Large-Scale Registration Operations
Four concurrent registration exercises from 2018 to the present — joint verification, continuous registration, headcounting for new arrivals, and biometric identification.
Screenshots of live humanitarian systems are not published. Where visuals appear, they are illustrative rebuilds created with synthetic data that reproduce the structure and design of the original work without exposing any operational or personal information — on data-protection grounds.
Context
Registration in the Cox’s Bazar response is not a single exercise. Several run concurrently, each with its own population, cadence and data requirements, all against the same underlying dataset. This case study covers four of them.
Approach
- Joint Verification Exercise (2018–2019). 800,000 refugees verified and registered in proGres during the emergency-scale joint exercise with the Government of Bangladesh, enabling issuance of identity documentation and access to assistance.
- Continuous Registration (2020 – ongoing). Ongoing registration across a 1.2M population — new birth registration, new registration, family composition, activation and inactivation, biodata updates, and identification of child-protection, legal-protection and specific-needs cases for referral.
- Headcounting for New Arrivals (2024). 200,000 new arrivals. I served as Focal Point, coordinating field teams, supervising enumerators and managing data collection and reporting end to end.
- Biometric Identification Exercise (2024 – ongoing). 200,000 refugees. I built the monitoring dashboard used to track throughput, coverage and data quality throughout.
Scale
These are publicly reported aggregate figures for the Cox’s Bazar response. No camp-level, disaggregated or case-level data appears anywhere on this site.
People registered or verified, by exercise
Publicly reported aggregate totals
Outcome
1.2M population supported
Across these four exercises, which run from 2018 to the present, the work covered a population of 1.2 million, with 800,000 verified during joint verification and 200,000 each through headcounting and biometric identification — delivered across eight registration sites, including repeated field missions to Bhasan Char between 2022 and 2024.
What I’d do differently
The recurring lesson across all four is that the data-quality cost of an exercise is set at collection time, not at analysis time. The exercises that went cleanest were the ones where enumerator training and validation rules were settled before the first day in the field — and where the monitoring dashboard existed on day one rather than being built once problems appeared.