Drive app installs from web traffic

You’re asked to increase app installs among users who land on restaurant menu pages from weblinks (not yet app users). 1) Map the funnel and define metrics: web_lander -> deep_link_prompt_view -> store_visit -> install -> first_app_open -> order_within_7d. Include guardrails: bounce_rate, add_to_cart_on_web, conversion_to_order (web+app). 2) Design an experiment comparing two variants: (A) soft interstitial with smart deferred deep link; (B) aggressive full-screen gate after add-to-cart. Specify randomization unit (session vs. user), cross-device identity pitfalls, power, holdout, and exposure caps. 3) Define success criteria and trade-off policy if installs rise but immediate orders fall on web. 4) Outline how you’d segment effect sizes by intent (e.g., cart_abandoners, repeat restaurants, delivery_distance) and avoid Simpson’s paradox. 5) If app-store policies or tracking limits block attribution, propose measurement alternatives (geo-level switchback, pre-post with CUPED, modeled install lift using synthetic controls) and how you’d validate incrementality.