Beneficiary/service data collection

Sound beneficiary data practice starts from one rule: register a person once, and reference that one record every time they receive a service or take part in an activity, rather than creating a fresh record at every touchpoint. Failing to do this is the single most common cause of inflated reach figures in programme reporting — not dishonesty, just an architecture that made double-counting easy.

A related discipline is capturing a person's identifying and demographic attributes (name, sex, age, location, and whatever else the programme genuinely needs) once, at registration, rather than re-asking and re-recording them at every subsequent service — which both wastes field time and creates the risk that the same person's recorded age or location silently disagrees with itself across different records.

A beneficiary is registered once per organization and then joined to one or more projects; service and event records reference that one person, so "unique beneficiaries reached" and "service events delivered" stay two genuinely different, correctly labelled numbers.

Geography, gender, disability and other attributes are captured at registration and carried through every downstream report and dashboard without re-entry.