The Current Employment Statistics (CES) survey is conducted monthly by the Bureau of Labor Statistics (BLS) for the purpose of estimating month-to-month changes in total payroll employment, and in payroll earnings and hours for production and nonsupervisory workers. BLS is exploring the feasibility of using an earnings concept that is more inclusive than payroll, and collecting earnings and hours data for all employees. In a series of in-depth interviews with CES respondents, we explored respondent understanding of key concepts and survey terminology. We debriefed respondents on their current reporting practices and inquired about their information systems. We found that respondents had some ambiguity in their understanding of our basic concepts. In addition, we asked respondents whether they could provide the proposed items. Although almost all respondents answered affirmatively, further inquiries suggested otherwise. By expanding "cognitive" interviews to include detailed information about data systems relative to specific concepts, we learned about data quality and potentially avoidable sources of measurement error. These findings will be used to better understand the results of a data collection pilot study.