This paper analyzes the effectiveness of Indonesia’s conditional cash transfer flagship programme, the Family Hope Program (PKH), in local level implementation. By examining the Bantaran District in Probolinggo Regency, it examines how communication, resources, implementer attitudes, and bureaucratic formalism affect programme performance. A qualitative case study: in‐depth interviews with 17 purposively sampled informants, programme facilitators, community leaders, and beneficiary families. In addition, field observations were made and programme reports, as well as local rules, were analysed. The results also show a strong implementation gap: while the PKH is institutionally embedded, there are clear constraints to its transformative potential. Crucial bottlenecks include a lack of facilitator capacity, problems with the communication process that cause beneficiaries to misunderstand conditionality, and very instrumental (rather than substantive) beneficiary compliance. A highly centralized bureaucracy and ongoing targeting inaccuracies also undermine the programme's credibility. An innovation of the study is to combine several strands of evidence to examine CCT implementation at the sub-district level, suggesting that, as Edwards (1998) observed, his variables are interactive and ecosystemic in nature. It points to an “implementation paradox” in which operational viability reinforces marginal developmental effects. The study provides evidence-based recommendations for re-jigging implementation systems and an analytical lens suitable for other social protection programmes in developing countries.