At its simplest, campaign evaluation examines the reasons behind the full or, most likely, partial achievement of objectives. Broadly speaking, evaluation findings are more likely to be defensible in cases of short-term efforts to bring about specific policy changes in processes which were of interest to only a small number of actors. Any increase in complexity – the time-frame and number of actors involved, the nature of the changes sought – weakens the confidence that can be accorded to an evaluation's version of events. Done well, evaluation places the efforts of the organisation being evaluated in the context of multiple change factors, deals in contribution not attribution and focuses on deriving lessons and on stimulating refinements to strategies more than on proving causality.
Even then, there are arguments that evaluation is an exercise in attempting to know the unknowable , which is not a reason not to evaluate, nor should it limit evaluations to small-scale, short-term, highly focused campaigns. What it should do though is to clarify understandings of what is being attempted. Evaluations of long-term efforts to bring about normative change – in the trade in arms or in access to medicines, for example  – may involve constructing timelines of the processes under review, but without claiming a definitive version of events. These are inherently speculative exercises, but should be embraced as such by both evaluators and those being evaluated.
Campaign evaluation is not that different from academic history, with a continuum running from evaluations of the adoption of treaties and policy change through the 'histories of the present' of Timothy Garton Ash – 'evaluations' of political change in Central and Eastern Europe in the 1990s  – to the putatively purely historical question of whether the Romantic or Positivist branches of Polish nationalism were more effective in overthrowing the country's 125 year occupation between 1793 and 1918. By extension, modern-day campaigners can learn much from examples seemingly distant in both time and social and political context.
 Evaluator says evaluation is a waste of time, never works again, Jim Coe, October 2012, http://www.coeandkingham.org.uk/strategy/evaluator-says-evaluation-waste-of-time-never-works-again/
 History of the Present: Essays, Sketches and Dispatches from Europe, 1990-99, Timothy Garton Ash, 1999.