The operational challenges of integrated ecosystem service (ES) appraisals are determined by study purpose, system complexity and uncertainty, decision-makers' requirements for reliability and accuracy of methods, and approaches to stakeholder-science interaction in different decision contexts. To explore these factors we defined an information gap hypothesis, based on a theory of cumulative uncertainty in ES appraisals. When decision context requirements for accuracy and reliability increase, and the expected uncertainty of the ES appraisal methods also increases, the likelihood of methods being used is expected to drop, creating a potential information gap in governance. In order to test this information gap hypothesis, we evaluate 26 case studies and 80 ecosystem services appraisals in a large integrated EU research project. We find some support for a decreasing likelihood of ES appraisal methods coinciding with increasing accuracy and reliability requirements of the decision-support context, and with increasing uncertainty. We do not find that information costs are the explanation for this information gap, but rather that the research project interacted mostly with stakeholders outside the most decision-relevant contexts. The paper discusses how alternative definitions of integrated valuation can lead to different interpretations of decision-support information, and different governance approaches to dealing with uncertainty.