Performance dashboards
1. Introduction
In OpenCRVS, performance dashboards surface near real-time performance and vital statistics data using configurable dashboards powered by Metabase.
They are used by operational managers, programme leads, and policy makers to monitor workload, coverage, timeliness, completeness, and data quality across locations and over time. Dashboards draw on aggregated, de-identified data rather than individual records, helping to answer questions such as how many events were registered, how quickly, and where there are gaps in coverage.
2. Feature overview
Performance dashboards provide a near real-time view of how the CRVS system is performing, using aggregated, de-identified data.
Core capabilities
With performance dashboards, OpenCRVS supports:
Configurable metrics and visualisations built on analytics-ready datasets from OpenCRVS.
Policy and planning insights based on trends in births, deaths, and other events.
Role- aware views so users see dashboards relevant to their responsibilities and jurisdiction.
Operational monitoring of workload, bottlenecks, and data quality by office, role, and event type.
Programme oversight for coverage, timeliness, and completeness of registration across locations and time periods.
Performance dashboards are:
Aggregated and de-identified — driven by analytics fields only, not raw personal data.
Metabase-powered — implemented as Metabase dashboards connected to OpenCRVS data sources.
Extensible — countries can add new charts, filters, and dashboards as their CRVS monitoring needs evolve.
3. Data sources and metrics
Performance dashboards rely on structured data from OpenCRVS, including only fields that have been explicitly marked as suitable for analytics (that is, form fields with the analytics = true property). This ensures that personally identifiable information (PII) is excluded from dashboard datasets.
They draw on:
Event registrations — births, deaths, marriages, and other event types.
Key timestamps — date of event, date of declaration, date of registration.
Locations — where events occurred, where they were declared, where they were registered.
Statuses and flags — Draft, Notified, Declared, Registered, Archived; late registrations; correction requested.
User and office information — which office and which role performed key actions.
From these data, common metrics include:
Volume — number of registrations by event type, time period, and location.
Timeliness — time from event to registration (for example, within 30 days, 31–365 days, > 1 year).
Completeness / coverage (proxy) — registrations per 1,000 population or compared against targets (where reference data is available).
Data quality — share of records rejected, corrected, or with missing key fields.
Process efficiency — time spent in each status (Declared, Validated, Registered), number of rejections or escalations.
Countries can decide which metrics to prioritise based on their CRVS strategy and available reference data.
4. Configuration overview
Performance dashboards in OpenCRVS are delivered using Metabase, an open-source business intelligence and analytics platform. Dashboards are created and managed in Metabase and then securely embedded within the OpenCRVS application.
From a user perspective, dashboards appear as part of the OpenCRVS interface. From a configuration perspective, they are defined and maintained separately in Metabase.
How dashboards are integrated
Dashboards are:
Embedded in the OpenCRVS UI via links in the side navigation.
Access-controlled by scope and role, ensuring users only see dashboards and data relevant to their jurisdiction (for example national, provincial, or district level).
What is configured in Metabase
Configuration primarily happens inside Metabase and includes the following components.
Dashboards
Collections of charts that answer specific operational or policy questions (for example registration volumes, timeliness, or completeness).
Each dashboard typically aligns to a use case such as:
Completness rates
Registrations
Operations monitoring
Data quality
Datasets / models
Analytics-ready tables or models that structure OpenCRVS data for reporting.
These commonly include:
Event “fact” tables (registrations, actions, timestamps)
Dimensions (event type, date, location hierarchy, office, role, status)
Pre-calculated indicators (for example timeliness buckets or monthly aggregates)
Using curated models improves consistency, performance, and reuse across multiple dashboards.
Metrics (data visualisation)
Charts and tables that present metrics in an interpretable way, such as:
Time series trends
Bar or stacked charts by location or event type
KPIs and summary cards
Tables for operational detail
Metrics are built on top of datasets and reused across dashboards where needed.
Filters
Interactive controls that allow users to narrow the data displayed, for example:
Date of event
Event type
Event location
Registration location
Filters can apply to individual charts or entire dashboards to support flexible exploration.
Access control
Role-based permissions determine:
Which dashboards a user can open
What event data they can view based on their jurisidciton ?!?!!?! (think there is a complex solution)
This ensures that sensitive operational insights are visible only to authorised users while supporting decentralised monitoring.
Extending and maintaining dashboards
Dashboards are designed to evolve as country needs change. Teams can:
Add new charts or metrics
Create additional dashboards for new programmes
Refine models as data structures mature
Introduce new analytics fields as forms are updated (with
analytics = true)
Because dashboards are configuration-driven rather than hard-coded, updates can be made without changes to the core OpenCRVS application.
< Example dashbaords >
5. Use and interpretation
Performance dashboards should be interpreted in context:
~~Data lags — some indicators may be based on partial data if registrations are still being processed.~~
Population denominators — coverage estimates depend on the quality of population estimates or target numbers.
~~Operational factors — spikes in workload or outages can affect short-term trends.~~
Dashboards are most powerful when used as part of a regular review and action cycle, for example:
Monthly or quarterly performance review meetings.
Joint analysis sessions between CRVS, health, and statistics stakeholders.
Targeted support or supervision for offices with persistent performance or data quality issues.
OpenCRVS provides the underlying data and integration to support these dashboards; each country decides which indicators to track, how often to review them, and what actions to take based on the insights.
Last updated