I just finished reading “Connected Agriculture”; a new report on ICT, mobile and agriculture by Vodafone and Accenture (with support from Oxfam). I found this report very interesting. The aim of this blog post is to share my views and comments, and see if there are other opinions in the community.
As part of our work on Web for Agriculture initiative, and our related projects (W4RA, VBAT, VOICES), I’ve read numerous reports on ICT for agriculture, or ICT for the base of the pyramid (see e.g. my last post on ICT for the BoP). I’m typically skeptical on the need for yet another study. However, this is one of the best, most easy to read, overviews of the potential of ICT/Mobile in agriculture. It is pretty rare to see such studies with a holistic approach centered on farmers, their needs, and elements that could improve their lives. It would have been helpful to add another chapter on how to scale up, in a sustainable manner, the services reviewed in the report so that they are directly usable by the majority of people at the base of the pyramid. This is where the current work of the Web Foundation shows promise, as discussed below.
The report covers four areas (information services, trading services, financial services, and supply chain services) that are rarely put together in one study. All of these areas are critically important for people. Experts in any or all of these areas might not find much that is new to them, but it is a really great global overview. So in few words, I strongly recommend reading it (and sharing your views here!).
Now, let’s go deeper into the analysis of the content. In each of the four topic areas, there is a detailed analysis of current systems and pilots, the identification of challenges, and the possible impact, with very detailed numbers. I know that people love such numbers. But I’m usually very cautious regarding numbers. It is extremely difficult to know if this is reasonable or not, particularly as the underlying model(s) used to make the calculations are not provided and explained. That said, I’m sure that there numbers are based on current pilot outputs. What is important for me is more the qualitative aspects. It is clear that the different topics addressed in the report are critical to improving the livelihood of small-scale producers, who are one of the most disadvantaged segments of population in developing countries.
In that regard, this report is a great contribution to the domain. It identifies and underlines the potential for ICT and mobile technologies in different dimensions to directly increase the income of people.
But I think we need more now. We need an implementation plan, we need to realize this potential and not merely talk and write about it.
The important question is HOW?
From my perspective, this report does not provide answers to this most important question. My understanding, and I may be wrong, is that the proposed action is to scale up the existing systems: m-Pesa has great results in Kenya … scale it up; SMS-based trading platforms have some impact … scale them up; farmer helplines are useful … scale them up.
I’m not sure scaling up these existing services is the only answer. Our experience with VBAT and W4RA, which are addressing farmer helplines and trading platforms, tells us that the current generation of systems are not scalable from the perspective of the technology, farmers abilities and sustainability. Some specific examples:
We are working with One World South Asia on Lifeline’s India, a farmer helpline in India. This farmer helpline is, as described in the report, based on human knowledge workers answering all requests.
Each worker can handle up to 70 requests a day, which covers close to 10% of the worker salary. The figure below show the evolution of sustainability as the service is scaling up.
It is obvious that this is not going to be sustainable. Now let’s look at the problem from another perspective. 85% of the requests that are placed have been asked and answered before, and thus are already in the system (no need for external expert intervention). So if farmers could automatically find the answer they need, without human intervention, the new scenario in terms of sustainability gives the following projection:
It is obvious to me that if we are able to empower farmers, and enable them to use technologies directly, we could help farmers in a way that is sustainable.
In the same way, for trading platforms, the current hype is around SMS services. How can we imagine that SMS will scale up? Some numbers that come from the literature: 4% of people in rural Gambia are able to use SMS (source). 20% of the total population in Bangladesh is able to use SMS (source). Our experience in Mali: where we work in the Mopti, Tominian, Bandiagara, Bankass regions, we’ve encountered almost no one able to use SMS. Here again, trading platforms will scale when farmers will be enabled directly with technologies that they can use.
The Web Foundation is exploring voice technologies as a way for many more people to access information through the Web . The voice channel works on all phones, and can be deployed immediately. In the future, icon-based interfaces on feature phone with IP connections might also become an opportunity. Overall to realize the potential that is underlined in the report, it is essential to invest-in and develop interfaces for people who cannot use text input and output.
It is exactly the same for financial services. There is a huge, and well-deserved buzz around m-Pesa. However, m-Pesa is not accessible to people who cannot use the text interface (SMS, USSD, etc.). So there is an element of exclusion in such initiatives even though inclusion is cited as their success factors. For those who cannot use text, they must rely on intermediaries who might be hard to find, dishonest or expensive.
Mobile technologies provide incredible opportunities to realize the promise of ICT for development. I am convinced that in agriculture and finance, some services that have been tested have demonstrated their ability to increase income and improve people lives. The current generation of services are not scalable, and not usable by all farmers at this point in time. It is therefore essential to focus on a new generation of mobile-based financial services, trading platforms, helplines, and other services. that are usable directly by all farmers, and that are financially sustainable. Most of these goals are in the roadmap of our Web for Agriculture and Voice Browsing initiatives.
October 31, 2011
Just around the "only 4% of rural gambians use SMS comment" - it seems a little strange that this is quoted as a sourced fact. The comment comes from a blog which publishes the data without supporting documented results and states that the localised results are to long to publish. Even if this statistic is correct it is only based upon 50 people without any demographic information. Further to this the actual questions do not appear to include a "do you use SMS" question - there does appear to be a "What facilities do you enjoy most from your mobile?" question (which seems to be a pick one feature) but that really isn't the same thing as "do you use SMS."As a result I feel that the statement is highly misleading and will end up being quoted endlessly by people who don't check their facts. If the demographics, locational split and question asked are all reasonable (and this needs investigation) - The correct statement should probably be - In a 2010 survey only 2 out of 50 people questioned in Bansang - Rural Gambia - said they used SMS
November 2, 2011
Hi Rob,you are making a valid point, and to be honnest, I shared, in some way, your feeling before making the post. I think there are different dimensions to consider.First of all, is this information right or wrong ? is it really misleading ? I don't think so. It is a comparative study between rural and urban users, and in that regard the difference in the result is qualitatively significant, even if not comprehensive. This leads to my second point. We are working in Sahel (Mali, Burkina, Northern Ghana) with farmers and our experience is that SMS usage is close to zero (see our report at http://public.webf2.wpengine.com/2011/01/Voices-W4RA_Public_Report_V3.0.doc ) (this is purely anecdotal, and concern a couple of thousands people only)Unfortunately, I haven't been able to find any real study that would investigate this exact point. The two I'm citing in the post are the only ones i've found in the last five years or so. I've been unable to find any other research that would show similar or different results that would let me think that the people and communities we are working with are specific or at the contrary quite representative of the situation in most rural context in developing regions.It is an essential information for further deployment of ICT/Mobile services at the farmer level. I'm sure that e.g. operators have more details about how many simcards are not sending any SMS, or what is SMS usage in rural areas, but I've been unable to get any detailed information. If anybody has any other studies or information on that matter, I would be very interested to get them.steph
Notes on the upcoming Gambian presidential election » oAfrica
November 8, 2011
[...] is not widely used in rural areas. Internet penetration remains low. Fewer than 10% of Gambians (less than 200,000) [...]
A Voice of the Web » Blog Archive » Scalability strategies in mobile ICTD
September 10, 2012
[...] that shows that you need to reach at least 50.000 villages before breaking even (see details in a previous post I wrote about this case). In such cases, there is no other choice to ensure a long term sustainability of a [...]
October 25, 2012
Stephane, about SMS agricultural service uptakes. CocoaLink in Ghana has good evidence on the upper limits of what can be achieved. In villages served by CocoaLink's Pilot information service, which I studied for a report for UNESCO-INRULED (ICT for Rural Education and Development, see http://www.inruled.org/en/, report forthcoming) the SMS service for cocoa farmers was being used by 25% of the adult population of the targeted villages, after just 12 months of the pilot. Original estimates of service uptake have been raised by two orders of magnitude. It looks probable that 20% of this agricultural sector in Ghana will be recruited to the SMS information and support service by 2014. The observation that some farmers are not able to use the SMS interface is correct, but there are many anecdotal reports of older farmers using their children or grandchildren to handle the SMS interface on their behalf.I have also studied projects where take-up is perilously low, almost non-existent. The use of well designed services, local partners, and careful project planning makes all the difference. It is great to see you discuss the sustainability of these services in proper technical detail. The pilot projects out there at the moment are (correctly) focussing on content and service establishment but will need much more automation to become viable. I am working with some services in SSA which are tackling this head-on, with business plans that aim at profit from running the services. I'm working in this area and would be happy to keep in touch with you about it, especially on your question about "dead" simcards in rural settings. Stephen Haggard
October 26, 2012
Hi Stephen,thank you for your comment. I'm looking forward to reading your report. About anecdotal evidences about literate intermediaries, you are right this is happening. But based on my experience, this has a limited effect and this is highly dependent on the context.For instance, in Kenya the SMS usage level is very high, so such phenomenon are more likely to happen. In Mali, less then 20% of the subscribers have ever used (send or receive) a SMS, so this is very unlikely to happen. Even in case it is possible, it is less than a satisfactory option as it still creates barriers between the targeted end-user and ICT.BestStephane