evidoria

← Back to browse

Good practice Imported

SITA — Nepal's AI Platform Turning Months of Survey Analysis Into Seconds

Nepal · Kathmandu · See the Nepal profile · See the Kathmandu profile

Evidence: Descriptive / self-reported Top 62% 53/100 · Ask Evidence Copilot about this practice

Nepal's Ministry of Health and Population and UNFPA, with Google and the British Embassy, built SITA, an AI platform for querying national survey data in plain language. A pilot on the Demographic and Health Survey matched human analysis 100% of the time, halving analysis time.

100%
Match rate vs human-coded analyses
50%
Estimated cut in analysis time
20%
Estimated reduction in survey cost
SITA — Nepal's AI Platform Turning Months of Survey Analysis Into Seconds

Details

Maturity
Pilot
Promoter
UNFPA Nepal / Ministry of Health and Population
Period
2025–2026 (pilot to Phase III upgrade)
Keywords
public health statistics, digital government, data analytics, survey research

Context

SITA ('Unlocking Data with AI') is a domain-aware AI platform built for Nepal's Ministry of Health and Population by UNFPA Nepal with UNFPA's global IT Services Office, Google and the British Embassy Kathmandu, unveiled in Kathmandu on 31 March 2025 after more than a year of development.

Results

SITA lets a planner type a plain-language question about a large dataset -- such as the Nepal Demographic and Health Survey -- and returns polished charts, tables and narrative text within seconds. In the pilot evaluation, SITA's outputs matched 100% of a parallel set of human-coded analyses, while UNFPA estimated a 50% cut in analysis time and a 20% reduction in the cost of running large surveys.

Conclusions

The project is still an early, single-country pilot reported mainly by the implementing partners themselves; UNFPA Nepal put a Phase III upgrade out to tender in mid-2026 to extend the tool to more ministries, universities and sectors.

Implementation
Indicative cost
Medium (€50k–€500k)
Time to results
Short (< 1 year)
Staffing & skills
UNFPA Nepal, UNFPA global IT Services Office, Google, British Embassy Kathmandu, Dryice Solutions (local developer)

Conditions for success

  • domain-aware AI trained on health survey data
  • plain-language query interface replacing weeks-long manual statistical write-ups

Commonly funded by

National / regional programmes Digital Europe Programme

Indicative funding routes for practices of this type — always check each programme's current calls and eligibility rules.

Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.

Data sources

Where this practice's information was retrieved from, and when.

Attachments

Similar practices you may find useful