Brazilian e-commerce app Dinda lifts direct open rate 25% and grows app revenue 60% over desktop with push notification personalization

Background

Beginning in October 2015, Dinda observed that an increasing proportion of its customer traffic was arriving via mobile devices and its app. The retailer's prior approach — sending undifferentiated push notifications to its entire user base — was not optimized for the diverse interests of its customer segments. Without personalization, notifications were generating opt-outs; without A/B testing, messaging quality was not improving systematically. Dinda needed to build a mobile-first engagement engine that could grow its active user base while reducing churn risk.

What Was Implemented

• Deployed Airship Engage for segmented push notification delivery, targeting users by product category interest
• Ran daily A/B tests on notification copy variables: title inclusion, emoji use, offer framing, and segment targeting
• Used Airship's Predictive Churn Analytics to identify users at low, medium, and high churn risk by device tag
• Built retargeting campaigns to re-engage medium- and high-risk users before they lapsed
• Monitored results via Airship Insight analytics, including Lifecycle tab for understanding notification influence on user behavior
• Implemented automated welcome series to onboard new app users and encourage push notification opt-in

Results

Dinda achieved a 25% direct open rate for push notifications (noted as above typical industry benchmarks) and generated 60% more revenue from its mobile app than from its desktop website over approximately two years following the October 2015 mobile strategy pivot. Medium- and high-risk churn users decreased as retargeting campaigns took effect. All figures are vendor-reported by Airship. An attributed quote from Rodrigo Santos, marketing coordinator at Dinda, is available in the Airship source. Note: The book describes the 25% figure as "a 25% lift in direct open rates" — implying it is an improvement over a prior rate. The Airship source presents it as the absolute open rate achieved, without disclosing a prior baseline. This is a potential mischaracterization in the book that cannot be resolved without a prior-state benchmark.

Lessons

• Category-based segmentation for push notifications ensures relevance, reducing opt-outs while improving the quality of engagement for those who receive the notification
• Daily A/B testing of copy — including emojis, titles, and offer framing — systematically improves open rates and creates an organizational learning loop
• Predictive churn identification allows proactive retargeting before users fully disengage, preserving the active user base
• Rich media and creative experimentation in push notifications can outperform plain-text alternatives; testing is the mechanism that reveals what works per audience