The focus of this paper is to share thinking about meta-level evaluation of action research (AR), and to introduce indicator domains for assessing and measuring inputs, outputs and outcomes. Meta-level and multi-site evaluation has been rare in AR beyond project implementation and participant satisfaction. The paper is the first of several associated with the Evaluative Study of Action Research (ESAR) in which we wish to establish the ways that espoused intents articulated in projects are realized and why certain approaches are adopted and seen to be effective. We seek to increase understanding of outcomes and impact of AR. Description is provided of multiple issues of complexity associated with establishing evaluative criteria and indicators categorized according to inputs, process, outputs, outcomes and impact. We explore theory associated with definition and practice of evaluation prior to presentation of the indicators. We think the time is ripe for deeper examination of AR practice and welcome the associated conversation and critique of our proposed indicators.