← Selected work · All case studies · How I work
Making an existing member benefit easier to notice
A relevant prompt helped eligible members activate an available offer. The next question was whether online activation also influenced store purchases.
My role: Led the Digital Optimization Team. The senior analyst communicated the findings; product and loyalty partners supplied the business context.
The team: A DOT experiment, with findings communicated by the senior analyst and business context from product and loyalty partners.
The test brief
The problem: Eligible, signed-in guests had bonus offers available during a member event. The team wanted to make the activation opportunity easier to notice.
The proposed solution (solution hypothesis): A site-wide prompt shown to eligible guests would make the available offers more visible and increase activation.
The primary success metric: Bonus-offer activation.
The secondary success metric: Revenue per visitor.
The full story
Start with awareness
During a member event, signed-in web guests had bonus offers they could activate. DOT tested whether a site-wide prompt could make that opportunity more visible to the people eligible to use it.
Match the measure to the question
Bonus-offer activation was the primary metric. Revenue per visitor was secondary. That distinction kept the immediate behavior—activating an offer—separate from the commercial effect the team also wanted to understand.
Read the team result in context
Amy Priola’s May 6, 2024 results announcement reported a 136.81% increase in bonus-offer activations and a 0.80% increase in revenue per unique visitor relative to control, across mobile web and desktop. After the test concluded, the winning experience was implemented through SAS Digital as part of the personalization roadmap. Amy communicated the findings and next steps; product and loyalty partners supplied the business context.
Notice the question the result could not answer
Leadership discussion raised the possibility that guests might activate an offer online and transact in a store. That was a useful follow-up question. The announcement did not establish an in-store effect, so the next analysis would need to connect behavior across channels before making that claim.