Access East Africa

Kai Wulff, CEO of Altech Stream East Africa, explains why fixed infrastructure has a bright future in Africa

Tags: KenyaKenya Data Networks (
  • E-Mail
Access East Africa Kai Wulff is optimistic about the fixed telecom sector in East Africa.
By  George Bevir Published  June 9, 2010

There is only one way to drive internet penetration in Africa above 10%, according to the boss of one of East Africa's key telcos. Kai Wulff, CEO of Altech's East Africa-based divisions, says that the continent will only burst through the double digit barrier when Africans can see that an internet connection can result in a tangible benefit.

"It has nothing to do with the cost of the device or the initial investment, it depends on how relevant it is to the life of somebody," Wulff says. "It must be a direct benefit - if internet providers can demonstrate that we will exceed 10% penetration."

According to the latest figures from the International Telecommunication Union (ITU), at the end of last year internet penetration in Africa stood at just over 8%, considerably lower than the world average of 27%.

Wulff suggests that the actual figure may be even lower. "If you count somebody who goes twice a week to an internet café, then you are right to estimate 10%. But the reality, the proportion of real internet subscribers and people who use it for useful things, you might see at only 2% to 3% penetration."

In Kenya, there has been a steady increase in the amount of mobile subscribers, up 19% from 16.2 million in 2008 to 19.3 million in 2009, according to figures released by the Communications Commission of Kenya (CCK) in March.

While mobile penetration is rising steadily, internet usage has not enjoyed the same rate of growth, perhaps at the expense of the steadily increasing amount of mobiles latching on to networks throughout Africa.

There were almost two million internet subscribers in Kenya, with the vast majority mobile subscribers; 8,349 were fixed internet users, a base that declined on the previous year by 1%.

The cost of using a mobile in Kenya has declined significantly, which the CCK said was a result of stiff competition from the four mobile operators.

According to the ITU's price basket comparison, Kenya's mobile cellular sub-basket as a percentage of gross national income (GNI) per capita 2009 stood at 12%, down from 24% in 2008. Meanwhile, a basket of fixed broadband services stood at 62% of GNI per capita 2009, down from about 296%.

The reductions propelled Kenya to the top of the ITU's list of ten economies with the greatest decrease in ICT prices (see table below), ahead of nearby countries Madagascar and Zambia.

Undersea access

Wulff says the drop in prices can be attributed to one factor in particular: submarine cables. Before they landed on the coast of east Africa the great swathes of the continent were reliant on VSat terminals for access to the internet, a costly setup that kept the internet out of reach of many, and that still exists in some landlocked countries.

"We brought in two submarine cables, so the cost per megabit duplex dropped from $5,000 to a purchase price now of approximately $68," says Wulff.

KDN is the the anchor tenant for Seacom in Kenya with an initial purchase of 10Gb, and it also has a 10% stake in the Teams cable giving it access to 140Gbps. With ready access to international capacity and the option to increase it further when the need arises, Wulff says KDN has been able to reduce prices, describing KDN as "the guys who destroyed the market prices".

"Because we are an infrastructure company, I want the water to be as cheap as possible so more people buy pipes from me," he says. "That is why we dropped the prices from day one and we said we'd give the full benefit of a submarine price drop to the market, and people hate us for it, to be honest, including even our share holders.

"In terms of access, the cheapest service you can now buy from me is free, because it is advertising financed," Wulff says. "I think we are a little bit ahead of the rest of the world in terms of marketing and product range, because the cheapest product you can buy from me doesn't cost you anything.

That is pretty good, but it still does not give you more than 10% penetration, even if I give it for free. It is not an elastic demand and you still have to create relevance, even if you give it for free."

Digital village

Wulff returns to his theme of demonstrating a "tangible benefit" in order to drive up penetration. One of the programmes KDN is involved in is the government and private enterprise-led ‘Digital Village' initiative, which he gives as an example of the different way of working in Africa.

"I gave a speech to Bill Gates and a few other people where I put an empty Coke can on the desk, and I said them if you can tell me what that is I will stop my speech. Everyone said, ‘it is an empty coke can'. I said, ‘no, it is a raw material.' And we have to see the same in every business case in Kenya," he says.

KDN tries to demonstrate how the internet can be relevant to the lives of Kenyans by drawing up a business case to show people how they can make money from the web, packaged with a financing arrangement. An important aspect is that no individual controls the intellectual property of each of the business cases, so that others can replicate the business case.

The example Wulff gives is a village with some solar power and a receiving antenna with WiFi redistribution of the capacity, together with some kind of financing scheme to allow people to buy cheap end user devices.

"We started [Digital Villages scheme] three years ago," he says. "It is always hard to explain to your shareholders why you want to use a business case that only returns your money in three years...but we have done several and they are working well. We have one in a slum here, he had one when he started and now has 46. He has financed the 45 computers himself by the money he makes through the Digital Villages programme."

The programme ranges from equipping schools with several computers, to a solar cell on top and two laptops and a WiFi retransmitter. Most of the users access the network with refurbished laptops over a WiFi connection.

"It is the cheapest technology to reach them," Wulff explains. "There is no interference that far out, you are the only one broadcasting.

"With a CPE you can get about 4-5 km line of sight, if you have a nano station you will get six or seven kilometres. It's not WiMAX range where you get 30km but for the amount of money we can put a lot of WiFi transmitters out there."

Another advantage over WiMAX is that with the 4G technology users need a radio, you need a computer and you need power. "We are the biggest WiMAX operator in Africa, so we have 100s of base stations, so I have nothing against the business case. It is just a fact that you need things people can afford and can power the CPE at the other end."

Backhaul bonanza

"We bill about 7,000 blue chip corporate customers, but we don't bill Butterfly customers because they use a prepaid scratch card service and we deal with that at arm's length, so we would be sitting there at 60,000 subscribers maybe but it is hard to say precisely because some months they buy, and some months they don't buy," he adds.

Wulff says that KDN accounts for approximately 80-90% of Kenya's international internet capacity, carrying traffic for the country's big operators and as the link to Seacom.

"In terms of subscribers, if you count the ever-changing 3G subscribers and not their backhaul then we might have 10-15%, because 3G is these large numbers but not really large in capacity.

"Out of the 3G backhaul we do about 80-90%. Maybe the billing is with (mobile operator) Safaricom, but the transport of the network is with KDM."

Providing these links is what Wulff initially set out to do, but when he began installing fibre in Kenya, it prompted some people to question his sanity. "When I laid fibre here the first time people said I was crazy. Now I tell people that I want to sell a service where I actually pay people to take the service, with a business case, they call me even crazier but, I think it will make money."

KDN works with a host of operators across the region, including Etisalat, PCCW, Bharti Airtel, Zain and Qtel, providing access in and out of Kenya. It also provides hosting services in a data centre, back up for data for operators and fibre backhaul from Kenya.

"100% of Seacom goes through our network and roughly 60%-70% of the Teams traffic goes through our network. If you take people who are adding us and the redundancy you are close to 80%-90% and with Teams and Seacom you would have 95% market share because the rest is on Vsat."

On its website, KDN claims to backhaul 75% of Kenya's mobile traffic, a figure that Wulff describes as pure "marketing".  "I would say it is more, but it doesn't really matter," he says.

It is not the amount of traffic that passes through KDN's pipes that is of interest, Wulff says, but the market share in terms of infrastructure that has been deployed and access to customers.

"We want to maintain a 50+1% market share wherever we go as an infrastructure provider, because then you are relevant, you are the bully in the market and you have the lowest operational cost because fibre is a game of size. As long as we keep investing the same amount of money as all of our competitors combined I think we will continue to be fine. It is all about divide and rule."

He says the business case is simple: roll out fibre cheaper than anybody else and we maintain it better and cheaper than anybody else. "There is no magic trick - if you roll out 1000 km of fibre, your project overheads are maybe 15%. If you roll out 10,000 km of fibre, then your project overheads are maybe 3-4%.

"If you maintain it -100km or 1000km - it is not a factor of ten, it is maybe a factor of two between 100km and 1000 km. To monitor a fibre network you need two fibre nodes anyway, irrespective of if you have 10km or 10,000km of fibre. The software to monitor the network and the incremental cost of adding more boxes is very little, compared to the millions of dollars you have to spend on some carrier-grade monitoring systems and umbrella management systems, so it is just a matter of size," he says.

"You are not really more efficient because you have found the magic bullet of how to do fibre. It is simply that you were bold enough to do fibre when nobody wanted to do it, and by the time the competition wanted to come in you have so many kilometres of fibre that your project team and everybody is 100% efficient and your overheads are low."

KDN estimates that it installs 10 km of fibre every day, a figure that Wulff says could actually be higher. When asked if KDN will spend more on capital investment than last year, he declines to comment, other than to say KDN's "vision is to maintain absolute market leadership in terms of deployed infrastructure and you have to keep it that way".

According to the firm's last set of results, capital expenditure was $20 million, part of which went towards the construction of a 1,500 km fibre optic cable from Mombasa to Kampala in Uganda to run the capacity bought later in the year from Seacom.
Pushing into neighbouring, landlocked countries will provide KDN with another opportunity.

The company has already moved into Rwanda, where it now has microwave, and Wulff says a fibre network is now 99% completed. "We have some issues with the government in Rwanda for the approval process, but Rwanda will be on submarine fibre this month [May] hopefully," he says.

Wulff adds that Tanzania is also connected by submarine fibre. KDN will also be connecting Burundi and Zambia after it has "switched on" Rwanda. Wulff adds that Malawi will also be connected by the end of the year and plans are underway to connect Mozambique, Southern Sudan and Somalia.

"If we can push our way from Kenya to SA and do back connections to all landing stations and at the same time run a similar business case in west Africa, encompassing Nigeria all way down to Angola, and cross connect back ends of fibre to landing stations, in the next three or four years, you have a fibre backbone that is not much worse than in Europe or US," he says.

3099 days ago

I agree with Thomas. I think KDN CEO and management team are onto something big. The law of "first mover" applies here. This CEO had a vision and guts to do the unthinkable: Believing in Africans. Where else in Africa do you see a CEO giving Raw material (Bandwidth) for almost nothing? Can't wait to read more about this topic. Thanks for this great post.

3196 days ago

I think the guy makes real sense .. We have sen what he has achieved so far and I am sure he will come up with even bigger surprises! The real topic should be how he manages his team ... This is the magic bullet he denies to have.

Add a Comment

Your display name This field is mandatory

Your e-mail address This field is mandatory (Your e-mail address won't be published)

Security code