Demonstrating the impact of Knowledge Transfer activities and advisory services aimed at farmers is important to justify their use in the achievement of policy goals and in the spending of public funds. Experience gained in evaluating Farming Connect in Wales, the programme delivering these services using a wide range of activities, is drawn on to illustrate issues faced in detecting economic impacts at farm and sector levels, in particular in establishing the counterfactual. The various methodological tools employed are compared, including the common but 'naïve' approach of asking farmers about impacts on their business, and the 'quasi-experimental' one of comparing samples of beneficiaries and non-beneficiaries. Rather different results are obtained (farmer responses suggesting a far greater level of impact) and reasons for these are sought. Lessons have been learned that need to be taken into account in the future design of monitoring and evaluation and for the selection of the types of activities that receive public support.