Behavioral Science for Digital Health and Product Teams
The Problem You Are Facing
Users complete onboarding and then drop off. Feature adoption is low despite positive user research feedback. Engagement declines after the initial period. The product works technically, but people are not using it the way the design intended. You have tried gamification, notifications, and UX improvements, but the behavior data has not shifted meaningfully.
These are behavioral problems, not product problems. The product works. The question is why people are not performing the target behaviors (daily usage, feature engagement, sustained engagement over time), and that requires behavioral diagnosis before design intervention.
Why Product-Led Approaches Miss Behavioral Barriers
Product teams typically approach engagement through the UX lens: make the experience intuitive, reduce friction, add delight. This addresses one dimension of the COM-B model (physical Opportunity: making the product easy to use). But if the barriers are in other dimensions, UX optimization alone will not move the needle.
Post-onboarding dropout often reflects a motivation barrier. During onboarding, the product has high salience: the user is focused, guided, experiencing novelty. After onboarding, the product competes with everything else in the user's life. The habit of not using the product is stronger than the intention to use it. This is the Plan vs. Impulse dynamic applied to product design.
Low feature adoption often reflects a capability barrier combined with a motivation barrier. Users cannot hold all features in working memory (cognitive load), and the effort of learning a new feature rarely feels worth it at the moment the option arises (effort-reward calculation). Tutorials and tooltips address capability. They do not address the motivation calculation.
What a Behavioral Design Approach Adds
The SHIFT framework applies the same diagnostic precision to product behavior that it applies to organizational behavior. Specify the target behaviors (not 'increase engagement' but 'user completes one core task daily within the first 14 days'). Diagnose the barriers using COM-B. Design interventions matched to the barriers.
Practical applications: the 'Slowly Increase Difficulty' strategy translates directly to progressive engagement design. Week one: one core task. Week two: add a second. Week three: connect to existing workflow. Each stage builds competence and confidence before adding complexity. Implementation intentions can be embedded in the product ('When I sit down at my desk Monday morning, I will open [product] and check my dashboard'). Social proof features (BCT 6.2, Social comparison) show users how peers with similar profiles are using the product.
For digital health specifically, the behavioral barriers are often compound: low health literacy (capability), environments that do not support the health behavior (opportunity), and competing short-term motivations that override long-term health goals (automatic motivation). Effective digital health interventions layer techniques across all three COM-B domains rather than relying on any single approach.
Product engagement is a behavioral design problem. If you have optimized UX and engagement still lags, the barrier is likely not physical opportunity (product friction). Diagnose with COM-B before adding more features or more notifications.
