When designing a biodiversity monitoring programme equal attention should be paid to the relevance of expected sample data for management and the choice of indicators as to questions of sampling design and analysis.

Measuring changes across different levels of a cause–effect chain, from changes in ultimate stressors (e.g. management practices), to changes in proximate stressors (e.g. forest structure and function) and changes in biodiversity, is accompanied by an increase in cost and technical difficulty as well as increased uncertainty in the interpretation of sample data.

Structural indicators support a coarse-filter approach to forest management and are relatively cheap and quick to measure but offer no guarantee that conservation targets have been achieved. There are relatively few empirical tests of the importance of specific structural indicators for biodiversity.

Species data represent the highest standard for evaluating progress towards conservation goals, yet many types of species data and biological indicators do not provide reliable sources information for management. A lack of threat information and persistent sampling difficulties mean that endangered species are rarely a viable option for monitoring. There is little empirical evidence to support the concept of a biodiversity indicator that is capable of capturing changes in other species. There is similarly little scientific support for the focal species as a practically useful concept for monitoring. Ecological disturbance indicators are useful for monitoring because their dynamics can be linked to management impacts and underlying changes in ecological condition, yet they are rarely chosen on the basis of standardised selection criteria.

Baseline measurements of indicator variability at reference sites are essential for interpreting management impacts on biodiversity but are fraught with both practical and technical difficulties.