Desktop App Mobile App Web App
Staff Growth Product Designer
Adobe

Commerce Growth Tests

Three strategic B2C & B2B experiments targeting conversion, activation, and revenue across Adobe's platforms.

Team Growth design team
Timeline 2021–2024

As Staff Growth Product Designer, I led a series of strategic experiments targeting conversion, activation, and revenue across Adobe's B2C and B2B surfaces. Three tests are detailed below.

AI & Credit Paywalls Free Users on Account Recommendations in Admin Console

Test 01

AI and Credit Paywalls

Adobe's generative AI features were seeing high engagement but low conversion at the credit-limit paywall. We hypothesized that the existing paywall experience created friction at the wrong moment — interrupting flow rather than building intent.

I designed a series of paywall variants that contextualised the upgrade proposition at the point of value: showing users exactly what they'd unlock, surfacing relevant plan comparisons, and reducing the steps to checkout. The winning variant drove a measurable lift in paid plan upgrades while maintaining positive sentiment scores from user testing.

Test 02

Free Users on Account

Enterprise accounts often had a large number of unactivated free-tier users — a missed opportunity for both product adoption and potential upsell. The challenge was to nudge account admins to bring these users into their paid plans without creating a pressure-driven experience.

I designed an in-product activation flow surfaced within the admin dashboard that framed free users as an untapped resource, made bulk activation simple, and provided clear visibility into seat usage. The test resulted in increased seat fill rates and positive feedback from IT admins in follow-up interviews.

Test 03

Recommendations in Admin Console

Admin Console users — primarily IT managers and procurement leads — had low awareness of adjacent Adobe products that could benefit their organisations. The standard cross-sell approach of banner ads had low engagement and was frequently dismissed as noise.

I designed a contextual recommendations module that surfaced product suggestions based on actual usage patterns across the account — framed as insights, not ads. By tying recommendations to concrete usage data ("Your team is using X heavily — teams like yours also use Y"), the module achieved significantly higher click-through than prior cross-sell placements.