Framework · Behavioral science

Choice architecture, defaults and sludge

How choices are ordered, defaulted and framed shapes decisions; no layout is neutral. Defaults are the strongest single lever (moderate evidence, very variable), while nudges on average are small once publication bias is removed. Use honest defaults, all-in prices and a friction audit.

StartPay

Plain label: honest defaults and friction auditAlso known as: Nudge, Thaler and Sunstein, defaults, sludge audit

Core idea. A nudge steers choices without banning options or changing incentives much. Sludge is friction that makes it harder to do what is in a person's own interest, such as a maze to cancel. F04

Evidence. A meta-analysis of 58 default studies (about 74,000 people) found sizable shifts (about d = 0.63-0.68) with very wide heterogeneity Evidence grade B. Moderate evidence: solid studies with some limits.. Across choice-architecture studies, bias-corrected averages shrink toward zero, and large, consistent effects should not be expected outside a few exceptions such as defaults Evidence grade C. Limited evidence: few or small studies..

How it plugs in. Pre-select the plan, setting or reminder time most informed users would pick, changeable in one tap. Structure complex choices with a recommended option and a stated reason. Show the full price first. Count the steps to cancel vs subscribe and to export vs import; ratios near 1:1 protect revenue quality.

Watch-outs. Expect low single-digit effects that often decay. Defaulting buyers into paid add-ons tends to come back as refunds and chargebacks.

Keep reading

Framework · Behavioral science
Hick's law, cognitive load and choice overload
fewer, clearer choices
Decision time rises with the number of equally likely options, and clutter burdens working memory. Choice overload is real only under specific conditions. One primary action per screen and 2-4 clearly different plans are good defaults. Hick A (simple tasks); load theory B; overload C.
Framework · Behavioral scienceEvidence grade B. Moderate evidence: solid studies with some limits.
COM-B and the Behaviour Change Wheel
capability, opportunity and motivation diagnosis
A behavior needs capability, opportunity and motivation, and each gap points to different fixes. Use it as a completeness check when diagnosing why a step leaks, especially to catch opportunity problems mistaken for knowledge problems. Moderate evidence as a diagnostic (B), less as an outcome predictor.
Framework · Behavioral scienceEvidence grade B. Moderate evidence: solid studies with some limits.
Endowed progress effect
credit real progress already made
People persist more when a task looks already started. In the original car-wash field study, completion was 19% for a blank 8-stamp card versus 34% for a 10-stamp card with 2 stamps given. Credit only genuine actions. Moderate evidence (B), with few direct replications.
Framework · Behavioral science
Feedback interventions and gamification meta-analyses
what feedback and game layers really do
Feedback helps performance on average (d about 0.41), but over a third of feedback interventions made performance worse, mostly when feedback shifts attention from the task to the self. Game layers in learning show moderate gains (g about 0.25-0.49). Prefer task-focused feedback and near-peer comparisons. Feedback A; gamification B.
Framework · Behavioral science
Flow and the challenge-skill balance
difficulty that matches skill
Deep absorption comes with clear goals, immediate feedback and a perceived balance between challenge and skill. Too hard breeds anxiety; too easy breeds boredom. Use it to find boredom hotspots among veterans and anxiety hotspots among novices. Antecedents B; flow causing performance C.
Framework · Behavioral scienceEvidence grade C. Limited evidence: few or small studies.
Fogg Behavior Model (B = MAP)
motivation, ability and prompt model
A behavior happens when motivation, ability and a prompt meet at the same moment; when it fails, at least one is missing. Hooked calls the prompt a trigger (B = MAT). Use it to triage a drop-off step, working ability first. Evidence for the model as a whole is limited (grade C).

Sources

When and why defaults influence decisions: a meta-analysis of default effects · Jachimowicz et al., Behavioural Public Policy, 2019
Evidence grade B. Moderate evidence: solid studies with some limits.
The effectiveness of nudging: a meta-analysis of choice architecture interventions · Mertens et al., PNAS, 2022
Evidence grade C. Limited evidence: few or small studies.
No reason to expect large and consistent effects of nudge interventions · Szaszi et al., PNAS 119(31), 2022
Evidence grade C. Limited evidence: few or small studies.

Source IDs: F04

Book content is paraphrased, never quoted at length. Last verified October 24, 2026.

Your product

Find the stage your product is stuck in

Gamify It reads your product, picks the one move that matters at its stage, and sizes the test to your traffic. Connecting is free; a Pro subscription starts it.

Add Gamify It to your assistant Take the 2-minute check-up

Related

Hick's law, cognitive load and choice overload

COM-B and the Behaviour Change Wheel

Endowed progress effect

Feedback interventions and gamification meta-analyses

Flow and the challenge-skill balance

Fogg Behavior Model (B = MAP)

Sources

Find the stage your product is stuck in