Mobile Web Content in East Africa [Report]

Vodafone recently concluded a policy paper on “Broadband in Emerging Markets”, also titled “Making Broadband Accessible for All“.

The position and reason for this paper is best summarized below.

“The success story of mobiles in the developing world is well known. Yet in the case of extending data services in emerging markets, there is a real danger of some serious policy mistakes. As in developed markets, broadband strategies in developing countries have tended to focus on investment in fibre. This is too simplistic. This focus on fibre may miss an opportunity for a transformational change built on the capabilities and in particular accessibility of mobile broadband. The early evidence suggests that mobile internet is spreading as quickly, in some developing countries, as mobile telephony did originally.”

Traditional definitions of broadband have a narrow focus on bandwidth and speed. This paper uses a wider definition, as broadband policy needs to consider the entire ‘eco-system’ of internet and data services from both a demand and supply-side perspective.

Content Sections

  • Mobile Internet usage and demand in Kenya: The experience of early adopters (David Souter)
  • The potential of mobile web content in East Africa (Erik Hersman)
  • Spectrum policy and competition in mobile services (Thomas W. Hazlett)
  • Rethinking mobile regulation for the data age (Martin Cave & Windfred Mfuh)
  • Building next generation bradband networks in emerging markets (Luk van Hooft)

The Diffusion of the Mobile Web Across East Africa

Mobile web content is growing at an astounding rate. It rose 2.6-fold in 2010, nearly tripling for the third year in a row. Official Kenyan industry statistics show that mobile internet subscribers will grow by approximately 843% for the 12 months to September 2011.

What I like about papers like this is that I get to use words that normal people don’t use. I make a case for international content and platforms as “drivers of diffusion” of data across East Africa. That simply means that these platforms and content are helping to spread the use of data more deeply into the region, and allowing local players to get in at lower costs.

International web content is by far the most widely available and used in East Africa. This is in large part due to the ease of finding and disseminating this content, as well as its normalized licensing schemes and reliability. International platforms also carry a majority of the content that is currently being viewed on mobile phones. The following are the types of content that are most important to consumers in East Africa, according to our interviewees:

  1. International entertainment news (sports, gossip, lifestyle)
  2. Local news
  3. Breaking news
  4. Facebook (and to a lesser extent other social network tools such as Mig33, Mxit and Twitter)
  5. Jobs
  6. Dating (chat and relationships)
  7. Religion
  8. Local video/media

The reasons are that international platforms, such as Facebook, Yahoo!, BBC, CNN, Google and Wikipedia, have already been tailored to work on the most widely used data- enabled handsets. This contrasts with local content providers, many of whom have yet to tailor their websites for mobile access. In addition, local content less available at present, not as easy to license, and often cannot be reliably guaranteed as a long-term source.

Local Content

I interviewed a number of executives from Kenya, Uganda and Tanzania. There was a clear belief that while international content, increasingly localized for the market, is currently king, local content has the greatest growth potential because it is more highly valued by consumers.

While local content developers lack scale they have advantages that the global platforms do not. For one, they understand the local tastes and culture so customers value their content more. The consumer benefits of truly local content and platforms could be large.

The Government Role

There is still a lack of concrete government policies for government services or content to be made available or accessible via the mobile in any country in East Africa, even though this is the primary channel by which citizens could access services online. There is a solid case to be made for mGovernment, instead of just eGovernment.

To underline this, the most popular Kenyan Government website (Kenyan Revenue Authority) is shown as seen on a PC screen, a smartphone (HTC Desire) and a typical 2G internet enabled handset (Vodafone 350). The website is most clear and easily accessible via a PC interface (and consumer interaction primarily is through downloadable pdf files). There are no browsing problems when accessing through a PC-based browser. The KRA website is also accessible via the native Android browser in the HTC Desire Smartphone. The HTC Desire also allows downloading and viewing of pdf files. However, the native browser on the Vodafone 350 (a basic 2G EDGE handset) does not present the KRA website in a usable format. As can be seen, the website is badly rendered and quite impossible to navigate.

Possible government services to be made available via mobile web:

  • Paying bills
  • Service delivery questions and concerns
  • Taxes – access, information and filing
  • Health – access or appointments, information
  • Public job search

An argument can be made that m-government services would have a greater impact if the focus were on supplying tools for small businesses to interact with government, rather than only making services available for citizens in general. By removing the barriers to entry for small businesses, the government would be providing a service that increased usage, decreased business costs and had a potential tax revenue increasing effect due to filing and paying on time.

Summary

East Africans are accessing the web primarily through their mobile phones. The new medium is enticing them online with the new services and content provided through a new medium. Broadband penetration rates are low enough in this region that we are not yet seeing the displacement of newspapers, radio and TV seen in other, more connected regions of the world. However, as with all network technologies, there is the potential for reaching a tipping point. This will depend on the provision of enough mobile web content that is valued by East African consumers.

The content driving East African users online is currently largely provided by international news and content sources, such as Yahoo! and the BBC, and also by global internet platforms, such as Facebook and Google’s Gmail. Even taking into account the decreasing data costs, falling data-enabled handset costs, and the increased availability of broadband, there would not be enough traction locally to get to the critical point if the content were not available.

These international content sources and global web platforms generate demand, and therefore allow the mobile network operators to decrease costs as more users come online. International content is thus providing a pathway for local content creators. While local content is in high demand and there is a rapidly increasing user base, the mobile web content space in East Africa is in its early stages, and there are no
clear leading content providers. At present the key trend is the provision of increasingly localized content by the leading global companies.

This paper has identified two important barriers to the further diffusion of mobile internet usage across East Africa: lack of m-government policies; and, more important, an absence of charging mechanisms which share the cost of mobile internet access between end-users and content providers. If governments embraced mobile-based provision of services and provided access free of usage charges to end-users (sharing the efficiency gains through payments to network operators), the potential impact on internet access could be dramatic. The challenge for governments and local developers of mobile web content is to utilize their local cultural understanding and ability to maneuver quickly to make their content more relevant and affordable to end-users.

(Note: This is summary of my section. Download the full 2Mb PDF report to read the section in its entirety, and to read the other 4 sections of the paper.)

Pay Attention to the Mobile Web

In 2008 we saw the scales begin to tip with imports of data enabled phones being larger than that of non-data enabled phones.
In 2009 we saw the undersea cables hit East and Southern Africa in a big way.
In 2010 we saw the mobile operators get serious about data availability and cost packaging for everyday Africans.

2011 is upon us, and with it brings a new type of data-enabled mobile user in Africa. It also brings the mobile web to center stage.

Mobile web content has been defined as any internet-connected or browser-based access to the internet and as digital content connected to a database that passes through a handheld device connected to a wireless network.

Simply put, the mobile web is the same data that the web layer brings to you on a computer, just now on your phone.

The mobile phone is the most ubiquitous instrument there is in the market. Usage is no longer limited to sending and receiving calls and texts, especially with the increase of data enabled phones, increased bandwidth availability and decreasing data costs. The convenience in terms of use-anywhere-anytime has made access to mobile web content easier, accelerated by dropping rates of mobile handsets and data.

What does it look like?

Here are a couple of examples:

  • Consumer content such as movie times and restaurant reviews, such as Flix and EatOut.
  • Consumer focused transaction sites and classifieds like Dealfish and Pigia.me.
  • Content, such as news, blogs and aggregators like Afrigator.
  • Business information for consumers and businesses, such as Mocality.
  • Mobile-specific communities, such as Motribe, Facebook and Twitter.
  • The ability to pay via mobile payment methods or credit cards, brought to you by mobile payment aggregators like PesaPal.
  • Advertising done by the likes of InMobi and AdMob.

You can see that it doesn’t look all that different from it’s purely web-based counterparts. It’s the same data, just more accessible on your phone.

There are strong plays to be made in all of these fields, as there are few leaders in any country, much yet regionally… yet. The reason for that is we’re just on the front end of this sea change, so even the leaders only have a very small slice of the pie.

While there will always be a place for client-focused mobile applications (Android, iPhone, Ovi, etc.), there is just too much friction there to scale. Friction for the developers who build the applications, and friction for the users who need the “right” phone to access the apps.

For more brain food on this topic, I suggest reading Fred Wilson’s post, Counternotions and alternate thoughts from Diogenex.

Motribe: The Mobile Web Community Builder

The Mobile Web is the future of mobile apps, and it’s not surprising to see Vincent Maher and Nic Haralambous, from South Africa, on the front end of it. Motribe is a simple community building platform for the mobile web. You can easily get a site up and going in an hour that allows chat, photo sharing, private messaging and mobile blogs.

That bit about the mobile web is important, since it means you can browse to it on most phones, and you don’t need a special app for it built on all the smartphone platforms, like iPhone, Android, Ovi, WinMo and Bada – as in, there’s one less barrier to entry.

I asked Vincent why he chose mobile web, his response:

“Mobile is the killer internet platform for Africa, but also the rest of the world. We have found that our younger users prefer using an ipod touch to surf the web than a PC. Motribe works on 4000 devices (or more) and the Motribe plan is to change the way people use social networks in emerging markets.”

Initial funding was raised 4DI Capital, and they’ve got a clear business strategy, which is to sell their product. Pricing ranges from $10 to $50, and each level gives you a greater ability to customize and “own” the mobile social network that you’ve built. There is also an enterprise level available for bigger brands and companies. Motribe also has a free plan with core features and a 100-user limit for you to get started quickly.

Its built on Amazon EC2, S3, RDS and Cloudfront using PHP, Codeigniter, Google Charts, JQuery and Cassandra. Vincent stated that, “Cassandra is the most interesting of the components because its going to be the key to scaling to millions of users.”

Giving it a Test Run

I went ahead and signed up to give Motribe a whirl. My test site is AfriGadget.Motribe.mobi, where I’ll put up some stuff from AfriGadget and see if a community grows around it. Just getting going, I can see that a lot of attention has been put behind this platform (as would be expected with veterans like Vincent and Nic).

Some notes:

  • Signup: done easily, nice little touch to provide a QR code directing to a URL for login.
  • Setting up a community: simple, see image below.
  • Access code: for when you want only certain people to join
  • Test mode: for making sure your community is setup right before it goes live
  • Themes: many simplified stock themes available out of the box.
  • QR code generator: there’s a neat QR code generated for the URL of your new site. (Would be nice to have this as an embed code for websites)

There are a couple example sites already going – emofwendz.com is the one they ran for the pilot, and it has some fantastic engagement stats, like an average of over 100 pages viewed per visit (the norm for web sites is about 5) and average visit lengths of around 60 minutes. Today, Vincent said, an Afrikaans-language site was created for Christians http://ekerk.motribe.mobi, its a good example of exactly what they people to do with the platform.

Some Thoughts

If there’s any platform that’s come out of Africa in the last year that fills a global need, it’s Motribe. I won’t be surprised to see this go big at all.

There are always teething pains, experimentation and adjustments when a new platform goes live. I found a few issues, like when I went to upload my logos they threw a bug (I was a pixel off on the size, thus the issue). Not unexpected in a brand new platform, and I’m sure it’ll be fixed shortly.

I wasn’t able to test out the “Custom URL” and “Advertising Manager” features, though I would like to see how each is implemented. It might be worth having a section on the website to preview at least the Advertising Manager in more detail to see if it’s worth upgrading to.

There isn’t any SMS functionality yet, and I’m not sure there needs to be either. As Vincent said, “we don’t have a need for SMS right now but we may well integrate SMS at a later stage depending on whether we can find some good uses for it.”

Worth reading: other posts by TechCentral and the Daily Maverick.

Mobile Web East Africa – Stream

[I probably won’t be able to keep this up all day, but I’ll try to blog/stream what’s happening here at Mobile Web East Africa as best I can. Refresh for updates]

It’s day one of the Mobile Web East Africa conference in Nairobi. This is a new conference, started up in South Africa to great success, and now spreading to other regions of the continent.

Paul Kukubo of the ICT Board

Paul Kukubo, of the Kenya ICT Board, is talking about the future of tech in Kenya, and how the government’s aim is to be a major hub for technology in the region. Explaining how the changes in the industry are brought into context for the government’s vision 2030. He talks about mobile payments, digitizing of government documents and processes, developing software standards and the growing tech community within Nairobi.

Paul continues with mentioning how their approach is to influence policy formulation, intellectual property, data protection, linkages to venture capital and basically catalyzing growth in the ICT sector between government, the public and business.

Rick Joubert of Yonder Media

Rick Joubert, from Yonder Media, “the mobile phone is the most ubiquitous consumer device in the world.” He goes on to talk about how the phone is now even more spread through South Africa than radios. There are 2x as many phones as TV sets. There are 6x more mobile phone subscribers than internet users (in South Africa).

Rick defines the Mobile Web this way:

  • Tier 1: The WAP internet
  • Tier 2: The mobile web application internet
  • Tier 3: Web browsing on phone

**Interruption**
PS Ndemo, who I know and like, is going to give a short address. This isn’t cool, as he’s interrupting Rick Joubert mid-talk (and a very interesting one too). Case-in-point for why government needs to get out of the way more than anything else in the technology field… [Yes, I note that this is probably the American viewpoint on equality coming out].

PS Ndemo is talking through how there were 3.5 million internet users in Kenya last year. Now, with the cost of laptops dropping, we now see 500 laptops sold per day (there were only 20k per year sold before).

Kenya also has the digital villages project (Pasha) with the World Bank is seeing long lines of individuals in far off places coming in to try the internet, get on Skype and figure out how to set up email and other services.

PS Ndemo, ever the gracious person, has at least apologized for the interruption and made amends to the speaker and the conference as a whole. There’s a reason I like him… :)

**Back to our scheduled program**

“The Apple iPhone is the number one handset on every continent in the world, except… Africa”. The Nokia 3110 and Samsung E250 are the two biggest phones on the continent. The fact of the matter is, the real money is being made in the inexpensive DVD/Nollywood areas, not in the mobile web yet. Services that play to USSD, Voice and SMS are where the real opportunities lie.

Driving forces:

  • Growth in data networks and coverage
  • Mobile data access charges
  • Local content
  • Better quality handsets shipping at the cheapest possible price
  • Mobile wallets, mobile commerce, mobile banking

Business models and monetization routes:

  • Commerce
  • Transactions and financial institutions
  • Content
  • Advertising

Rick projects that the “size of the prize” in mobile advertising is approximately $8 billion per year in Africa.

Rick finishes showing a video from LynxEffect on how consumers see mobile web, the seductive side of it.

Eric Cantor of AppLab Uganda & Grameen Foundation

I wrote about AppLab and their work with MTN and Google last year. Eric wants to talk about a critical look at a critical space, the 95% of the African population that doesn’t own a smartphone.

(Get the full presentation of Eric Cantor’s slides as a PDF)

“There are more people having conferences and running too many pilots around the use of social mobile work than there is real practical applications and scaling of the products in the market.”

Technology: Be Patient
SMS is not the only way. It’s very challenging and very expensive to work with SMS. One way to adjust this focus is into voice – people like people, and want to talk with each other.

Handsets need to evolve. Nokia 1100 vs Java 1680 ($20 vs $60) – we’re waiting on the $40 smartphone.

Eric reminds us that we need to get back to the Four-Ps of marketing. We can’t forget user experience, the services might be serious, but still need to be fun to use. At the AppLab they don’t believe what they hear (because everyone says “yes, this product will be great in our market”). They try to dig deeper, learned from Google, on what customers really want and see what people are really using.

Question time

I’ve asked the question for Eric Cantor about why we’re not seeing very simple data hooks built into some of the USSD and SMS applications running in Uganda. (more on this here: “Should we be building SMS or internet services for Africa?“). Eric agrees that there is a lot of upside in that space, and that they’re trying to push more towards the data channel, but until we start seeing more data-enabled handsets in ordinary people’s hands out in the villages, it’s just not a main priority yet.

Robert Alai asks what is driving advertising growth in South Africa? Is it the large companies, or smaller organizations?
Rick responds saying that it’s large companies trying to reach their customers, from banks to Coca Cola and everyone in between. Businesses build grow in this space to find solutions for that, and that’s primarily small innovative companies (like his own), small nimble startups.

Agatha Gikonda of Nokia East Africa talks about the Ovi Store and the opportunities for local developers to create applications and put them on the store to make money.

Peter Arina of Safaricom

What are we going to do to drive interent usage in Kenya? Some stats:

Mobile users are estitmated at 19.05m subscribers. Kenya population estimated at 40m with 22m being the addressable market (15yo or older).

70% of mobile data users spend less than 20ksh on a monthly basis. Industry data enabled handsets estimated at 5m or 26% of GSM users. Cost of a 3G handset is 3x higher than that of a non-data enabled handset. Computer prices are way too high compared to the disposable income of majority of Kenyans.

“The cost of devices that access the internet is the biggest barrier to entry for ordinary Kenyans.”

Cost of broadband (price) is prohibitive due to infrastructure investment.

Local content – the most popular sites accessed on the Safaricom network is Facebook and YouTube. Limited content which is highly priced, is also a barrier locally. There’s a need for high quality data enabled handsets to get good experience.

Conclusion
Mobile data users estimated to reach 10m in the next five years subject to availability of affordable devices, increase awareness, local content development and drop in data prices. Safaricom is trying to work directly with the manufacturers to get more data-enabled devices into normal Kenyan’s hands.

There is a need for more local content that is relevant at affordable rates. Need for reduction in frequency costs, a creation of daily usage habits among users and a need for the government to remove VAT on modems.

Questions
@Kahenya is asking a question. Safaricom is trying to become more affordable, it’s still the most expensive network in Kenya. It’s still has no fixed rate for the mobile data network access. It doesn’t work for small and medium sized business, is Safaricom doing anything to change this?

Peter Arina says they are trying to be cautious. They’re trying to focus on quality (bull$%@& as they have the worst network connectivity in Kenya). He says that they have plans to reduce the cost of data as well, but he has no details on it.

The Safaricom rep says that their main goal is to provide services to the Wananchi (the ordinary/mass Kenyan). The question remains then, why is the cost so high for all of their services?

Paul Kukubo asks when Safaricom will open up their network for value added services for developers and other companies. He’s wondering why the revenue share is so high here (currently if you partner with Safaricom, they’ll take about 60% of revenues), meanwhile elsewhere in the world, like Japan, give 70% to the developer.

Paul asks about the issue with the networks taking advantage of the developers who are out there who come to them with ideas and new products.

The Safaricom rep states that this is not the case any longer. They partner with MobilePlanet and Cellulant (as examples, but it’s a poor one because they’re established companies now). He says that at first they start off with a big chunk of the revenues, but as the product does better, then the developer will get more of the share.

Basically, we get no straight answers from Safaricom and only promises of better things in the future with no details.

Jose Henriques, Executive Head: Online Product Management, Vodacom South Africa

6.65% of the African population currently uses PC internet. The top ten countries make up 85% of that.

Some more stats:

  • Africa represents 15% of the world population, but only 3.9″% of the world’s PC internet usage.
  • Africa’s PC Internet users have increased by 1359% from 2000 to 2009.
  • The global service revenue generated from subscriptions to mobile internet access are forecasted by Informa Telecoms & Media to rise from $57 billion in 2008 to $120 billion in 2013.
  • Mobile ad revenue is estimated to be at $2 billion by 2014. Total value of marketing spend on mobile to be around $6 billion.
  • Mobile subscription rose from 54 million to almost 350 million between 2003 and 2008.
  • On average there are 60% mobile penetration in the world. In developing countries the figure stands at 48% , which is 8x bigger than in 2000.
  • Lack of fixed-line access will drive huge mobile internet usage and revenues.
  • Vodacom generates 49 million ad impressions per month in South Africa (big opportunity).

(Full presentation by Jose Henriques from Vodacom South Africa)

Questions
What has been defined as a smartphone, is not what we define one as today. How would you define it?
Cheapest data enabled devices are about 2000Ksh in Kenya. Safaricom thinks that these are smartphones.

Mpesa… Why is Safaricom unable to cooperate with and provide third-party access (opening their APIs) to developers in Kenya for Mpesa?

The Safaricom rep says that they are willing to do this, and that they’re hungry for people to come in with ideas and products. No specifics given on this. @TMSruge, the moderator, asks her to provide details on how they are actively trying to seek out and help grow this as there is no API or SDK.

@wanjiku says she’s heard Safaricom saying that they have a tendency to do well with big companies, but holding smaller company money for 3-4 months, hurting their cash flow.

Steve Vosloo asks what types of local content are people really willing to pay for?
The Safaricom rep is out of touch… she states that, “no one is willing to pay anything for mobile content”. This is bunk.

Rick Joubert comes in to state facts on how much money there is being made in South Africa in mobile content, $540 million is the real number just in SA. It’s not whether people will pay or not, it’s whether they find value in local content.

A question was asked of Safaricom, why they don’t open up the ability for third-party service providers to bill consumers? The answer by Safaricom is that they are. (I can’t confirm this)

We have Zap, Mobile Pay, Mpesa, etc… when are we going to have an agnostic system to send/receive money? by @kahenya

MTN rep says to come to Uganda to see this working. It’s there working on the MTN system. It’s a serious issue of not having your payment system to go beyond your own network.

Mpesa is a wall gardened. Kahenya and Teddy Ruge ask when there will be a need to NOT walk around with 3 handsets to send money within each one.

Safaricom states that they can already do this within their system. They lay the blame at the regulators feet for why it hasn’t happened.

**Lunch**

Brett StClair of AdMob

Brett starts by asking, “what is mobile internet?” It’s a website that is built for mobile handsets. Admob puts banner advertisement on these sites. They server 12-14 billion advertisements into this network each month. The man on the street can earn revenues start advertising today. There’s a 60% payout to publishers.

Have access to 53 countries in Africa. Monthly ads serves is 750 million in Africa alone.

African Mobile Web currently has South Africa, Nigeria, Libya, Egypt and Kenya as the top 5. Data prices do have a huge effect on the advertising revenues available in Africa.

Nokia 3110c is the most pervasive phone in the market (3.8%), Samsung E250 is at 3.7% penetration. Top smartphones are the Nokia N70 at 10.8% and then Nokia 6300 at 10.3% and then the iPhone at 8.2%.

Top reasons why South Africa is working:

  • 5 million fixed line internet vs 10 million mobile internet users
  • Strong operator billing infrastructure
  • However, mobile money is not mature yet
  • Early adoption by premium traditional publishers
  • consumers traditionally have had a fast adoption rate
  • Due to vast competition for impressions average CPC pricing grew from $0.03 to $0.27 in a year.

Is Africa next? The rest of the world thinks it is, but we need to get the local people to understand this.

Cheapest inventory in the world is in Africa… global accesss average is $0.03, in Africa it’s at $0.01. Local contet providers will benefit as they understand the local African consumers.

Key to making this work:

  1. 3g network coverage
  2. Cheap data pricing
  3. GPRS enabled handset penetration

What are the opportunities in Africa?
Strong tend to follow the West and South Africa. Paid for content, reliant on operator billing. Free content, which is ad funded. The top publisher types are communities, portals and downloads. The top categories are music, religion (15%), games and brands.

African traffic is made of 54% Nokia handsets, then 18% Samsung handsets. iPhone requests make up 18 million impressions in Africa.

Debates on the Mobile Web at MobileActive ’08

We just finished a really good conversation on the the future of the mobile web at MobileActive ’08. Toni Eliasz of Ungana Afrika moderated a discussion where one side of the room was charged with arguing against the mobile web, and the other half for the mobile web. I sat on the “for” side of the room.

MobileActive '08

My Position

The web is made up of data, and we generally think of it as what we access via the PC. However, that same data can be accessed and added to through mobile phones as well. Whether its basic SMS, Java apps or direct web browsing. Data is data – how you access is what matters.

Some of the issues holding back penetration of the mobile web:

  • Accessibility – though this gets better every year
  • Cost – The reason why you can’t directly compare interaction or development of apps and services that use the mobile phone to the PC is because of the cost associated with data and SMS costs on mobile right now.
  • Interface – usability can be a major problem on Java apps, and 160 characters is very limiting.

But the basic truth remains. If you can access and contribute to the global databases of content, then you are in fact on the mobile web.

The mobile web is already here. It’s happening now.

Mobile Web Questions

Mobile Web questions
The questions we debated.

Rabble’s and Blaine’s Positions

Rabble, creator of Yahoo’s Fire Eagle, and Blaine, the original architect of Twitter, continued the discussion with me afterward. The claim here is that the only truly mobile web device is the iPhone, all else is negligible – maybe not in theory, but in action.

Rabble explaining how we access the Mobile Web

Rabble tells me that it’s much like saying that if you could get the web through this blurry glass, even if it’s feasible, it’s not useful or likely. He’s got a good point…

[final note: I was preoccupied while trying to post this with Rabble and Blaines’ conversation…]