Service Level Agreements (SLA) for multi-service Information Technology (IT) outsourcing contracts contain vast amounts of textual information. The SLAs provide details about a specific service, Key Performance Indicators (KPI) to measure its performance; as well as process elements, such as activities, events, and resources that are integral in achieving performance goals. However, KPIs and the process elements may be interrelated. The knowledge of such interrelationships is often tacitly present in the SLAs. The aim of our research is to extract this hidden information from IT service contracts and analyze them to empower customers of IT services to make better performance management and incentive decisions. We apply an Ontology- Based Information Extraction (OBIE) approach in developing a prototype decision support framework, named SLA-Miner. The results, obtained from analyzing a set of Industry SLAs, demonstrate the utility of SLA-Miner in identifying KPI interrelationships, deficiencies, and impacts of various process elements on individual KPIs.