Community health monitoring and evaluation at scale
A government healthcare programme needed to scale community health monitoring from 2 blocks to 7 districts across Uttar Pradesh—tracking 589 mobile health workers' performance in real time without centralised oversight. Catholic Relief Services (CRS) and Sarathi Development Foundation (SDF), in collaboration with the National Health Mission, implemented
ASHA Sangini Supportive Supervision Application (ASSSA) on Infoveave: a live interactive health monitoring dashboard with 80 tracked KPIs, 6 custom report formats, and multi-level role-based access. Supervisors could benchmark performance against peers and make data-driven decisions within hours instead of weeks.
At a glance:
- Industry: Healthcare — community & population health M&E
- Scale: 7 Uttar Pradesh districts · 74 blocks · 589 mobile users
- Platform: Infoveave BI — real-time supervision dashboards, peer benchmarking, bilingual reporting
Multi-Level
5-tier hierarchical access (Individual ASHA → Supervisor → Block → District → State) with role-based dashboards
Real-Time
Live dashboards tracking 80 KPIs with peer benchmarking against peers and district/state averages
Automated
6 custom bilingual report formats—from ASHA daily activity to state monthly trends
In this story:
Client Overview
The story began with ReMiND (Reducing Maternal and Newborn Deaths), a supportive-supervision pilot run in 2 blocks of Kaushambi district between 2012 and 2015. Its model was promising: give frontline health supervisors timely evidence, and they could coach ASHA workers more effectively. The next challenge was to make that model work across a much larger and more complex programme.
Catholic Relief Services (CRS) funded and oversaw the scale-up, while Sarathi Development Foundation (SDF) led implementation on the ground. The National Health Mission (NHM), Uttar Pradesh provided government alignment and data integration, and Noesys supplied the Infoveave unified data platform. Together, the partners created the ASHA Sangini Supportive Supervision Application, or ASSSA.
The move from pilot to programme changed the scale of monitoring. The system had to serve 589 active mobile application users and 523 ASHA Sangini supervisors across 74 blocks in 7 districts—Bareilly, Faizabad, Kannauj, Kaushambi, Lucknow, Mirzapur, and Sitapur—supporting maternal, newborn, and child health services for a population of approximately 1.77 crore. What had worked through close coordination in 2 blocks now needed a dependable monitoring and evaluation system that could operate across an entire district hierarchy.
The Challenge: Fragmented M&E at Scale
Before ASSSA, supervision data moved through manual collection and reporting cycles. Information about pregnancy care, maternal deaths, infant mortality, and ASHA activity sat in separate datasets. By the time teams consolidated that information into reports, the data could be weeks old. Supervisors were looking backward when they needed to decide where to intervene next.
Scaling made the delay more consequential. Every day, 589 mobile users generated transaction-level records across 74 blocks. Yet an ASHA Sangini could not readily compare her team's performance with another supervisor in the same block. Block coordinators lacked a consolidated view across supervisors, while district and state teams needed separate reports prepared for their level of the hierarchy.
Reporting also had to fit the way the programme actually operated. NHM followed an April–March fiscal calendar and used government-standard terminology. Field teams needed Hindi, while programme and funding stakeholders also required English. Without one governed analysis layer, differences in names, classifications, and reporting formats could travel up the hierarchy and weaken confidence in the results.
The programme therefore needed more than a dashboard. It needed an M&E system that could turn daily field data into consistent measures, preserve the context of each administrative level, and shorten the distance between an emerging performance gap and a supervisory response.
Business Requirements
The requirements followed directly from those operating realities. At the centre was a common definition for more than 80 KPIs covering ASHA activity, maternal mortality, infant and child mortality, and high-risk pregnancy. Those measures had to remain consistent whether a user viewed one worker, a block, a district, or the programme as a whole.
The same data also had to answer different questions. A supervisor needed to see which workers required coaching and how her team compared with peers. A district manager needed to identify geographic patterns and mortality hotspots. State administrators and programme funders needed trends they could trust without waiting for a new round of manual consolidation.
That led to three practical design conditions: daily integration from the CommCare-based mobile application, 6 report formats that followed the programme hierarchy, and native Hindi and English presentation using government terminology. Automation would only be useful if the resulting information felt familiar to field users and remained comparable at every level.
Ready to see Infoveave in action?
Book a personalised demo and discover how Infoveave can unify your data and drive measurable outcomes.
Book a DemoThe Solution: ASSSA on Infoveave
Infoveave BI was selected because it could combine low-code configuration, bilingual presentation, geospatial analysis, and role-based reporting in one environment. It also brought relevant health-programme experience from an earlier CRS Ethiopia deployment for the Joint Emergency Operations Programme.
From mobile activity to governed health data
The data journey started in the CommCare-based ASSSA mobile application. Across 74 blocks, users recorded pregnancy registrations, antenatal care, institutional deliveries, newborn visits, immunisation, high-risk pregnancies, and maternal or infant deaths. Talend ETL scripts extracted those transactions daily and loaded them into a MySQL data warehouse.
The warehouse did more than store records. Shared dimension tables aligned block names with census standards, consolidated facility types, and standardised ANC complication categories. Fact tables organised ASHA activity, maternal mortality, infant mortality, and high-risk pregnancy data. Explicit codes distinguished missing values from incomplete records, while age-at-death rules placed cases into consistent reporting bands.
These controls mattered because a small classification difference could alter a rate when data rolled up from worker to block, district, and state. Central data-quality rules also prevented loss-to-follow-up cases from distorting performance measures. The result was a governed foundation on which every level of reporting could rely.
Turning the same data into decisions at five levels
Infoveave connected to the warehouse and calculated more than 80 KPIs on demand. Some were deliberately derived rather than copied from a source field. For example, the percentage of ASHAs completing a first-day newborn visit depended on CASE-WHEN logic across more than 10 joined tables. Defining that calculation once meant every dashboard and report used the same interpretation.
The insights and data visualization layer then adapted those measures to a five-tier hierarchy: individual ASHA, ASHA Sangini supervisor, block coordinator, district manager, and state administrator. Supervisors could move from a team-level comparison to an individual worker. District teams could use GIS heatmaps to locate maternal or infant mortality patterns. State users could follow trends without losing the ability to drill into the contributing districts and blocks.
Six automated report formats mirrored that hierarchy. They ranged from daily or monthly worker-level views to Sangini summaries, quarterly trends, block-to-supervisor comparisons, district-to-block reports, and a state-to-district view. Alert indicators drew attention to daily, weekly, or monthly anomalies, while peer benchmarking showed each supervisor how her team compared with other supervisors and with district or state averages.
Designing for use in the field
Hindi was treated as part of the product design, not as a translation added after implementation. All 80 KPIs were documented in both Hindi and English, with separate report templates and government-standard terminology. This allowed block-level supervisors and government staff to work directly with the dashboards rather than depending on an English-speaking analyst or IT team to interpret them.
Planning and requirements were completed in November 2018, including the KPI definitions, report hierarchy, and bilingual specifications. The initial Talend, MySQL, Infoveave, and user-acceptance work followed a 9-week development cycle. Rather than treating launch as the end of the project, the partners continued refining the system through the programme period to mid-2023.
80 KPIs Tracked Across 4 Domains
The 80 KPIs gave the programme a connected view of frontline activity and population health outcomes. They were organised into four domains so that users could move from an operational signal—such as a missed visit—to the wider maternal or child health pattern it might affect.
ASHA worker performance
Ten monthly activity measures covered the work supervisors were expected to coach: first-day newborn visits after home deliveries, home-based newborn care, immunisation and VHND attendance, institutional delivery support, childhood illness management, nutrition counselling, malaria and fever cases, DOTS support for tuberculosis, VHSNC participation, and family-planning referrals.
Maternal and child mortality
Maternal death surveillance distinguished deaths during pregnancy, the 0–42 day postpartum period, other causes, and complication types. Infant and child mortality measures added gender, age bands, place of birth, and place of death. Because each record retained its geographic hierarchy, district teams could see whether a concerning pattern was concentrated in a particular block rather than hidden inside an overall total.
High-risk pregnancy
High-risk pregnancy monitoring followed identification, previous and current complications, antenatal follow-up, birth preparedness, pregnancy outcomes, and the mother's post-delivery condition. It also surfaced a useful absence of data: an ASHA reporting no high-risk pregnancies could indicate a genuine result or an identification gap that needed supervisory attention.
One definition from field entry to state reporting
Every KPI came from the same standardised dimension and fact tables. Missing and incomplete values remained distinguishable, quality rules ran centrally during ETL, and historical trends followed the April–March government financial year. This governance made comparisons meaningful: a supervisor and a state administrator could view different levels of detail without using different versions of the measure.
Results and Impact
From reporting delay to a supervision rhythm
The most immediate change was time. Data that previously took weeks to consolidate became available within hours of field entry. That speed changed what a supervision conversation could be.
One ASHA Sangini noticed that a worker on her team had stopped making first-day newborn visits. She contacted the worker that afternoon rather than waiting for a monthly report meeting. It turned out the worker had started a small business and was deprioritising health visits. They reviewed the data together—including how her performance compared with peers—and problem-solved. The gap closed within a week. Before ASSSA, the same shortfall would have remained invisible for three weeks.
Block coordinators gained a cross-supervisor view within 24 hours. District managers could investigate geographic hotspots instead of waiting for a manually assembled report. In Kannauj, the heatmap showed a cluster of high-risk pregnancies in a sub-block where villages sat 8–15 km from the primary health centre. ASHA workers were identifying high-risk cases, but referral-to-care was low—the barrier was transport, not awareness. The programme partnered with a local transport operator to provide subsidised rides. Within three months, referral-to-care in that sub-block rose from 34% to 71%, visible as diminishing intensity on the same heatmap.
This changed reporting from a retrospective administrative task into part of the supervision cycle. Automated consolidation reduced spreadsheet handling and gave supervisors more time for field coaching. At higher levels, the same governed dataset supported district and state reviews, so decisions no longer depended on reconciling several versions of a report.
Peer benchmarking added another layer of accountability—and a different kind of motivation. Each supervisor could see her team's performance alongside peers in the same block and against district or state averages. The comparison was powerful because it was transparent and relational. When one Sangini saw that another team's first-day newborn visit rate was 92% while hers was 71%, the numbers did two things at once: they proved 92% was achievable, not theoretical, and they raised a practical question—what is she doing differently? Knowledge began to move between supervisors. Performance gaps closed because peers wanted to match peer performance, not only because a target said they should.
The bilingual, government-aligned reports also reduced friction between field implementation and formal review. NHM users could work with April–March reporting periods and familiar terminology, while CRS and SDF retained visibility across all 7 districts. Information could move through the hierarchy without being translated, reformatted, and recalculated at every step.
Scale and sustainability
The scale-up was designed to spread infrastructure costs across a much larger population. Its incremental annual cost was estimated at INR 4.39 per capita (US $0.07), compared with INR 31.4 per capita for the 2-block pilot—a 96% reduction attributed to infrastructure amortisation. The estimate was equivalent to approximately 6% of Uttar Pradesh's annual Reproductive and Child Health budget and 3% of the WHO-recommended investment for universal maternal, newborn, and child health coverage.
The health-impact figures associated with this programme come from the earlier ReMiND model, not from the ASSSA dashboard alone. A peer-reviewed evaluation by Prinja et al. (2018) in BMC Pregnancy and Childbirth projected that the ReMiND intervention could avert 312 maternal deaths and 149,468 neonatal deaths over 10 years, at INR 12,993 (USD 205) per DALY averted. ASSSA's role was to extend the monitoring and supervision infrastructure behind that model from 2 blocks to 7 districts; it should not be read as an independent attribution of those projected deaths averted to Infoveave.
Post-launch refinement
The system continued to evolve as field teams used it. Fifteen user-driven change requests shaped the product after go-live. Several of those requests show how closely the platform had to track field and government reality: aligning block names with government census standards so state reports matched official records; consolidating facility types to match NHM classifications; improving age-at-death categories for more precise mortality analysis; and correcting Hindi terminology to match how supervisors actually spoke about health work. Later changes also strengthened PDF reporting, audit logging, KPI drill-downs, and mobile responsiveness.
That refinement is part of the outcome. Large-scale population health monitoring did not succeed because the programme published one fixed dashboard. It succeeded because the partners created a governed feedback loop: field data informed supervision, supervision exposed new information needs, and those needs shaped the next version of the monitoring system.
Related success stories:
Ready to scale your health programme M&E? See how ASSSA automated supervision across 589 mobile health workers and 74 blocks. Book a demo to learn how Infoveave can unify your data, enable peer benchmarking, and drive measurable health outcomes.
About the Authors
Infoveave Product Team is a member of the Infoveave Product and Solutions Team — specialists in Unified data platforms, agentic BI, and enterprise analytics. Infoveave (by Noesys Software) helps organizations unify data, automate business processes, and act faster with AI-powered insights.
Ready to revolutionize your data journey with Infoveave?
Related Success Stories
- Infoveave collaborates with Project Balance & Catholic Relief Services (CRS)
- Infoveave builds dashboards for ICAP to monitor HIV care & treatment in Kenya
- Program effectiveness and survival analysis
- Centralised Healthcare Data Transformation Across 17 Hospital Sites
- Infrastructure Estimating on a Unified Data Platform (Australia)
Frequently Asked Questions
Ready to see Infoveave in action?
