Understanding and reliably predicting the ionisation of the winds of early-type stars is in several respects of key importance:

1. Unsaturated UV resonance lines, such as the trace ions N V lambda 1240, C IV lambda 1550 and Si IV lambda 1400, are very sensitive mass-loss diagnostics. Mass-loss rates as low as 10^{-9} M_{odot}/yr can be measured. However, current uncertainties in the theoretically derived ionisations of these species may lead to uncertainties in the mass-loss determination of up to an order of magnitude.
2. Theoretical predictions of mass-loss rates, based on the theory of radiation driven winds, critically depend on the run of ionisation of metal ions. Here the iron ionisation is most important as this is the dominant driver. Consequently, changes in the ionisation of Fe may lead to significant changes in mass loss.
3. Ionising fluxes of early-type stars are important in the study of emission nebula around individual stars as well as of giant H II regions irradiated by young starburst regions and galaxies. Subtle effects of line-blanketing may strongly influence predicted ionising fluxes in e.g. the He I continuum, while shocks may dominate in determining the number of He II ionizing photons. Both line-blanketing and shocks make their presence felt by modifying the ionisaton. In this review I will address the predictive power of current models with respect to ionisation, by comparing theoretical results with observed ionisation properties.