An increasing number of U.S. law enforcement agencies have disclosed that they have large numbers of untested sexual assault kits (SAKs; also called "rape kits") in police property storage. Whether previously untested SAKs should be tested for DNA evidence has been the subject of considerable public debate. To inform policy and practice regarding rape kit testing, the current study tested a sample of 900 previously unsubmitted SAKs from Detroit, Michigan, and documented the DNA forensic testing outcomes associated with those kits. We assessed how many SAKs yielded DNA profiles eligible for upload into CODIS (Combined DNA Index System), the federal DNA criminal database; how many resulted in a DNA match (termed a "CODIS hit"); and how many of those hits were associated to other sexual assault crimes (i.e., serial sexual assault hits). Overall, there were 259 CODIS hits, 69 of which had DNA matches to another sexual assault case. The potential utility of a DNA profile and CODIS hit may vary depending on whether offender was known or unknown to the victim, so we examined these outcomes separately for SAKs associated with stranger- and non-stranger-perpetrated sexual assaults. We also present six case study examples of how DNA testing and CODIS hits helped identify serial sexual assaults in both stranger and non-stranger sexual assault cases. Implications for rape kit testing policies are discussed.