Abstract Classic rule-based systems appeared to have enormous potential for accounting and auditing problems. Unfortunately, there has been limited adoption of those systems. Why have those systems not lived up to their promise? This paper pursues one potential explanation. Typically, expert systems require that the user provides it with environmental assessments. Unfortunately, those assessments vary based on who is making the assessment. This paper finds that a group of partners and managers, when compared to a group of staff, each generated different environmental assessments to the same situation. However, in a rule-based system, different assessments can lead the user to different recommendations. As a result, different users in the same situations can generate different solutions from the same system. Accordingly, the quality of the recommendations from an expert systems are not “user neutral.”.