The Blank to Count Index: A Missing Measurement Layer for the Evaluation Enterprise

This paper argues that the evaluation field, despite its rich proliferation of frameworks, suffers from a fundamental architectural gap: it lacks a measurement layer that verifies, quarterly and independently, whether resources actually reach the household. Monitoring and Evaluation (M&E), Results-Based Frameworks (RBF), Randomised Controlled Trials (RCTs), Difference-in-Differences (DiD), Cost-Effectiveness Analysis (CEA), and Theory-Based Evaluation (TBE) each answer a distinct question but none measures diagnostic sight at the household level, converts blindness into a financial quantity, or enforces the temporal frequency required for course correction. The paper uses the metaphor of a building with six floors and no basement. The ground floor is M&E, the first floor RBF, the second RCTs, the third DiD, the fourth CEA, and the fifth TBE. The building has no foundation layer that verifies whether resources traversed the causal chain at all. The state measures inputs at the ground floor and estimates impacts at the fifth floor, but it cannot verify whether the money passed through the intermediate floors to reach the household. The Blank to Count Index (BCI) is proposed as that basement as a measurement layer that sits beneath the evaluation architecture and supplies the household-level, quarterly, binary, independently verified data each framework needs but currently lacks. The paper specifies the BCI's methodology: the Master Formula, the binary epistemology (Blank = NO, Count = YES), the frequency rule (F/4), and the monetary translation into Value Secured and Value at Risk. It then demonstrates how the BCI complements each of the six frameworks: With M&E, the BCI becomes the receipt module, distinguishing distribution from receipt. With RBF, it becomes the verification node, confirming whether the causal chain was traversed. With RCTs, it becomes the delivery verification module, distinguishing delivery failure from intervention failure and enabling estimation of the local average treatment effect on the treated (LATT). With DiD, it becomes the delivery verification layer, validating the parallel trends assumption and separating treatment effects from delivery effects. With CEA, it becomes the verified unit of benefit, enabling cost per verified unit rather than cost per reported unit. With TBE, it becomes the observable node, identifying where the mechanism broke down. The paper concludes by presenting a mixed-methods evaluation architecture that integrates all seven layers with the BCI as the foundation and the six frameworks as cumulative floors, arguing that this architecture is not a luxury but a necessity for rigorous evaluation. The BCI is the diagnostic infrastructure of the evaluation enterprise: it does not replace the frameworks; it supplies the data they need to function effectively. Key contributions include: A reframing of the relationship between evaluation frameworks as complementary layers in a cumulative architecture, with the BCI as the foundational measurement layer. A testable instrument for measuring diagnostic sight at the household level, with a specified formula, binary epistemology, frequency rule, and monetary translation. A feasible, low-cost pathway for governments and independent institutions to restore the course-correction function of evaluation and secure billions in public resources currently Value at Risk.