Due to complex, competitive and crucial nature of call center jobs, organizations in services industry are spending more resources than ever on staff training and development. This is the case also in Call Center Representative Training. However, although organizations invest billions of dollars every year in training, no concrete evaluation framework exists to adequately quantify the impact of Call Center Representative (Henceforth CCR) training on actual job performance. Filling this gap, current study attempts to develop a framework to evaluate training programs in the context of call center industry using Kirkpatrick’s learning and training evaluation model. Developed framework is then implemented in actual training programs of the case company to develop insights on evaluation of training programs and their limitations. The study is based on actual data of three call centers of a leading Telecom Company in Pakistan. These call centers answer approximately 72 Million calls a year. Study analyzed data of almost 627 CCRs who were trained in 34 different training programs by 18 different certified trainers at three locations. CCR training was selected as research setting because of two reasons. Firstly, high turnover of CCR’s in call center industry necessitates frequent and extensive training which makes CCR training a big chunk of resources utilized in call center industry on training and development; secondly, standardized scrutinizing procedures followed in the call center industry for hiring CCR’s enabled and facilitated implementation of training evaluation framework which is suggested in this paper. Data was scientifically recorded for the entire year 2012 and different aspects of training were recorded to ensure that Kirkpatrick model could be applied. By successfully applying Kirkpatrick’s learning and training evaluation model, the study developed a framework to gauge effectiveness of training program in call center using Kirkpatrick model. Our investigation of training programs using the developed framework revealed that training programs get very high scores at initial level. Trainees are inclined to rate trainings as excellent at level 1 (Reaction) of Kirkpatrick model but as we go deep with levels (Learning, Behavior) of model, it was identified that effectiveness of training programs deteriorate subsequently. Decline of almost 20% was recorded between the effectiveness of training at Level 1 (Reaction) and Level 3 (Behavior). These results suggest that reaction of trainees is an inadequate measure to evaluate training programs and training programs should be evaluated at a deeper level to get a realistic picture of training effectiveness. Though scope of this study was limited to call center trainings where results at each level of Kirkpatrick model could be gathered objectively, the study opens an interesting and challenging area for management researchers about exploring and improving quality of training programs. It shows the need to study further his field by developing and implementing effective evaluation models in diverse training fields, specifically in areas such as social and leadership training.