Experiments

REMOVE BREADCRUMBS

The Experiment in Action

Hypothesis

If a merchant’s navigation is effective, breadcrumbs become redundant. Removing the breadcrumb trail will reduce visual distractions during checkout and result in higher RPV.

This is one of three breadcrumbs hypotheses being tested. Review early results from the SIMPLIFY BREADCRUMBS and SUPPRESS PRODUCT-LEVEL BREADCRUMBS experiments.

CONTROL – MOBILE

TREATMENT – MOBILE

Remove Breadcrumbs Experiment

How It Works

We are removing the breadcrumb trail on the treatment group and monitoring results.

Why We Are Testing It

Most merchants implement breadcrumb trails on their websites. They are a standard convention to help shoppers sort product views by category, determine which category the product is from, and find their way back if they make a wrong turn in their search. But even trusted conventions should be tested and optimized. We wanted to test the effects of removing the breadcrumb trail to see if it reduced distractions for the buyer during the checkout process.

Results: Remove Breadcrumbs Experiment

Average performance numbers are based on results from March 14, 2019 to September 15, 2019 from 6 merchant sites.

Average Experiment Results So Far*

Device Type

Average Lift

Total Visitors

Revenue per Visitor
(RPV)

Conversion Rate
(CR)

Average Order Value
(AOV)

Treatment

Control

Mobile

-9.57%

-13.90%

+5.22%

101,470

175,072

Desktop

-1.23%

-6.32%

+8.54%

121,288

190,292

Tablet

-8.35%

+5.28%

-8.09%

7,997

27,317

Results: Remove Breadcrumbs Experiment

Average performance numbers are based on results from March 14, 2019 to September 15, 2019 from 6 merchant sites.

Average Experiment Results So Far*

Mobile

Average Lift
Revenue per Visitor (RPV)	Conversion Rate (CR)	Average Order Value (AOV)
-9.57%	-13.90%	+5.22%

Total Visitors: Treatment: 101,470 | Control: 175,072

Desktop

Average Lift
Revenue per Visitor (RPV)	Conversion Rate (CR)	Average Order Value (AOV)
-1.23%	-6.23%	+8.54%

Total Visitors: Treatment: 121,288 | Control: 190,292

Tablet

Average Lift
Revenue per Visitor (RPV)	Conversion Rate (CR)	Average Order Value (AOV)
-8.35%	+5.28%	-8.09%

Total Visitors: Treatment: 7,997 | Control: 27,317

Distribution of Average RPV Lift for All Traffic

*These results reflect the average findings for all participating merchants and are not guaranteed for all users.

Data Analysis: Remove Breadcrumbs Experiment

What It Means

We are seeing a significant negative impact on revenue when breadcrumbs are removed entirely from merchant sites. Few results are indicating that this treatment is providing a positive user experience.

Things to Consider

Shoppers love breadcrumbs! If a merchant currently does not use breadcrumbs on their site, they should consider implementing the feature to see if there is a positive impact on revenue.

Who Should Try It

If a merchant has a flat or compact product catalog, it could be worth testing the removal of breadcrumbs — particularly if users are less inclined to browse for other products.

Who Should Try It

If a merchant has a flat or compact product catalog, it could be worth testing the removal of breadcrumbs — particularly if users are less inclined to browse for other products.

Related Experiments

SUPPRESS PRODUCT-LEVEL BREADCRUMBS

SIMPLIFY BREADCRUMBS

Interested in Learning More?

Get deeper insights into over 130 experiments focused on increasing mobile eCommerce growth available through the Initiative.

JOIN THE INITIATIVE