Replacing a keyboard can be stressful for operators who take calls all day, particularly when special functions must be sacrificed. This study applied several participatory design techniques to involve over 3,000 operators in the redistribution of function keys on their replacement keyboard. Video analysis and manual data logging methods were surprisingly accurate when measured against software logs. User acceptance was excellent. In the initial cutover program, operator performance levels returned to normal very quickly, resulting in over-confidence and unrealistic expectations of post-implementation performance for the remaining call centres because the focus was on the technology rather than on the impact of technology on business benefits. As a consequence, business plan requirements for future projects have now been changed to accommodate predictions of post-implementation performance and taking into account other changes happening at the same time.