Mobile Meals – moMEALS is an m-Marketing system implemented by UgaBYTES and supported by CTA. moMEALS processes started with pondering the issue of hunger and poverty – MDG No.1. Realities of rural people were explored, as a starting point. The feelings of food vendors visited. And the role of middlemen, analyzed. Farmers want to be able to tell middlemen to buy or forfeit the produce to the next buyer. But, not being told to let rot their produce or sell at low price, by middlemen. Process 2 involved a review of the ICT tools that could be used in the fight against hunger and poverty. No single technology could be comprehensive at serving the purpose. So - Mobile devices, the web and reduced telecentres, were adapted. The slogan – Let every hut smile – would become the driving statement. Now – process 3 is covering the basic functions of moMEALS database. Basic statics underlying the rational for the system design are also relayed.

How the system is structured and will be used

MoMEALS platform is built on a SMART texting system. It starts with a user typing in ‘buyer’ – for food vendors or last mile consumer; or ‘seller’ – for small scale farmers. The message then triggers off requests for further information – in a set of three questions; farmer’s or buyer’s name, location by district and nearest town. Subsequently, the user is assigned a login pin number as a system password. The user’s mobile number is his or her user name.

The user posts an advert after signing in; if using the mobile phone, the user states if he wants to ‘buyproduct’ or ‘sellproduct’. This triggers off a couple questions e.g, what you want to sell or buy, quantity on sell or for buying and price desired. If using web panel, the user accesses two things; the home page and match. Here, the user is able to 1) create products; 2) see his or her goods; 3) match products; 4) search products; or 5) see product posts from other users. The three stage process ends in an advert that states the users; name, product on sell or purchase, price of good desired, location of user, mobile phone number of user and product code. The registration information and product advert information are linked in the summary. Each user’s product offer is treated independently. The user can post unlimited adverts.

The second part of the system is the query service. The user, either a seller or buyer, queries the system by typing in MATCH ME or MATCH ALL. The request aggregates all available adverts that match a specific user’s product or all available products matching a user’s products. An advert can be updated depending on changes in market conditions. Alternatively, an advert may be deleted or deactivated, if the sale in completed. Information is handled by a database. It is accessed by mobile phones and web. Users at the micro telecentres can print out or write on community market boards all the offers that are available.

Homogenous users by geography, type of product produced or purchased may constitute an SMS posting group. The list allows giving updates to the group that meet specific user needs. Updates are posted by the administrator and or authorized managers. Posting may be done via a mobile phone or via a web interface, which has ability to schedule messages that would be appearing more than once. The web generates progressive reports and supports massive user registration.

How the registration form devolved

The first form had 67 simple entries – even if this would be for good intentions – only three entries would trace into the SMS system. The 67 entries would have helped in matching users and in future evaluation of the system. That is why fields like; farmer’s name, phone number, district, village, sub county, parish, main 3 cash crops, main 3 food crops, main 3 commercial animals, type of inputs and throughputs and associated costs to the farmers, marketing chain, gender of user, age, highest education and many other were included in the first form. But the bigger the form the high the user would fork to use the system. So – entries had to be collapsed, leaving only relevant and necessary entries. The new form would thus only have; name of the farmer, district – listed for selection, and nearest town – listed for selection. The phone number would be captured automatically. Meaning that, users would register on the system in 3 steps – involving 3 SMSs and associated costs. But the first 3,112 users, all farmers, gave information to the 67 entries. The information was collected in a physical survey. So – some of this information may continue to be collected as long as it is not collected using SMS system e.g. if it is by email, groups, commissioned enumerators, web, etc, so as to firm M&E processes.

Product that will be offered via the system

One would think that food crops would be different from cash crops – and yet largely not, for small scale farmers. Implying that commercialization of small scale farms would have to be sensitive to nurture a balance between commerce and subsistence growth to avoid reducing poverty while hunger get on the rise, as a result. For the 3112 potential users talked to, a list of 31 cash crops was stated, and more than half of it, 16 crops, would be re-stated as food crops, in a proceeding entry. But the response counts were 8595 and 8284 for cash crops and food crops respectively. Meaning, rural people produce the same for food crops and diversify a little for cash crops. Evidently, 80.7% of the cash crops – coffee, 17.9%; mangoes, 2.2%; cassava, 6.3%; banana, 11.4%; beans, 10.8%; tomatoes, 9.4% and maize, 22.7% - come from 22% of the total crops grown for commerce i.e. 7/31 as stated by farmers. Meaning, the remaining 79.3% of commercial crops would be seen as pseudo. On contrary, the food crop sector was a little better.  92% of the food crops – banana, 20.9%; potatoes, 5%; cassava, 27.4%; G. nuts, 3%; maize, 5.5%; tomatoes, 2.6%; beans, 13.1%; and S. potatoes, 15.4% - came from 50% of the total food crops. Meaning, 50% of the food crops would be seen as pseudo.

By no doubt, the results are representative, but it would be bound to change by district – due to natural agricultural zonation and socio-cultural traditions of the people living in each area, as the system maps over a bigger national coverage. So – the comprehensive list of crops would be used, to include those agricultural products, outside the survey districts. Even then, all crops would not be added in the beta, database – as on-going lessons would inform future product additions and deletions.

Product categorizations would, thus, be adapted as a way of collapsing the long list. Crop product types, adapted would include: fruits with 12 products; vegetables with 15; grains with 11 and cash crops with 8 products. The animal sector would have 9 product types: animals with 7 products; animal meats with 7 products; animal milk with 3 products; Skin with 4; Eggs with 7; manure with 6; fish with 3 and bees with 5 products. Farmers said that they mainly rear cows (31.2%), goats (13.9%), and chicken (31.8%), representing 76% of the total, which would be coming from 37% of number of animal reared (3/8). The value though, is expected to be significantly affected, if scaled over a national coverage. Summing it – a total of 14 product types with a total of 93 products would be listed in the system.

Likely hiccups existing presently

It would be frustrating if it were stated that the system wouldn’t get hiccups. For, indeed farmers pointed out a number of them.  The most significant challenges would be that of; unreliable electricity supply for charging phones (21.7%), high tarrifs (19.4%), unstable telecommunication network (13.2%), unavailability of airtime cards in the village (11.7%), battery wears out very fast (9.3%), lack of money to buy mobile phones (8.7%), failure to use phone applications or functions (5.7%), and unsupportive customer care service centres (3.3%). The responses come from 3120 farmers interviewed in Mukono, Mpigi, and Nakaseke. The percentages were based on 6308 counts.

Mukono district, which is next to Kampala, was least affected by unstable network at 4.5% while the other two districts were evenly affects 10.8%  for Mpigi and 12.8% Nakaseke. Meaning, unstable telecommunication networks would extensively affect moMEALS rollout as it moves to far rural sites and users. High call tarrifs affects urban areas more than the rural parts – thus it is expected that as we move deeper in the villages, the problem will become less and less. The underlying reason may be that, rural people use to phones strictly for business transaction while the urban users have diversified the use of phone across every part of life. Also battery wear out is move of an urban problem than rural. However, the problem of unreliable power evenly affects semi rural and rural districts. Unavailability of airtimes is a major problem for the rural users – thus may be extensive as moMEALS moves far rural.

The social investor of moMEALS is CTA while the implementer is UgaBYTES.