YC W17 Launch: Wifi.com.ng – Uncapped ISP for Africa | Hacker News Reader

YC W17 Launch: Wifi.com.ng – Uncapped ISP for Africa

blog.ycombinator.com

143 pointscraigcannon75 comments

How about we replace "Africa" with "Select African Countries". This is as odd as a company saying "Uncapped ISP for Europe"; it's an entire continent!

It's only Nigeria actually.

It's only the Lagos region actually.

given that this is a launch announcement, it would seem to imply that they've started in Lagos (probably due to its relatively high density of internet users) and will expand out into other regions in Nigeria and other African nations as they grow.

Most people don't realize how big of a land mass Africa is. US fits in Africa 4 times over. It's the second largest continent after Asia.

Taking a look at different map projections when I was young was the first time I realized that visualizations can be deceptive.

the mercator projection has a lot to answer for.

Yes, the tendency to label Africa as if it were a country (rather than 54 of them) is highly annoying. Clearly this won't be available for me (Zimbabwe/South Africa)

Not to mention that most ISPs in SA offer uncapped (if shaped) internet.

Oh My. Never knew something like this existed, Similar to Wifi Dabba which was here a few days ago. As a Nigerian, this will definitely make life a whole lot easier but a particular Achilles heel for startups in your line of business is pricing. What do you charge? and how do you intend to convince Nigerians that you are better than the likes of Spectranet, Swift, mainone, telcos like Glo, Mtn etc

Looks like the monthly plan is ₦1500, or $30 USD.

Even at official rate, $30 is ₦9307.50. At actual rate, $30 is ₦12,000

Where did you get that from? Not Google or XE, obviously.

abokifx.com

We charge $100 for setup and $30 monthly subscription. WRT convincing Nigerians, our product sells itself as we offer unlimited internet with no caps.

To whom does your product sell itself? ie, for the average (including less central/wealthy cities) Nigerian, this is about 10% of income.

Is Lagos just that much wealthier than the rest of the country? Is there a wealthy professional subset that can afford this? Or is it mostly business users who need lots of bandwidth?

Can customers share their connection by using n devices e.g. a router and share it with their neighbours?

The title here is misleading (I am from Uganda, was excited ), this service seems to be aimed at Nigeria (only Lagos at that) , see the .ng .

Could the OP or Mods please change this to a less catchy and more accurate title like `Uncapped ISP for Lagos` .

You guys should try to get NGO's as customers to provide them connectivity at Western rates and use the surplus income to subsidize the local customers.

This is something we could explore.

IITA would be all over that! Their internet is really bad.

I'd avoid NGOs, if you look over at DSLReports you'll see NGOs complaining about the free connectivity they get, and their ISPs chiming in about how its costing them a couple grand a month to supply a few Mbps to them since they slam the connection 24/7.

Costs are high and anything less than perfect, low latency connectivity supporting 20+ HD Skype calls is unacceptable results in NGOs bitching about the service they get for free.

They are probably "bitching" because they are struggling against titanic obstacles when many of them could have cushy desk jobs in Indianapolis. If an ISP wants to help them, it would be to participate in their mission, not just to avoid "complaining."

A common theme for expats: if you're there only for the money, you're gonna have a bad time.

From what I hear at work at least some NGOs create lots of problems in other ways as well.

TBH, their pricing is already Western rates (at least UK).

9500NGN = £25/month, which would get you 40/10 VDSL2 uncapped in the UK.

Ask if they are a charity and charge them more if they are?

I couldn't see any info about the founders or their company on the website, but it looks like the company is Tizeti Network Services, founded by Kendall Ananyi and Ifeanyi Okonkwo: https://www.tizeti.com/

The company website is https://www.tizeti.com and the customer-facing one is http://www.wifi.com.ng.

Very cool! Many of us have dreamed of doing this in the USA. Seeing it done in Africa where it meets an actual neeed rather than just a hacker fantasy is particular cool!

How are the nodes connected? Is it a mesh?

How about the backhaul? How's that handled and is it still under the control of the telcos?

It's not a mesh. We do a point to point between towers and then a point to multi point to customers. Mesh halves throughput, so we stayed away from it.

Single radio meshes halve throughput, but a multi-radio mesh does not have that same hop speed halving effect.

Or you could bond the two radio channels and have twice the throughput, which is effectively the same thing as halving the throughput vs equipment and spectrum costs.

Would love to know any details about the last point to multi-point part segment.

Do customers need to install wifi receivers on their roof tops ?

Can you go inside flats ?

Do you need line of sight to the towers from homes ?

How far can a house be from the tower ?

> One of the top 3 infrastructure challenges in Africa is internet connectivity (the other 2 are electricity and healthcare).

What about roads and water? "According to a report for the World Bank, average road density on the continent is 204 kilometres of road per 1000 square kilometres of land area – only a quarter of which is paved. In contrast, the world average is 944 kilometres per 1000 square kilometres with more than half paved." [1]

Not that I expect a YC company to solve African roads...

[1] https://www.newscientist.com/article/mg22129512-800-africas-...

With enough electricity, healthcare and internet there isn't nearly as much need for roads and water.

I'm sorry, but need for basic infrastructure like WATER does not disappear when you receive internet and electricity.

enough electricity lets you desalinate, DUH

desalination requires infrastructure, and most of the water needs are in places like the sahel which are super far from the sea, the problem isn't desalination, it's that there's no water in the ground in some geographic locations

As someone who spent a lot of time delivering fertilizer and seed in rural Kenya, you're really wrong about this Sam.

Even the best regional roads in Africa are two lane roads. Unlike the US, cars have highly variable top speeds though, which means nearly constant passing. Cars will drive in the wrong lane for significant periods of time and hastily merge back. But there are large number of buses that are heavily incentive to go quickly at any cost, so they will often off road and suddenly pass on the left. I've seen a two lane road with 5 lanes of traffic on it.

That's the best case scenario. A large chunk of the country is only accessible via rough road spurs. If you're trying to deliver, say hypothetically, 10 tons of fertilizer and seed on the back of a truck, it means you're going 10mph instead of 45mph.

My family worries about me getting murdered in Kenya or the government suddenly melting down, but really traffic is what they should be worried about. And this is Kenya, which is one of the most developed countries in E. Africa. There's a real need for some more roads.

Holy shit. 30% coverage for a city the size of Lagos is already an impressive success.

Thanks. We intend to cover the entire city of Lagos.

How do you intend to get roof rights to do so? I'd be adverse to using roofs without a roof access agreement, otherwise whenever a customer moves, your apt to lose a node & its coverage/the sites it relays bandwidth for.

I know Lagos has wifi enabled kiosks on the street, for downloading movies from a Netflix style subscription service, so I suspect they are doing something similar to that, instead of rooftop access.

Is Lagos flat?

Relatively flat

What's the backhaul from one of these base stations? And what sort of real world speeds are people getting?

Right now we are doing 300Mbps on the backhaul. People get 7.5Mbps burstable to 15Mbps but the equipment can support up to 75Mbps per user.

The 300Mbps connection doesn't come from the local telcos or their subsidiaries?

What kind of backhaul? Fibre? Or point-to-point wireless?

Pricing seems to be about 48 USD for setup and 31 USD per month for the service [1]. Is that correct?

Edit: The currency, the naira, lost quite a bit of its value last June [2] when the central bank allowed for it to become (more) market-driven [3].

[1] http://wifi.com.ng/pricing/

[2] http://www.xe.com/currencycharts/?from=NGN&to=USD&view=1Y

[3] https://www.bloomberg.com/news/articles/2016-06-14/nigeria-c...

Is that cheap for the country, though?

At the moment we're going after the middle to upper class and would roll out plans in a few weeks to meet the lower end of our spectrum.

Are you planning a prepaid hotspot model, or individual connections? I'd think the cost per Mbps from your upstream and the overall cost to backhaul each Mbps across your network would be the limiting factor for lowering your price point further.

Setup is about $100 and monthly subscription from $30.

According to your pricing page, there's an offer for 50 USD instead of 100 USD of setup costs, though.

Sounds quite expensive considering the average monthly household income in Nigeria is less than 280 USD.

$280 is outdated. And too optimistic.

Right now at N400 per dollar (it was 450 barely a month ago), $280 is 112,000 naira. Top earning high school teachers get N50k monthly. Entry level bankers get N70k. I know an accountant for a media house (Masters in finance) who earns N80 and works 10 hours per day for six days a week.

Congrats! We're based in Uganda and dealing with Internet is a huge pain point for us - we have a backup provider because reliability is so poor and we're paying much more than $30 USD/month. All the best and hope you expand Eastward soon!

East Africa is on our road map. So we'd be seeing you

Can't wait! Whats the timeline look like? Prices out here are insane

Best of luck folks! Looks like a super cool problem to solve and work on!

Thanks! We have been at this at a while and hope to solve the connectivity issue in Africa.

What equipment/tech are you using for the point-multipoint segment?

Good luck! Although, if I'm honest I fail to see the "YC-grade" innovation in this project versus the myriad of WISPs that have been so popular for more than a decade. Does it all boils down to the unmetered business model?

What do you mean, uncapped?

So the user can consume any amount of data from the global internet?

How do you manage that?

This is the most important question. Answering this on their site / sales materials will boost conversion than any other thing I can think of.

The next question is, "What areas of Lagos have coverage?"

Their site doesn't say. Only gives pictures of buildings of their customers.

Uh... I hate it very much when companies claim "uncapped" data. While they don't artificially limit how much data you can use, you are always limited by your speed.

Yes, your data is unlimited but but your speed of 128k limit how much you can actually use. In this case, after a short chat with their online reps they offer 2-5 mbits. At 5 mbits this translates to roughly 1.6 Terabytes of data over 30 day period.

Don't get me wrong, what they are doing seems great! I just wish they would show their speeds upfront and the download/upload ratio.

This is really cool. Is the tech similar to MonkeyBrains? MonkeyBrains charges $35/month and pays their technicians Bay Area wages. I'm sure there are good reasons why you can't drive the price down lower than $30, but they are not obvious. It it just to manage demand as you get off the ground? Is the upstream fee extremely expensive? Something else?

Upstreams are still lower than what they would get on enterprize on our competitors network. We can actually drive our cost below $30 however most important for us our goal is to build out our network that's essentially what our raise is for

Why is this trending? It is nothing new. This has been ongoing in many african countries for years

I fail to see a competitive edge for this company? If providing uncapped internet is profitable at those rates, can't the existing telecom providers just start offering uncapped plans? That renders the whole point of this company moot.

Not really - it gives you a chance to grab market share. Even if the competition catches up, you can have already gotten a lot of loyal customers.

This is really risky because it takes so little for competition to catch up. They just offer another pricing tier in their billing system that shuts off caps and they are done.

In kenya a company called zuku started providing uncapped home internet at about the same prices. Now its well on its way to owning the market for Internet at home, despite all the efforts of local telcos to compete.

At the same price the biggest telco is only willing to give about 12gb capped data. And it's no where near the quality of a zuku connection.

I think it might have to do with the physical infrastructure. A cell tower can be seen as an access point shared by millions and therefore has to have limits on what each device can send and receive. Home internet is via fibre and an an access point shared by only memebers your household.

How would you compare your service to Express Wifi (or similar), is it competitive or does it serve a different market segment? I'm also curious how you manage forex risk and the regulatory system there.

Reply on news.ycombinator.com