This research examines how suspects attempt to influence interviewers during investigative interviews. Twenty-nine interview transcripts with suspects accused of controlling or coercive behavior within intimate relationships were submitted to a thematic analysis to build a taxonomy of influence behavior. The analysis classified 18 unique suspect behaviors: the most common behaviors were using logical arguments (17% of all observed behaviors), denial or denigration of the victim (12%), denial or minimization of injury (8%), complete denials (7%), and supplication (6%). Suspects’ influence behaviors were mapped along two dimensions: power, ranging from low (behaviors used to alleviate investigative pressure) to high (behaviors used to assert authority), and interpersonal alignment, ranging from instrumental (behaviors that relate directly to evidence) to relational (behaviors used to bias interviewer perceptions of people and evidence). Proximity analysis was used to examine co-occurrence of influence behaviors. This analysis highlighted combinations of influence behaviors that illustrate how different behaviors map onto different motives, for example shifting attributions from internal to external to the suspect, or to use admissions strategically alongside denials to mitigate more serious aspects of an allegation. Our findings draw together current theory to provide a framework for understanding suspect influence behaviors in interviews.