Those who distribute child sexual exploitation (CE) material in the public Internet potentially face greater risks of detection. While public distribution is prevalent, little is known about the structure of these websites. We investigate whether websites take steps to hide their purpose, and, if so, what steps are taken? We analyze 634 websites directly or indirectly, via hyperlinks, connected to websites hosting known CE material, and compare our findings to an automated examination of the same websites. We determine whether the initial visual representation is congruent with the underlying structure and content identified in the automated data collection. Implications for understanding cybercriminal processes are discussed.