In a previous post I wrote about there being more Facebook users than Internet users in South East Asia. I also said that this was not the case in South Asia.
But I was wrong. I had relied on data from Bangladesh, India, Pakistan and Sri Lanka. It was only recently that I looked at the data for Nepal. Looks like Nepal lines up with the SE Asian countries on this issue.
At the Ford Foundation supported course we had a good discussion on this. But no explanation that we could all agree on.

4 Comments
Ujjwal Acharya
Thanks for writing about Nepal but your statement is, sadly, based on mismatched data and thus I find your statement invalid.
The number of Facebook users in Nepal that you have posted (4 million) is data of around August, 2014. The ITU data was probably recorded sometime in 2012 when Nepal’s internet penetration was ranged around 15 per cent.
According to Nepal Telecommunications Authority, in August, 2014, the total internet subscribers in Nepal was 8,946,569. (I understand there are some duplication due to multiple SIM subscription and there are multiple users to fixed connections to homes and office, but we can safely assume that number of Internet users was around 8 million).
If you compare data from around same time, its sensible; but if you compare data from August 2014 to data from 2012, it’s going to give you false comparison and a non-sensible finding such as “Nepal is only South Asian country where Facebook users outnumber Internet users”.
Rohan Samarajiva
Good point re data reporting points.
We were not reporting “subscribers” which is what is reported by NTA. We were reporting “users” which is a category that relies on demand-side data, or is calculated by applying a multiplier to the reported subscriber data. The ITU Internet user number does not differ from what the LIRNEasia method yields.
However, do you accept as fully accurate the NTA numbers? Seems to be based on a very broad definition.
Ujjwal Acharya
I don’t completely rely on data provided by NTA but I also don’t rely on appropriate data given by Facebook to a potential advertisers. FB data is more unreliable than data of NTA because simply you don’t know what definition they apply to their users.
Shreedeep Rayamajhi
The data provided by NTA is very arguable in regards of active users which is more or less subjected to subscription. Until and unless we have specific provision of life span or definition of active SIM, its very hard to determine the actual active user rate.
Request for Proposals: Development of a website for the Asia AI4D Observatory
LIRNEasia is inviting proposals from qualified web development firms, service providers and consortia of individual professionals to design, develop and deploy a public-facing website for the Asia AI4D Observatory. The website is intended to serve as a searchable, interactive regional knowledge repository and engagement platform, making evidence and resources on responsible artificial intelligence easy to discover, use and share.
LIRNEasia is hiring: Intern – Data Annotation
LIRNEasia is looking for a talented individual to join the team as an Data Annotation Intern. The full job description is available below.
Sri Lanka at WAIC 2026: Aligning with a More Inclusive Vision for Global AI Governance
The World Artificial Intelligence Conference (WAIC) 2026 was held in Shanghai from 17–20 July and brought global attention to the rapid development of artificial intelligence (AI), as well as the increasingly important question of who gets to shape the rules governing it. This analysis examines the alignment between Chinese President Xi Jinping’s speech and Sri Lanka’s Deputy Minister of Digital Economy Eng.
Links
User Login
Themes
Social
Twitter
Facebook
RSS Feed
Contact
9A 1/1, Balcombe Place
Colombo 08
Sri Lanka
+94 (0)11 267 1160
+94 (0)11 267 5212
info [at] lirneasia [dot] net
Copyright © 2026 LIRNEasia
a regional ICT policy and regulation think tank active across the Asia Pacific