CASE STUDY
Increased E-Commerce Conversion Rates through Iterative A/B Testing
Efficiency
Conversion Rate
Challenge
Proctor & Gamble asked TargetClose to help increase e-commerce conversion rates for their oral care brands, OralB and Crest. The key requirement was to quickly strategize, test, and report on new UX/UI variations for both brand websites.
Challenge
Proctor & Gamble asked TargetClose to help increase e-commerce conversion rates for their oral care brands, OralB and Crest. The key requirement was to quickly strategize, test, and report on new UX/UI variations for both brand websites.
Solution
Solution
TESTING STRATEGY
TargetClose created over 40 unique tests on the homepage, product pages, checkout, and the navigation bar to optimize conversion rates. Many of the tests were iterative - building upon the success of previous tests. The tests ranged from changing the button colors, to adding quick checkout options, to revamping the entire cart page.
CONTROL
TEST
PAYPAL FIRST
CONTROL
TEST PAYPAL FIRST
Development
We built page variations to test as 1) separate pages on the brand sites and 2) via custom javascript and CSS in Google Optimize. The choice depended on the completixy of the development. Google Optimize was favored for simple tests, such as button color changes or hiding and adding elements on a page. Turnaround time to start these tests was 1-3 days.
Payment Options
Cart & Checkout
Social Proof
User Journey
Cross Selling
Product Order
Analytics & Tracking
Analytics & Tracking
TargetClose created weekly reports for P&G to have constant, easy access to the test’s performance. At the conclusion of each test, TargetClose provided a recommendation and/or further optimizations based on the data findings.
Outcome
A/B tests with statistical significant winners were rolled out on the Crest and OralB websites and became part of the default user experience. Assuming all tests rolled out and had a cumulative effect, the total lift on mobile would be 9.6% and the total lift on desktop would be 11.7%.
*Assumes all tests rolled out and had a cumulative effect
Crest Highlights
Numbers Sold Test
Observation: Product pages lack a social value proposition.
Outcome: Adding social proof text “353 sold in the last hour” led to a 25% increase in CVR and a 31% lift in revenue.
Additional Tests
Observation: 95% of site visitors check out with credit card, while only 5% check out with PayPal.
Outcome: Repositioning “Check out” before PayPal in the mobile checkout flow led to a 6.5% lift in CVR.
Observation: Shipping times are long and disclosed to users.
Outcome: Hiding “5-8 days” under shipping method led to a 13% lift in CVR on desktop.
Oral B Highlights
Brush Head Navbar Test
Observation: The Brush Head tab in the navbar does not lead to a shopable product page.
Outcome: Redirecting the Replacement Brush Head tab to a product page with best sellers at the top led to a 35% lift in CVR on Mobile.
Additional Tests
Observation: The value of the deal customers receive on products is unclear.
Outcome: Red strikethrough on original price on deals page led to a 31% lift in AOV and a 18% lift in revenue on mobile.
Observation: The X-out on the promotional banner interrupts the customer journey.
Outcome: Changing the UI led to 6.8% lift in revenue (monthly impact: $20k) in mobile.
