Many people consider the bugs which are expected to appear in software at the change of the century to be a normal maintenance issue. In this paper, we show that there are some significant differences. Taking advantage of these differences may help to lower migration costs. We look at how organizations adapt their maintenance capabilities to the maintenance demand. Under normal conditions, this strategy helps to simultaneously cope with the occurring bug load and to keep costs down. Century-change-related bugs, however, also occur in large quantities in programs that have had very low error rates in the past, bringing organizations into a situation they are not prepared for. We deduce that more mechanical approaches may be more appropriate and more cost-effective. A list of criteria is deduced to find the best approach in a given situation.