives. Choose one place first, then add more only if the first one works.

Onboarding example: the Shapa weight loss app

Irrational Labs describes how Shapa asks users to step on a scale every day. Weighing daily is not a habit most people already have, so Shapa uses onboarding to set it up. It asks users to imagine the next day and pick the moment that suits their morning routine. Then the app sends weigh-in reminders at that time. The question turns a new behavior into a plan that fits the user’s own day.

Planning example: a sedentary behavior app

The SedVis study, a smartphone app built to reduce sitting time, asked participants to write action plans. The app design shows that the when and where parts can be guided by the app itself, while the user supplies the details.

Mental health app example

The lesson for designers is that a simple question, asked at the right time, can serve as the plan prompt.

How do you write the prompt?

A good prompt is short, specific and easy to answer. Follow these steps when you design one.

  1. 1

    Name one clear behavior

    Start with one action in plain words. Sue Behavioural Design recommends making the target behavior as concrete as possible. “Eat healthier” is vague, but “choose the salad instead of chips at lunch” is something a person can do.

  2. 2

    Ask for the when

    Ask for a day, a time or an event. For example: “When will you take your first walk this week?” Offer a date picker or a list of common times so the user does not have to type a long answer.

  3. 3

    Ask for the where

    Ask for a place or a situation. A place helps the cue feel real. “Where will you be when you start?” works well, and a short list of options can help people who are unsure.

  4. 4

    Ask for the first step

    Ask what the very first action will be. Keep it small. A first step such as “put the bag by the door” is easier to start than “go to the gym.”

  5. 5

    Show the plan back

    Repeat the finished plan in one sentence, such as “If it is 7 a.m. on a weekday, then you put on your shoes.” Let the user edit it. Use the same plan later in reminders and status messages.

Tip

Ask one thing at a time. A form with five fields feels like homework. Two or three short choices usually get more honest, useful answers.

What are good and weak prompts?

Here are three pairs. The weak prompt is easy to write but hard to act on. The strong prompt asks for the three parts of a plan.

  • Exercise

    Weak: “Do you want to exercise more?” The answer is yes or no, and no plan follows. Strong: “On which days this week will you walk, and at what time?”

  • Saving money

    Weak: “Remember to save.” This is a reminder with no moment attached. Strong: “When you get paid, which day will you move a set amount into savings?”

  • Reading lessons

    Weak: “Will you keep learning?” This asks for a feeling, not a plan. Strong: “After breakfast, which lesson will you open, and where will you study?”

The strong versions have something in common. Each one ties a behavior to a moment the user already lives through. That is what makes the plan usable later, when a reminder arrives and the user has to act.

What are the risks and limits?

Elicitation is helpful when the plan serves the user. It can go wrong in three ways.

Plans can feel like chores. Ask for a plan when it matters, not at every screen.

Plans can be used against the user. A prompt that pushes someone toward a purchase or a setting they did not choose is not helpful. It is a trick. Read the course lesson on dark patterns before you add any question that steers a decision.

Plans do not fix a weak product. If the action is hard, unclear or unpleasant, a plan will not make it easy. Use elicitation together with clear buttons, short steps and good reminders.

Tip

Let the user decide who made the plan. So offer suggested plans, but always let users change them.

Check Your Understanding

Try these two questions. Read the explanation after each answer.

Quick check

Which of these is an implementation intention?

Quick check

Why does the when and where part of a plan help?

Summary

Eliciting implementation intentions means asking users to build an if-then plan inside your product. The plan names a cue, a place and a first action. It helps because the decision is made ahead of time, not in a busy moment.

  • Ask for when, where and first step

    Use short, specific prompts. Turn vague goals into one concrete behavior and one moment.

  • Place the question at the right time

    Onboarding, goal setting and reminders are common spots. Choose one, test it, then expand.

  • Keep it light and honest

    Let users edit their plans, avoid repeated chores and never use the plan to push a choice they did not make.

Sources

How is this lesson?

Last updated on 17 November 2026

Learn with me

Every lesson is free. I add new ones as I learn. Have a question, or want help with your own work? Book a short call.