excellent answer @laksh211in
-
Amazon Deals - ToS - Like us?
Why FTTH speeds are not growing in India
- Thread starter Kyle Crane
- Start date
- Replies 130
- Views 2,397
@laksh211in True, we don't need it. I mostly see people only using the base plan of their ISP even when there are much better plans available. So, why would ISPs bother with decreasing price or getting new plans which would require hardware upgrade too.
ISPs are getting more for their base plan, with less bandwidth use. That's an advantage for them. No way they would ruin it for few internet enthusiast people.
ISPs are getting more for their base plan, with less bandwidth use. That's an advantage for them. No way they would ruin it for few internet enthusiast people.
unlike USA and Europe the base of net users is India is focused only simple stable and cheap connectivity; they make up with sheer volume of users in millions. exception is a small 5%-10% power users who need high speed fiber similar to USA or Europe standards in ping and routing standards.
en.wikipedia.org
we are way ahead today compared to the dark 90's which i saw with 28 kbps and 56kbps dial up connections. today's broadband is a blessing I would say.
Internet in India - Wikipedia
we are way ahead today compared to the dark 90's which i saw with 28 kbps and 56kbps dial up connections. today's broadband is a blessing I would say.
it is limited use case but expecting that speeds pan India will take decades i feel as entire back-end network has to be upgraded to 10G.
i think most people aren't getting the point that right now bandwidth is the cheapest as it has ever been, so offering those 30 mbps plans are like throwing a bone . They can easily set the base higher but they are milking their customers at this point.
in our country where many people buy provisions for 3500-5000 rs a month do you think they will pay for 1 gbps internet instead? 30-40 mbps is for normal users and generally 100 mbps as mostly people do email, browsing, and YouTube and streaming 90% of the time. i am not talking of ultra rich though.
Last edited:
Interesting discussion. I am not going into the part whether higher speeds are needed or not.
Mr Parag Kar is largely correct when he points to ARPU. Almost everyone (from Airtel to Jio and even smaller players) are using GPON. Shared PON branch is 2.5Gbps down/ 1.2Gbps up. Thus selling 1Gbps as base plan may cause issues. It won't cause issues if people don't actually use it but even with two users can cause issues for 64-128 users. For now networks have kept 1Gbps as premium option. If someone is willing to pay 5-10x of what others are paying networks can workaround this limit by putting these users in seperate PON branch / litting more fibre / putting less users on the shared PON port etc. But in absence of larger uptake, all this effort is not worth it. XGS PON is the right technical solution but ONU cost is typically $100 outside of India (unsure of Indian pricing), OLT also has a premium.
Some numbers:
Airtel on the lowest plan (which actually sells a lot) is 3000 INR for 6 months for 40Mbps.
1. This includes 18% GST and 8% AGR, so remove 26% from it (left with 2220 INR)
2. CPE itself is close to 2000-2200 INR (left with zero)
3. In LCO areas (which there are many), LCO would take 30-40% cut.
4. Drop cable even with say 50m is probably 4 x 50m = 200 INR.
5. Three patch cords - one at pole, two at the home say 150 INR
I am not even bringing the opex cost of the infra deployed to the street, cost of bringing bandwidth to the pole etc.
Thus it's all negative for now on lowest tier and likely break even is well after 6 months. For some players cost can be lower but largely these numbers do not make sense. Hence you see the annoying push for FWA / airfibre product.
Likely we will see better speeds when:
1. Lower cost of rolling out infra - last mile roll out is a big deal. It doesn't logically make sense for everyone to be rolling out there own fibre. Ultimately we will get to situation where it will be common set of players rolling out fibre (say 1-2) and everyone uses them. This can be overhead fibre and with a plan to go underground someday. Realise that when you have 10-12 ISPs in an area, there are more than 12 fibres hanging on poles and cost has to be recovered somehow.
2. Cheap backhaul / NLD capacity - While cost of NLD is not a deal breaker for Airtel/Jio but it's there for smaller players. If it's not a deciding factor (like case of American market with many neutral dark fibre players), small players can give faster speeds and that would push competition and larger guys will also upgrade the speed.
3. Demand - for many people lowest tier just works. BSNL and Airtel were able to sell 256Kbps ADSL and stuck with 2-4Mbps for a very long time simply because there wasn't a very high demand for higher speeds. Ignore GPON limits for a min and imagine Airtel/Jio sell 1Gbps for 699 INR today. What will happen? Will more people buy their connection? How many people (who can buy today) are not buying because they must get 1Gbps at that price? In reality not a lot. But a small set of premium users (content players, rich folks) who are paying 3k for 1Gbps - what will happen to them? Can they be given 10Gbps for that price? Technically not. Hence this will kill higher tier without adding any fresh users on the lower tier. This market dynamic has to change before major speed upgrade happens.
Mr Parag Kar is largely correct when he points to ARPU. Almost everyone (from Airtel to Jio and even smaller players) are using GPON. Shared PON branch is 2.5Gbps down/ 1.2Gbps up. Thus selling 1Gbps as base plan may cause issues. It won't cause issues if people don't actually use it but even with two users can cause issues for 64-128 users. For now networks have kept 1Gbps as premium option. If someone is willing to pay 5-10x of what others are paying networks can workaround this limit by putting these users in seperate PON branch / litting more fibre / putting less users on the shared PON port etc. But in absence of larger uptake, all this effort is not worth it. XGS PON is the right technical solution but ONU cost is typically $100 outside of India (unsure of Indian pricing), OLT also has a premium.
Some numbers:
Airtel on the lowest plan (which actually sells a lot) is 3000 INR for 6 months for 40Mbps.
1. This includes 18% GST and 8% AGR, so remove 26% from it (left with 2220 INR)
2. CPE itself is close to 2000-2200 INR (left with zero)
3. In LCO areas (which there are many), LCO would take 30-40% cut.
4. Drop cable even with say 50m is probably 4 x 50m = 200 INR.
5. Three patch cords - one at pole, two at the home say 150 INR
I am not even bringing the opex cost of the infra deployed to the street, cost of bringing bandwidth to the pole etc.
Thus it's all negative for now on lowest tier and likely break even is well after 6 months. For some players cost can be lower but largely these numbers do not make sense. Hence you see the annoying push for FWA / airfibre product.
Likely we will see better speeds when:
1. Lower cost of rolling out infra - last mile roll out is a big deal. It doesn't logically make sense for everyone to be rolling out there own fibre. Ultimately we will get to situation where it will be common set of players rolling out fibre (say 1-2) and everyone uses them. This can be overhead fibre and with a plan to go underground someday. Realise that when you have 10-12 ISPs in an area, there are more than 12 fibres hanging on poles and cost has to be recovered somehow.
2. Cheap backhaul / NLD capacity - While cost of NLD is not a deal breaker for Airtel/Jio but it's there for smaller players. If it's not a deciding factor (like case of American market with many neutral dark fibre players), small players can give faster speeds and that would push competition and larger guys will also upgrade the speed.
3. Demand - for many people lowest tier just works. BSNL and Airtel were able to sell 256Kbps ADSL and stuck with 2-4Mbps for a very long time simply because there wasn't a very high demand for higher speeds. Ignore GPON limits for a min and imagine Airtel/Jio sell 1Gbps for 699 INR today. What will happen? Will more people buy their connection? How many people (who can buy today) are not buying because they must get 1Gbps at that price? In reality not a lot. But a small set of premium users (content players, rich folks) who are paying 3k for 1Gbps - what will happen to them? Can they be given 10Gbps for that price? Technically not. Hence this will kill higher tier without adding any fresh users on the lower tier. This market dynamic has to change before major speed upgrade happens.
@securetty Telecom sector is over-regulated. Any minimum speed enforcement of levels say 1Gbps minimum would be a bad idea. I would argue speeds are rather slow because of over-regulation. Your example of markets like France, Poland usually have termination monopoly. I agree with you that large Indian telcos don't have incentive but at the same time they are there for profit. Nothing wrong with that. Unless increasing speeds generates more revenue/improves profit why would someone do that. We all like to follow same logic in our personal life but somehow expect telcos to simply leave profit & loss statement out. If regulation is reduced you will see more and more infra build out and overall lower per bit delivery cost.
@Kyle Crane
(Ignoring your pessimist tone about India and keeping them outside of this reply for now)
While even I get impacted with FUP when choosing connection and have to go out of way to avoid FUP, but let's discuss it. As of now average monthly data usage is between 300-500GB on FTTH (lower in rural areas, higher in urban areas). Current FUP is usually 10x of that value. While I would love to have no FUP plans, challenge with that is - someone who needs it right now is actually using over 10x of average user. All that is perfactly fine in developed markets where ARPU is very high. A few users use 10-20x of average, sure doesn't matter. But in India when ARPU is so low, most of ISPs tell me that they actually want such users to churn.
Another part of problem is Hotels/small offices. ISPs want them to buy leased line (many them already do). FUP is largely the differential if you compare a leasedline of say 20-30Mbps for 50k an year Vs 100Mbps for 9k an year. Again, not ideal scenario but if FUP is removed, there would be a churn on leasedline and loss of revenue.
Get average user to pay 5x more - build a business model around it, I can bet you live without FUP and possibly with much higher speeds. It's just hard reality of the market.
Funny story on PMWANI:
Once an ISP shared how he has stop selling completely for PMWANI backhual. He found a few connections sold were used to backhual fibre OLTs in nearby buildings. Someone got 3-4 x 100Mbps connections and starting pushing a dozen user traffic. All that is illegal on paper but it is what it is. While just like you, i would love to have say a 1Gbps connection for say 600 INR a month with no FUP, I can bet you will find many smaller ISPs re-selling them and person selling it will simply loose on the revenue.
Similar threads
