Open source and mobile banking?
by Mark Pickens : Monday, February 23, 2009
I’m just back from the Mobile World Congress. Tomorrow I’ll write about banking agents and share our presentation from Barcelona. Meantime, I came home with a question: I can load any software I want onto my PC, so why can’t I do the same with my phone? Better yet, why can’t poor people?
I’d love to see a boom of cheap m-banking software, designed by people who know how poor people want to use their phones. Although lower-income, non-Western users make up 80% of the world’s new mobile consumers, the guys in Finland, Sweden and South Korea still decide how people’s phones look and feel. But for how long? I’m interested, because I expect usability to be one key in how fast poor people are willing to adopt mobile-based financial services (which CGAP believes can blow open the frontier for access to finance for the poor).
We already know open source software has sparked much of the innovation in computers. The world’s #2 web browser, Firefox, is an open source project, and one which Microsoft has borrowed from extensively. You like tabbed browsing? Thank Firefox. Without it, you may very well still be browsing the way Redmond, Washington thought you wanted back in 1998.
So why can’t there be an equally open market for the software on phones? That might open a world of cheap and culturally specific m-banking applications, designed by people who understand how poor people use their phones, in local scripts where people don’t use the roman alphabet.
Some would point to Apple’s iPhone store where customers have downloaded 30 million applications onto $300 iPhones. Maybe it is. To date, handset manufacturers and big operators seem willing to offer that freedom as a perk for people who can afford a $300 iPhone. And in the end, you can only download an app into Apple’s otherwise carefully controlled operating world. Far more interesting to me is Google’s Android, a totally open source operating system for mobile phones. The first Android-enabled phone hit the market in October 2008.
So why isn’t this happening yet for poor people’s phones? I spoke with half a dozen handset engineers in Barcelona, and the answer is straightforward. Up to now, Nokia, LG, Samsung would rather sell souped-up smart phones aimed at European, North American and East Asian consumers. Their math isn’t wrong: sell $30 handsets to Bharti Airtel’s 85 million clients, who probably buy a new phone once every 5 years, or pitch smart phones for 10 times the price to Verizon’s 72 million clients, who in good times upgrade their phone every couple of years. To command top shelf pricing, manufacturers offer phones with lots of add-ons embedded in a locked operating system. To top it off, your average operator like Verizon also wants to see the phone as a controlled space – they want you seeing shortcuts to all their new music, video and gaming services. The idea of you short circuiting this by blithely downloading another application is unappealing. So far, this means phones stay locked and handset manufacturers are focused on developed markets.
But for how long? 80% of the new mobile users are coming from developing countries. The typical mobile customer is now low-income not rich, young, from a non-Western society, and has a totally different experience of technology than you, me or the engineer in Helsinki can imagine. For instance, I have a deep familiarity with the concept of navigating through levels of menus, thanks to Microsoft.
But the Indian farmer? He may very well think in terms of a map, with multiple links between “locations” of information and applications. And when it comes to managing his money through his phone, do we know yet what he wants? I’d much rather see a 1,000 flowers blooming in the form of open source software designed by people thinking about the Indian farmer – then we’ll find out what makes a highly usable, friendly mobile experience for him. Lots of interesting insights into how poor people use mobile phones are available at Afrigadget.
It would be great to see manufacturers create app stores for emerging market users, and try out opening up their handsets to customization by code writers in poor countries.
February 24th, 2009 at 9:42 am, Simon Cavill ()
Hi Mark,
Great article, but you miss out on one critical point – Its one thing to write mobile apps but its totally another to try and deploy them! Java capabilities vary widely across handsets, manufacturers and even batches of the same device, so you end up deploying at least 50 plus versions for a European deployment.
Even in the “West” only about 50-60% of handsets in common use are configured correctly to run a mobile data session necessary to download an app in the first place. Elsewhere, mobile data coverage is patchy at best and can be expensive. Also many handsets are second-hand with incorrect data settings. Add to that mix a whole slew of cheap handsets with no Java capability at all and its no surprise why java deployments are so difficult to put together in the developing world.
Instead, we and others deploying mobile money applications utilise universal technologies such as USSD, IVR and SMS to communicate with end users phones and only deploy side-loaded secure Java apps to agents on a pre-configured phone. Even here, our apps auto-detect if a mobile data service is available (not often) and switch over to either encrypted SMS or USSD strings to communicate with our servers.
Best regards,
Simon


9 Comments
February 23rd, 2009 at 2:26 pm, Putting people first » Open source and mobile banking ()
[...] Read full story Leave a Reply [...]