How do you digitise a civil registration system?

I've walked into a lot of registration offices. The picture is often the same: shelves of hard-copy registers, some decades old, some quietly rotting, and a registrar doing their best with a system that was never designed for the population it serves.

Digitising that is very doable. It just isn't mainly a technology problem.

The short answer: digitising a civil registration system means moving the registration of births, deaths and other life events onto a secure digital platform, then connecting it to health, identity and statistics systems. The usual sequence is legal review, scoping, configuration of a CRVS platform, a pilot in a few locations, then national rollout, with historical records digitised as a separate workstream. A pilot can go live in about three months. National rollout typically takes six to 12 months. Legislation and procurement are the usual causes of delay.

Key facts:

• OpenCRVS's 2020 Bangladesh pilot ran in two locations, one rural and one peri-urban, for 12 months (UN ESCAP)
• That pilot took registration to communities through a mobile app used by community health assistants (UN ESCAP)
• Pilot to go-live: about 3 months. National rollout: 6 to 12 months

The steps

1. Check the law. Does current legislation allow digital registration, digital signatures and digital certificates? If not, start the amendment now. It's the longest pole in the tent.
2. Scope the first release. Which life events, which districts, which channels, which one integration. Keep it tight. See choosing a first use case.
3. Choose the platform and partner. A CRVS digital public good plus an implementation partner with regional experience.
4. Configure. Forms, workflows, certificates, languages, user roles and your country's administrative geography.
5. Pilot. Two or three locations, with real registrars and real families.
6. Integrate. Health facilities for birth and death notifications, then ID, then statistics.
7. Roll out nationally, district by district, with training at every step.
8. Digitise historical records as a parallel workstream, not a prerequisite.

The blockers that actually slow you down

Legislation. Some countries need to update their civil registration or identity law before a new system can go live. I've seen that lag run to years. If your law still assumes a paper register and a wet signature, fix that first.

Procurement. Many governments, particularly in frontier and emerging markets, aren't set up to procure open source. Their templates assume a licence, a vendor and a long contract. Changing procurement rules is often a political job, not a technical one.

Connectivity. Rural registration points may have poor or no internet. The system needs to work offline and sync later.

People. Registrars have done things one way for decades. Training, support and visible benefits for them are what make adoption stick.

My take: if your project plan has more technical risks than legal and political ones, it's probably wrong.

Where should registration happen?

One of the biggest wins of digitisation is moving registration closer to people. Instead of families travelling to a district office, births can be declared at health facilities or by community agents on mobile devices, then validated by a registrar.

That's what the Bangladesh pilot tested. It's also where most of the gains in registration rates come from.

What about the old paper records?

They matter, but don't let them block go-live. Start registering new events digitally, and treat historical records as their own project with their own budget. It's a big enough job that I've written a separate guide on it.

How long does it take?

More detail in how long a CRVS implementation takes.

Sources

• OpenCRVS goes live in Bangladesh, UN ESCAP

Planning a CRVS digitisation programme? Let's talk.