Chapter 15: Keeping the Metrics Usable: Narrowness and Timing

Accountability metric bodies lose leverage the moment they cease to function as shared infrastructure and become identified with one side. Narrowness preserves multi-sided usability. Timing determines whether that usability coincides with moments when elite calculations are already fluid. Both disciplines decide how far organised information actually travels into decisions.

Narrowness is the first requirement. A body that publishes a small set of recurring, comparable numbers—uptime by feeder or district, early-grade learning scores by cluster, process days and side-payment frequency, transfer arrival or leakage rates—can be cited by rival factions, consulted by external financiers, checked by intermediate brokers, and used by neighbouring officials. The same body, once it adds general political commentary, broad anti-corruption rhetoric, or open alignment with a particular faction or opposition figure, becomes easier to dismiss as partisan. Competing elites no longer treat the figures as neutral ammunition; external actors discount them; brokers whose positions depend on the incumbent distance themselves. Methods that were transparent enough to deter manipulation are now read as selective. Multi-sided usability collapses, and with it the capacity of the data to shift cost-benefit calculations. The versions that have remained useful treat the numbers as infrastructure: fixed indicators, fixed schedule, methods open to inspection, and no accompanying campaign. That restraint is what allows the same series to serve rivalry, precedent, and benchmark verification simultaneously.

Timing supplies the second discipline. Demand for credible comparative numbers rises sharply during succession and generational turnover inside ruling coalitions. When an incumbent weakens, a managed transition approaches, or a younger cohort of officials and business figures seeks distinguishing claims, actors need evidence they can use to demonstrate competence or to renegotiate internal bargains. Metric bodies that already possess a track record of narrow, factual output become valuable precisely then. New organisations created in the moment lack both the time series and the reputation for restraint; they are more easily painted as instruments of the contest itself. Bodies that have operated for months or years at district, municipal, or single-sector scale are more likely to be already running when the window opens. They require less fresh permission, face lower start-up risk, and can release the next scheduled comparison without appearing to have been invented for the crisis. National-scale efforts that wait for central authorisation frequently miss these intervals.

The two disciplines reinforce each other. Narrowness kept over time builds the reputation that makes the data usable during turnover. Operation at district or sector scale makes it feasible to maintain that narrowness without constant national bargaining, and places the series in position before the window appears. When both are observed, organised groups possess an information product that rival elites, conditional financiers, and performance-linked brokers can actually deploy. When either is abandoned—when the body widens into advocacy or when it exists only on paper until a national opportunity is granted—the product loses the multi-sided character that gives it leverage. Protests may still draw attention to figures or defend the collectors. They cannot restore usability once the numbers are widely viewed as belonging to one camp, nor can they create a time series that was never built. The practical implication is therefore sequential and restrictive: establish and protect narrow, recurring comparisons at a scale that can be sustained, keep them free of general advocacy, and maintain them so that they are already available when elite succession or revenue stress creates demand for exactly that form of evidence.