Real-time welfare monitoring in action: Lessons from the ground

Timely information is no longer optional — it is essential. Monitoring welfare in real time can help policymakers adapt more effectively in a world of shocks. But how do these innovations work on the ground? A new World Bank volume, Measuring Welfare When It Matters Most: Learning from Country Applications, draws practical lessons from diverse contexts. Together, these cases illustrate both the promise and the challenges of implementing real-time monitoring approaches for better policymaking.

Decentralized surveys for greater resilience in Malawi

In rural Malawi, the Rapid and Frequent Monitoring System (RFMS) shows how locally embedded enumerators can transform data collection. By hiring and training residents in target villages to conduct monthly surveys, the system provides continuous information on food security, coping strategies, and vulnerability to shocks.

This approach proved invaluable when cyclones struck southern Malawi. Because questions were quickly adapted to capture flood impacts, policymakers had timely insights into which households were most affected. The RFMS also built trust within communities and reduced costs by eliminating the need for costly travel. The trade-off, however, was significant upfront investment to establish a central management system. The Malawi experience highlights that sustainability depends on shared ownership — across government, donors, and communities.

Ethiopia’s rapid pivot to phone surveys

When COVID-19 disrupted face-to-face surveys, high-frequency phone surveys were rapidly deployed in Ethiopia to track the crisis. Over the course of 11 survey rounds, more than 3,000 households were repeatedly contacted, alongside firms and refugees. The surveys generated near real-time insights on job losses, income shocks, food security, and access to services.

The Ethiopian case shows how quickly governments can pivot when crises demand information. But it also underscores limitations: phone surveys risk underrepresenting poorer households without phones, and high attrition over time can erode representativeness. The lesson is clear: phone surveys are powerful tools in emergencies, but they need to be carefully designed and interpreted, ideally linked to recent baselines of face-to-face surveys.

Listening to households in Central Asia

In Kazakhstan, Tajikistan, and Uzbekistan, World Bank-supported listening surveys — nationally representative, high-frequency phone surveys — combine welfare indicators with public perceptions of reforms, which is rarely captured together in traditional surveys. By revisiting the same households monthly, they track short-term changes in income, food security, and well-being, while also gauging trust and legitimacy around reforms. Thousands of households across the three countries participate in the surveys, making them one of the most reliable tools for assessing public opinion.

They have provided valuable lessons: uncovering gender bias in perceptions of fair pay, showing that cash incentives for vaccination can backfire, and demonstrating how text-message campaigns during COVID-19 improved protective behaviors. The design also emphasizes affordability and flexibility, building on a baseline of face-to-face surveys but sustained through low-cost phone panels with a modular questionnaire. 

Using geospatial data for rapid disaster response in Pakistan

When catastrophic floods hit Pakistan in 2022, traditional surveys would have taken months to capture their impacts. Instead, the World Bank combined geospatial flood maps with household survey data and vulnerability models to estimate welfare losses within two weeks — supporting rapid emergency response, resource allocation, and international advocacy.

The Pakistan case illustrates how big data can fill gaps when time is of the essence. Yet it also highlights important caveats: results depend on model assumptions and need to be validated with ground data. Geospatial tools are powerful complements, but not substitutes, for conventional surveys.

Five key lessons from country experiences

What do these country applications teach us? Several lessons stand out:

1. No method is one-size-fits-all. Different methods work better in different contexts to answer different questions. Policymakers need a toolkit, not a single method.

2. Baselines remain essential. All approaches depend on having recent, reliable survey data to anchor estimates. Without strong baselines, even the most innovative methods risk producing misleading results.

3. Trade-offs must be managed. Speed, cost, coverage, and representativeness cannot all be maximized at once. Each approach comes with advantages and drawbacks that need to be carefully balanced against policy needs.

4. Partnerships build sustainability. From Malawi’s locally hired enumerators to Ethiopia’s collaboration with private survey firms, success depends on strong partnerships across governments, communities, and development partners.

5. Use drives value. Data has the most impact when it is actually used — whether to guide cyclone response in Malawi, shape COVID-19 relief in Ethiopia, or inform reforms in Central Asia. Real-time systems must be designed with a clear eye to decision-making needs.

Frontier innovations and the road ahead

These country experiences highlight both the opportunities and limits of real-time monitoring. Researchers are testing frontier approaches using novel data sources such as satellite imagery, mobile phone metadata, and digital trace data. These methods are already showing considerable promise. For example, one model leveraged new advances in deep learning methods to incorporate news reports to better predict food crises, significantly outperforming both traditional models and expert forecasts.

But challenges remain: dynamic changes in welfare over time are harder to capture than cross-sectional updates, risks of bias are real, and uncertainty needs to be clearly communicated. Innovation is expanding the frontier of what can be measured, but careful validation and continued investment in traditional surveys remain essential.

For policymakers, the message is clear: while real-time monitoring is not a substitute for strong statistical systems, it is an increasingly important complement that can strengthen the evidence base for adaptive and resilient policies.

The full set of experiences and insights can be found in the World Bank publication, Measuring Welfare When It Matters Most: Learning from Country Applications, available for download here.