When the Norwegian Refugee Council commissioned AfroTech Horizons to conduct a Multi-Sectoral Needs Assessment across 4 districts in Kigoma, one challenge was immediately clear: internet connectivity in many target areas was unreliable at best, absent at worst. Running a 4-district survey with zero data loss required a methodology built for offline-first operations from the ground up.
Here’s how we do it.
Tool Selection: Offline-Capable Data Collection Platforms
AfroTech uses mobile data collection platforms that store responses locally on the enumerator’s device and sync automatically when connectivity is restored. This means field teams can work all day in areas with no signal, and data uploads the moment they return to an area with connectivity — without any manual intervention.
All forms are pre-loaded onto devices before teams enter the field. Form updates, if needed, are distributed remotely and synced at the next connectivity window.
Enumerator Training: The Most Critical Investment
Technology is only as good as the people using it. AfroTech invests heavily in enumerator training — not just on the data collection tool, but on interview technique, informed consent procedures, confidentiality protocols, and quality control.
For the Kigoma assessment, we trained enumerators for three days before any field work began. Training included mock interviews, supervised practice sessions, and a thorough review of every question in the instrument to ensure consistent interpretation across all enumerators.
Quality Control: Real-Time and Post-Collection
AfroTech implements a three-layer quality control system:
Layer 1 — Form-level validation: Required fields, logical skip patterns, and range checks built directly into the data collection form prevent the most common data entry errors at the point of collection.
Layer 2 — Supervisor review: Field supervisors review a random sample of completed interviews each day, flagging inconsistencies for immediate follow-up.
Layer 3 — Back-office cleaning: Upon data upload, AfroTech’s data team runs automated cleaning scripts to identify outliers, duplicates, and inconsistencies before analysis begins.
Lessons Learned
After multiple large-scale surveys in low-connectivity environments, our key lessons are: over-train your enumerators, build more time into your field schedule than you think you need, and never assume connectivity will improve. Design for zero connectivity and treat any connectivity as a bonus.
AfroTech Horizons is available to support your next data collection exercise, however remote the context.