Pattern #91: Forced Action

Pattern #91  Tested 4 timesFirst tested by Marcos Ciarrocchi Recently tested by Tim Karcher on Mar 06, 2025

Based on 4 Tests, Members See How Likely Version B Wins Or Loses And By How Much

LOSSES
-5
-4
-3
-2
-1
FLAT
+1
+2
+3
+4
+5
WINS

Measured by the sum of negative and positive tests.

A B
Forced Action (Variant A) Forced Action (Variant B)

Expected Median Effects Of B

-

Progression

-

Leads

X.X%

Signups

(4 tests)

-

Engagement

-

Sales

-

Revenue

-

Retention

-

Referrals

X.X%

ANY PRIMARY

(4 tests)

All 3 tests meta-analyzed: +X.X% combined effect (p-val X.XXXX)

60.8% of 90% cumulative power target at 1% MDE from 3 tests Help replicate this with an A/B test

90%
↓
11.6% 46.9% 16.4%
Signup 3 tests: X%

Tested on

Tests

Pattern #91: Forced Action
Was Tested On Finn.com by Tim Karcher

Replaced

Isolated

Test # 580 on Finn.com by Tim KarcherTim Karcher Mar 06, 2025 Test link

Find Out How It Did With 36,898 Visitors

  • Measured by signup completed (prod added)

  • XX.X% (XXX successes out of XXX,XXX visitors)
  • XX.X% (XXX successes out of XXX,XXX visitors)
Zoom In
Forced Action - Variant A A
Forced Action - Variant B B
Signup
Statistical Power by Minimum Detectable Effect
7%0.5%
16.4%1%
50.2%2%
80%2.9%
99.8%5%
99.9%10%
99.9%15%
99.9%20%

In this experiment, the control shows an (X) close option within a signup flow, whereas the variation had it removed. Clicking the close icon would collapse the signup modal and show the product page. Users were still able use the back and forward functionality.

(The test was inversed in order to fit the pattern). Impact on signups was measured.

Get Access To See The Test Results

The Same Pattern Was Also Tested Here

Removed

Isolated

Test # 291 Elevate App by Jesse GerminarioJesse Germinario Mar 30, 2020 Test link

Find Out How It Did With 135,631 Visitors

  • Measured by successful trial account starts

  • XX.X% (XXX successes out of XXX,XXX visitors)
  • XX.X% (XXX successes out of XXX,XXX visitors)
Zoom In
Forced Action - Variant A A
Forced Action - Variant B B
Signup
Statistical Power by Minimum Detectable Effect
6%0.5%
11.6%1%
33.1%2%
80%3.7%
96.6%5%
99.9%10%
99.9%15%
99.9%20%

This experiment was ran on the initial onboarding screens of the Elevate App - right after installing and launching the app for the first time. The change was the removal of subtle "skip" links that fast tracked users to the signup/login screen (Get Started). Hence in the variation, all users had to scroll through the 4 introductory messages before being asked to create an account.

Get Access To See The Test Results

Replaced

Isolated

Test # 214 on Yummly.com by Marcos CiarrocchiMarcos Ciarrocchi Dec 07, 2018 Test link

Find Out How It Did

Zoom In
Forced Action - Variant A A
Forced Action - Variant B B
Signup

In this experiment, the highly prominent "skip" button was replaced with a less prominent text link. The copy of the skip text link also clarified the consequence of the action - losing out on personalization benefits. More so, the habitual top-right cancel icon was also removed.

Get Access To See The Test Results

Added

Isolated

Test # 193 on Yummly.com by Marcos CiarrocchiMarcos Ciarrocchi Aug 07, 2018 Test link

Find Out How It Did With 1,335,347 Visitors

  • Measured by post signups page visits

  • XX.X% (XXX successes out of XXX,XXX visitors)
  • XX.X% (XXX successes out of XXX,XXX visitors)
Zoom In
Forced Action - Variant A A
Forced Action - Variant B B
Signup
Statistical Power by Minimum Detectable Effect
16%0.5%
48.2%1%
80%1.5%
96.9%2%
99.9%5%
99.9%10%
99.9%15%
99.9%20%

In this experiment, the presence of an additional "skip all" text link was tested on a multiple step signup flow. The skip all link allowed users to bypass personalization questions and go straight to their app dashboard. The control (A) shows its presence, and in variant B we can see it was removed.

Get Access To See The Test Results

For each pattern, we measure three key data points derived from related tests:

REPEATABILITY - this is a measure of how often a given pattern has generated a positive or negative effect. The higher this number, the more likely the pattern will continue to repeat.

SHALLOW MEDIAN - this is a median effect measured with low intent actions such as initiating the first step of a lengthier process

DEEP MEDIAN - this is derived from the highest intent metrics that we have for a given test such as fully completed signups or sales.