Environmental monitoring programmes are often confined by limited funds. By use of Monte Carlo simulations and GLM procedures, it is examined to what extent trends documented for some Norwegian seabird populations could have been obtained with less field effort. Changes in both breeding and wintering numbers are analysed. To minimise bias from merging different populations, only regional trends are considered. The results strongly suggest that some populations can be monitored adequately by less effort than that applied, i.e. still ensuring that the counts will reveal the various sources of variation in reliable proportions to the total variance in bird numbers. This may be achieved either by reducing the number of plots or by counting them less frequently, for example in alternate years only. General advise with respect to what changes deserve special attention from conservation authorities, and a discussion of the principles and considerations which should be taken into account when designing monitoring programmes for seabirds are presented.