For more than 10 years, Kenya’s understanding of county performance has been primarily shaped by perception-based polls. Organisations such as Infotrak have played a helpful role in capturing citizens’ mood, offering a snapshot of how people feel about the service delivery they receive from county governments. These polls matter; public perception influences trust, legitimacy and, ultimately, political stability. Yet, perception alone is rarely a reliable proxy for performance. Feelings shift with headlines, politics and short-term experiences, while the underlying realities of governance often move more slowly and more quietly.
At CyMetricA, we believe the country is ripe for a new chapter — one where county performance is measured not just by what people think, but by what the data shows. This requires a shift from opinion to evidence, from episodic surveys to annualised analytics, and from static, decade-old datasets to dynamic, frequently updated ones. Our work over the past two years has focused on building a new county ranking methodology that does exactly this, using a fusion of proxy data, administrative records and satellite-derived insights. It is a fundamentally different approach from traditional polling and has the potential to redefine public accountability in Kenya.
The conceptual heart of our method comes from an unexpected source: the County Development Index (CDI) developed by the Commission on Revenue Allocation (CRA). Originally designed to identify marginalised counties for equitable funding, the CDI is a composite index built on a thoughtful blend of socioeconomic indicators. It is elegant, mathematically coherent and grounded in development theory. But it is also constrained by the nature of its inputs. Most CDI indicators are drawn from the national census or the Kenya Integrated Household Budget Survey — datasets that appear once every 5–10 years. This makes the CDI excellent for its original purpose, but poorly suited to annual ranking or performance tracking.
We saw an opportunity to borrow the architecture of the CDI — its structure, fairness and multidimensionality — and re-engineer it for the modern era. Rather than relying on static census values, we incorporate annually refreshed proxy indicators that move alongside real changes in county service delivery. For health, for instance, we turn to DHIS2’s annualised summaries of facility-level outcomes. For education, we look at KNEC’s examination results, which capture the effectiveness of local schooling systems. For infrastructure, we integrate the Kenya Roads Board’s yearly road inventory alongside electricity access data from EPRA. We complement these with economic signals from KNBS. We are also trying satellite-derived night-time light intensity from NASA’s VIIRS system, which serves as a powerful proxy for urbanisation, electrification, and economic vibrancy.
Introducing recency through proxy data enables us to build an annually updated model, providing a more honest reflection of county performance than is currently possible. Trends become visible. Improvements or declines become measurable. Policymakers gain a clearer understanding of where resources are working and where they are not. And because the method is grounded in real data rather than sentiment, it offers a stable benchmark that is less vulnerable to political noise.
The methodology itself is simple in concept but sophisticated in execution. It begins with aggregating multiple annual datasets from administrative systems, national statistics offices, and remote sensing sources. These are harmonised, cleaned and transformed into comparable metrics. Each indicator is then scaled to ensure consistency across counties and across years. The indicators are grouped into major dimensions such as Health, Education, Infrastructure and Economic Wellbeing, each synthesised through a weighted composite. Finally, an algorithmic scoring model blends the dimensions into a single annual performance index for each county.
We intentionally avoid disclosing the specific weighting schemes, proxy selection mechanisms, temporal smoothing techniques and calibration strategies we employ. These form part of CyMetricA’s intellectual property and represent a possible future investment asset. What we can say is that the system is designed to be robust, scalable and grounded in best-practice composite index theory.
The need for an approach like this is becoming increasingly apparent. East Africa is producing more data than ever before, yet much of it remains underused. Agencies publish valuable information in reports that few people analyse deeply. Counties generate administrative data that rarely feeds into comparative assessments. Satellite data offers extraordinary insight but is seldom operationalised in governance contexts. CyMetricA was designed to bridge these gaps, transforming raw data into actionable intelligence.
Of course, the path is not without challenges. Kenya’s data ecosystem remains fragmented, with public datasets scattered across agencies that do not always release information in aligned formats or on consistent timelines. Some county-level statistics are incomplete or irregular, requiring careful methodological safeguards. Administrative data quality varies significantly, and specific indicators remain partially inaccessible to the public. Even satellite-derived data, despite its richness, can be noisy or prone to seasonal distortions if used without proper correction.
Yet we see these challenges as opportunities. They push us to innovate — to design better pipelines, negotiate more effective data-sharing frameworks, and build models resilient enough to handle imperfect inputs. More importantly, they compel us to advocate for improved data governance practices across the region.
What we are building at CyMetricA is more than a ranking. It is a new evidence layer for governance. It has the potential to support annual county performance scorecards, objective governor evaluations, more informed media reporting, donor and investor decision-making frameworks and even county development compacts. Most importantly, it empowers citizens with a clearer understanding of real progress — not just perceived progress.
The future of county analytics in Africa will belong to those who can transform complexity into clarity, data into insight and insight into action. At CyMetricA, our mission is to develop models that reflect today’s realities rather than yesterday’s assumptions. By adapting the CDI framework for annualised performance tracking and infusing it with modern proxy data, we are taking a bold step toward a more transparent, evidence-based public sector.
This, we believe, is the future of county performance measurement. And we are only just beginning. Keep checking back for updates, or subscribe to receive regular progress updates.