Activation Gap: Turning Organic Web Traffic into Mobile Users

The Problem

A Fortune 500 financial institution had meaningful web traffic arriving at their Digital Tools pages and learn & grow articles, but a broken handoff to mobile app downloads. Certain “Download the Mobile App” CTAs were converting while others were being ignored entirely.

The Growth Framing

The instinct was to fix the call-to-action buttons. The real problem was the content surrounding them. Users were reading the articles and visiting the pages — they just weren’t downloading the mobile app. By focusing our assessment on the moments closest to conversion, we could pinpoint exactly where the web-to-mobile handoff was breaking down.

What I Did

Conducted a mixed-method funnel audit across the web-to-download journey:

• Collaborated with my team to analyze on-site behavioral data to map drop-off rates at each funnel stage.
• Ran user testing with current customers who were unfamiliar with the mobile app — targeting the exact activation gap we were trying to close.

Key Insight

Users weren’t failing to click — they were failing to understand why the app was worth downloading over what they already used. The friction was motivational and trust-based, not visual. The copy lacked specificity, the feature value was buried, and there were no trust signals at the moment of highest commitment.

Output

Delivered 18 prioritized recommendations across four growth levers:

• CTA visibility & placement (reduce friction)
• Message-to-intent alignment (meet users where they are in the funnel)
• Trust & confidence building (security signals, social proof)
• Motivation & context (differentiate app value from existing alternatives)

The recommendations were mapped to a prioritization matrix (scored by impact, implementation effort, and A/B test readiness). This helps the product marketing team have a clear sequence to include in their quarterly roadmap.

Challenges

Working within constraints: We knew the product marketing team had limited control over what they could update on their website. If we suggested something that required a massive technical overhaul, it would never get built. We had to understand the tech limitations and work around them so our recommendations wouldn’t be ignored.

Making sense of messy data: I didn’t have access to their main analytics tool (Adobe), and the data the client provided was not easy to analyze. I partnered with the SEO team to “hack” together the numbers we needed—like page views and drop-off rates—so we could make decisions based on facts, not guesses.

What I Learned

• Because the data was so unorganized, this project took longer than expected. If I were to do it all over again, I would leverage AI to give a first pass of patterns we can extract from the data we received.
• We didn’t have the time to review the prioritization matrix to determine which recommendations the client wanted to start with because the account contract ended. This conversation could have helped us prioritize immediate optimization opportunities.
• Some of the participants we recruited for user interviews were unqualified because they were already using the mobile app. The screener survey question could have been better-worded to make it obvious we wanted to talk to users who had an account with the bank, but not the mobile banking app.
• CTA links direct users to the app store vs the mobile download landing page depending on the mobile vs desktop experience. I missed this important detail when starting the research project. Next time I’ll make sure to test both experiences on different devices.