OBJECTIVES: As part of consulting projects IMSCG performs >5000 payer interviews p.a.  Interview analysis is typically focused on insights directly related to research objectives. However, often respondents provide perceptions/ unprompted insights unrelated to research objectives, which have never been analysed before. The objective was to test whether information obtained from unprompted payer responses in interview write-ups can be identified and analysed, to develop insights of payer perceptions of pharma and their impact on payer decisions. METHODS: We selected 100 projects performed 2008-2011 including double blinded interviews with national/regional payers/advisors, recorded in English in the IMSCG project database. Interviews were screened for quotes reflecting perceptions of pharma unrelated to the research objective. Only unprompted expressions were extracted and buzz-words created that allowed compression into a single word/phrase. Buzz-words were valued and categorized into perception clusters (reputation, interaction, employees, portfolio and research) to identify the cause of perceptions. Quotes linking perceptions directly to decisions were mapped separately. Respondents’ function and geography were interlinked with the information. RESULTS: We identified 543 interviews including 900 quotes with unprompted expressions of perceptions in 1900 buzz-words. Distribution of interviews was equal between national/regional payers, but 80% of quotes came from regional payers. Top5 EU accounted for majority of quotes.  Company specific and industry perceptions were evenly split. Data allows a variety of perception analysis: national vs. regional payer, pharma vs. company, company perceptions of national vs. regional payers. Perception clusters identify the cause of perceptions. Analysis of the impact of perception on payer decisions is feasible, based on 5% of overall quotes. CONCLUSIONS: The data allows a structured analysis of unprompted expressions of payer perceptions in retrospective research. Database processing is ongoing to derive in the future statistically significant answers related to the impact of payer perceptions on decisions and what these perceptions are based upon.