Skip to content

Technique · Hooked

Remove what the loyal segment doesn't use

Hide or remove features that pull new users away from the path your most habitual users took, and give the feature that path depends on the most prominent place.

Evidence grade C. Limited evidence: few or small studies.ValueStart

Plain label: subtractive changeAlso known as: subtractive modification, feature removal, simplify, declutter navigation

How it works

Every extra option competes for attention and adds a decision. The Modify step of Habit Testing includes taking features away, not only adding them: when the habitual segment ignores a feature, removing it shortens the route to value for everyone else.

What the research says

  • Too many options hurt choice mainly when options are complex, preferences unclear or the task hard; elsewhere the average effect is near zero.

    Chernev, Bockenholt & Goodman, Journal of Consumer Psychology, 2015

Before you start

  • You can ship changes behind a feature flag.

How to build it

Web
Move menu items the loyal segment rarely opens behind a single 'More' entry for new accounts, and lead the navigation with the path-critical feature.
B2B
Hide admin and advanced settings from end users until they need them; keep them one search away.
Mobile
Retire a tab the loyal segment ignores and promote the action that predicts retention to the main bar.

What can go wrong

  • Removing a feature a small but valuable segment depends on.
  • Cutting by overall usage instead of usage among the loyal segment.

Pairs well with

  • Mine the Habit Path first so you know what to protect.
  • Roll out behind a feature flag with a holdback.

In practice

Where it shows up

  • Superhuman · 2018

    Round

    Superhuman: 22% to 33% without shipping a feature

    “Very disappointed” went from 22% (summer 2017) to 33% by segmenting to the right personas, with no product change, and to 58% within about three quarters, per CEO Rahul Vohra (First Round Review, Nov 13, 2018). Company-described method.

    M1 ×Value
    Case grade for the number: B. Evidence grade for the mechanism link: B.Company-reported

Limits

What we don’t claim

Sources

  • Hooked · Nir Eyal

    Ch.8

Source IDs: HK-ACT43

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.