Pattern #78: Structured Information

Pattern Author: Kimberly Cheung - Product Manager at Yummly

Based on 2 Tests, Members See How Likely This Pattern Will Win Or Lose And Its (?) Median Effect

Almost Certain Loser
Almost Certain Winner
Structured Information
  1. Visible Tags Explaining

    This pattern asks the questions whether to show or hide metadata, tags or descriptors for each listing or tile card. 

Median Effects



Ex: First Action

(2 tests)



Ex: Leads, Quotes

(1 tests)



Ex: Trials

(1 tests)



Ex: Future Action



Ex: Transactions






Ex: Return Visits



Ex: Social Shares


Test #284 Tested on by Julian Gaviria Julian Feb 19, 2020

Find Out How It Performed With 241,428 Visitors

Listing Desktop, Mobile
  • Measured by company listing views   |   p-val (?)

  • Measured by company contact (RFI)   |   p-val (?)

In this experiment, structured data tags were displayed on a listing page to help potential buyers make better decisions. The additional information about the listed companies included: annual revenue, employee count, and year of establishment.

Get Access To See The Test Results

Test #195 Tested on by Kimberly Cheung Kimberly Aug 13, 2018

Find Out How It Performed With 371,274 Visitors

Listing Desktop, Mobile
  • Measured by clicks on a card and visits recipe   |   p-val (?)

  • Measured by full registrations   |   p-val (?)

Get Access To See The Test Results


Leak #59 from   |   Aug 19, 2020 Product

Amazon Discovers That Showing Structured Product Details Is Better

Amazon has been seen a/b testing the addition of product details or technical specifications, visibly above the fold. In this captured screenshot for example, we see such technical information as storage size, model numbers and types of interfaces. Interestingly, this change was rolled out when the experiment finished - hinting at a positive probability. View Leak

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.