When data is incomplete or of poor quality, even the best decisions risk producing waste, inefficiencies and loss of trust. Data governance is a concrete policy lever and analytics such as scenarios and ex-ante assessments help reduce allocation errors and reputational risks.
Some decision-makers—whether political or corporate—know this well: relying on partial or incorrect numbers means exposing oneself to waste, litigation and loss of trust from citizens, markets and regulators. In 2022, for example, a coding error in Equifax's systems, one of the world's three main credit reporting agencies, led to incorrect credit scores being sent to millions of consumers, with impacts on mortgages and loans and consequent legal action.
In the public sector, the European Commission estimates billions of euros wasted on environmental policies built on incomplete data.
When data works
The good news is that data, when well governed, changes policy outcomes. New York City uses predictive models to identify buildings at highest fire risk, concentrating inspections where they are truly needed and significantly increasing inspection effectiveness.
Cities like Singapore are experimenting with AI-based traffic management systems using real-time data, reducing travel times during rush hours and improving public transport efficiency. Here data quality becomes invisible infrastructure: fewer accidents, less congestion, fewer emissions.
How to avoid waste: the Sicily case
In 2024 the Energy Department of the Sicily Region asked OpenEconomics to assess ex-ante a new biomethane incentive for agricultural SMEs, to avoid over-financing oversized and inefficient plants. We cross-referenced tax, cadastral and satellite data to map approximately 2,800 potential plants, simulated three subsidy scenarios with a dynamic cost-benefit model and estimated the effects on emissions and employment by 2030.
The decision-maker chose the "mid-impact" scenario, containing spending, saving approximately 46 million euros of ERDF funds while still obtaining an estimated 18% reduction in emissions and approximately 1,250 new jobs. In practice, the same public euros generated more environmental and social value, with lower risks of challenges from citizens, the Court of Auditors and European institutions.
The three critical moments of decisions
In the cycle of a policy or an investment there are three junctures where data makes the difference:
• Problem definition: if the information framework is partial, needs are inflated and priorities are wrong. Integrated and verified datasets reduce bias and "blind spots.
• Measure or plan design: without quantitative scenarios there are risks of distorted incentives, over-budget and difficult-to-justify projects. Counterfactual models and sensitivity analysis allow testing options before committing real money.
• Monitoring and revision: with slow or closed data the course cannot be corrected and mistrust grows. Clear KPIs, updated almost in real time and, in the public sector, open data, make it possible to adjust course and explain choices.
The European data push
The European data strategy aims to create interoperable sectoral data spaces, with an expected impact of hundreds of billions of euros of additional GDP by 2030. Regulations such as the Net-Zero Industry Act and the Better Regulation guidelines require ex-ante quantitative assessments based on linked administrative micro-data, not on approximate averages.
For public administrations this means demonstrating with solid numbers why a measure deserves funds; for companies it means being able to show investors, banks and regulators that industrial plans and ESG strategies rest on credible scenarios.
What public administrations and companies can do
Whether it is a department, a utility or a large industrial group, the path starts with a few concrete steps:
• Conduct an inventory of existing databases, identifying gaps and inconsistencies that currently weaken decisions.
• Link an essential set of KPIs (Key Performance Indicators) to declared political or business objectives, defining from the outset how they will be measured.
• Introduce at least for the most relevant dossiers an ex-ante assessment based on scenarios and quantitative models, to be used later also in monitoring.
As we've learnt over the last years working with customers in complex sceenarios, for policy makers or company leaders investing in data quality and analytics is not a technological luxury: it is insurance against strategic errors, budget waste and reputational crises that are difficult to report.
.png){kind=small}
.png){kind=small}