The purpose of the pilot was to assess the feasibility and acceptability of the intervention, study design and procedures to inform the main RCT. This document outlines lessons learned and further reflections from the pilot study on the design and delivery of the intervention and study design. These are especially valuable insights as it proved not feasible to conduct the planned RCT. However, our lessons provide an intriguing perspective into information consumption and dynamics in refugee settings. Additionally, they might apprise future similar research designs seeking to deploy IVR technology to test the effects of information bombardment, information seeding, or source labelling.

Key lessons include:

• Reaching people through IVR is challenging: 67 percent of the individuals we called picked up at least one of the two IVR calls. Only 21-25 percent of the people stayed on to hear all three messages. Problems like network, suspicion, and lack of sensitization may obstruct the reach of and engagement with digital campaigns.
• People may categorize sources of information very broadly: in this case, various humanitarian organizations were simply viewed as the category ‘humanitarian organization.’
• What people hear when they listen to a message may depend on who they are: Social and psychological mechanisms could influence how individuals interpret the source of the message. Our intervention design did not account for these heuristics that influence the role of each source on information veracity.