Much of the published statistics are elaborate shams designed more to impress than convey intelligent data. As jakeyen has pointed out, Bangldesh, Pakistan and India are a case in point.
There are just three congregations in Bangldesh all made up of former Christians from different denominations, usually Roman Catholic, and all in the capital city Dhaka. There has been no penetration whatever into the vaster, rural Muslim population. For them the Watchtower is as mysterious as the man in the moon.
Pakistan is just the same, with 18 congregations, again all urban, Christianized and comfortable middle-class.
India is probably slightly better but again with a high preponderance of former Christians in the Watchtower fellowship. When I was there in the early 70s there was not a single Watchtower follower who was not from a non-Christian background. [Again mostly Roman Catholic]. There was a former Sikh, back in the 60s, who became a Watchtower follower, but this was only because he fell in love with and subsequently married a missionary girl.
Even in those relatively well off countries where the Watchtower operates, one could question the effectiveness of their evangelistic work. I challenge any Watchtower follower to line up a sampling of the general public on any street corner in London, Montreal, Sydney, Tokyo, and even in the Watchtower's own backyard, New York, and ask a simple question:
What do the words "Jehovah's Witnesses" mean to you?
The answers you will get will range from:
Are they a new rock group? Aren't they the ones who discovered golden plates or something? Are they the guys who kill their babies by prohibiting blood transfusions? and so on. Anyone who says "They are the only ones preaching God's kingdom"! can be shown to be a Watchtower follower anyway, which is statiscally possible, given a random sampling.
It is truly sad that after 131 years of frantic effort, this is all they have to show for it.