Achieving highly accurate social media measurement without patient level integration

A pharmaceutical company needed to measure a social-media campaign's real-world impact without access to patient-level user data, since the platform prohibited sharing exposed-user data and operated as a "walled garden" reaching millions of patients broadly. Symphony Health built a test-vs-control methodology at the Designated Market Area (DMA) level, clustering DMAs by demographics, diagnosis, prescription history, payment type, and HCP interaction via machine learning into five clusters split 70% test/30% control, running over 1,000 iterations to match disease prevalence and Z-tests for significance. With more than 99% statistical confidence, the campaign showed a 7.8% total patient lift and a 6.6% increase in new-to-brand patients in the test group versus control, with the method retaining 90% of the accuracy of Symphony's standard approach while meeting stricter privacy requirements.

The customer is anonymised in Symphony Health’s own case study — not withheld by us.

Symphony Health is a commercial claims and prescriber-data provider, recently divested from ICON plc to HealthVerity, supplying open/closed claims data for pharma commercial insight.

This sits in the Data Providers stage of the AI use-case lifecycle: Named examples from the real-world-data, claims and competitive-intelligence providers pharma AI systems are trained on, benchmarked against, or run inside of.