We highlight the flexibility of the IPSO experiment-design framework by contrasting its application to CMB, weak lensing and redshift surveys. We illustrate the latter with a 10 parameter MCMC D-optimisation of a dark energy redshift survey. When averaged over a standard dark energy model space the resulting optimal survey typically has only one or two redshift bins, located at z<2. By exploiting optimisation we show how the statistical power of such surveys is significantly enhanced. Experiment design is aided by the richness of the figure of merit landscape which means one can impose secondary optimisation criteria at little cost. For example, one may choose either to maximally test a single model (such as ��CDM) or to get the most general model-independent constraints possible (e.g. on a whole space of dark energy models). Such freedom points to a future where cosmological experiments become increasingly specialised and optimisation increasingly important.

5 pages, 5 colour figures, ApJ Style