GEOPOTATO

GEOPOTATO saves potato crops from late blight infetstation. Late blight is a devastating fungal disease which threatens food security of Bangladesh. GEOPOTATO is an early warning system, automatically computes weather data (humidity, temperature etc.) coming from Automatic Weather Stations and uses satellite image processing data to measure potato crop biomass growth. Then it determines whether the existing weather parameters are conducive for late blight attack. If the conditions are favorable to cause late blight, preventive advice is sent 3 days in advance via SMS and voice messages to subscribed farmers. GEOPOTATO service provides precise fungicide spray advice that reduces both economic and environmental consequence of fungicide use. Furthermore, due to the high precision of our solution, we provide spray advice only when compulsory, greatly reducing the frequency.

Potato is the 2nd largest crop in Bangladesh and the largest as a cash crop. It has strong bearing on country’s food security, and livelihood of 7.5 million potato farmers in Bangladesh. Accounting for 25-57% yield loss, late blight is the primary problem of potato cultivation in Bangladesh. Frequent and costly applications of fungicides have been the only resort for farmers to control late blight. However, farmers apply fungicide even when it is not necessary and incur economic and environmental losses due to over spraying. But in the end, often fail to prevent or cure late blight attack because they cannot effectively predict when it will occur.

Traditionally, potato farmers spray fungicide almost randomly as they cannot effectively predict late blight infestation, which depends on a complex dynamics of weather parameters and crop growth rate. In addition, farmers are often misguided to use inappropriate fungicide recommended by retailers who have vested interest to suggest fungicide that gives them higher profit.   

Potato is the 2nd largest crop in Bangladesh and the largest as a cash crop. It has strong bearing on country’s food security, and livelihood of 7.5 million potato farmers in Bangladesh. Accounting for 25-57% yield loss, late blight is the primary problem of potato cultivation in Bangladesh. Frequent and costly applications of fungicides have been the only resort for farmers to control late blight. However, farmers apply fungicide even when it is not necessary and incur economic and environmental losses due to over spraying. But in the end, often fail to prevent or cure late blight attack because they cannot effectively predict when it will occur.

Traditionally, potato farmers spray fungicide almost randomly as they cannot effectively predict late blight infestation, which depends on a complex dynamics of weather parameters and crop growth rate. In addition, farmers are often misguided to use inappropriate fungicide recommended by retailers who have vested interest to suggest fungicide that gives them higher profit.  

The GEOPOTATO is full functioning automated decision support service (DSS) for an optimal control strategy of late blight in potato for Bangladeshi farmers. 

The GEOPOTATO DSS automatically computes real time weather data (temperature and humidity) through Automated Weather Stations (AWS) installed at the service locations and determines whether the existing weather condition is conducive to a potential attack of late blight disease. Satellite image data is used to measure the potato crop biomass so that exposed newly developed biomass could be identified. Crop Growth model and Late blight disease model combine and analyse the weather data and crop biomass data. The analysis results in a timely spray advice provided to the farmers through SMS and voice messages. If the conditions are conducive to late blight, SMS and voice message is automatically dispatched to farmers with preventive advice. Presently GEOPOTATO services are provided to farmers in three most prominent potato producing districts of Bangladesh i.e. Rangpur, Dinajpur and Munshiganj. 

Innovativeness is at the core of GEOPOTATO service. It combines geodata like satellite image processing, automatic weather station etc. as input data. Based on this input data, state-of-the-art late blight disease model predicts conducive period with an acceptable standard of accuracy and scientifically. This prediction is then converted into highly localized fungicide application advice for individual farmers based on their planting weeks and geographic location. While the technology doing the computation in the background is very sophisticated, the delivery of the service is through SMS/ voice message taking advantage of high level of mobile network penetration among farmers. There is no internet or smartphone dependency to receive the service. GEOPOTATO’s innovation has been recognized by several entities such as (1) the International Telecommunication Union (ITU), a UN body, specialized body through its World Summit for Information Services (WSIS) Prizes 2019 Champion among thousands of innovations submitted to the competition from around the world, (2) and also a Winner of World Summit Award (WSA) 2018 competing with over 400 submissions from over 100 UN member in the category Environment & Green Energy. (3) GEOPOTATO also emerged as Champion of Tech4Farmers Asia Challenge 2018 of USAID.     

Late blight can be controlled to a certain degree by frequent and costly applications of fungicides. This however results in excessive use of fungicides which is detrimental for the environment and human health. It is also costly for farmers and the cost of frequent fungicide spraying is prohibitive for small holder farmers. Also, the resulting level of control heavily depends on (i) the timing of fungicide application and (ii) the use of fungicides in relation to local weather conditions, crop development and disease pressure. So, even frequent spraying does not have desired results always.

The efficiency of late blight control in potatoes can be considerably improved through informing farmers timely about the predicted infection periods and effectiveness of past spray applications. This The following is GEOPOTATO’s Theory of Change which has proved demonstrated that optimal late blight control is possible with a late blight alert .  

Figure: GEOPOTATO’s Theory of Change  

Analyzing 22 demo plots from 11 upazilas, Wageningen University reported that GEOPOTATO farmers were able to increase average yield increase by 15% in last season. Half of these demo plots were GEOPOTATO demo service receivers, who received GEOPOTATO alert and acted accordingly, and the other half (control group) sprayed fungicide according to conventional practice. In another study done by Wageningen University & Research it is concluded that the average additional financial benefit for individual farmer is around USD 220 per planting season for GEOPOTATO service receiving farmers in comparison to non-GEOPOTATO farmers. This study was conducted on 1,800 farmers from 18 upazillas.

The GEOPOTATO solution is currently has 41,812 registered farmers from 2018-19 potato season. In 2019-20 potato season we plan to reach more than 75,000 farmers. After 5 years, we aim to reach 500,000 farmers which is around three-fourth of smallholder potato farmers of Bangladesh.

The mission of GEOPOTATO is to optimally control late blight in potato by minimal and timely fungicide application by subsistence farmers. The vision of GEOPOTATO is to improve food and nutrition security by sustainable farming practice in potato cultivation.

Right now, we are at a stage where we have demonstrated impact of the GeoPotato service with funding support from donors and research organizations at beneficiary level (reported by Wageningen University), and also at agro input company level with respect to their branding (supported by independent study by Boston-based impact research NGO PAD). Now, in next steps we plan to employ a two-pronged approach. The first is B2C approach to reach individual potato farmers as subscribers through farmer associations, cooperatives, microcredit providers etc. And second is a B2B approach to acquire organizational customers of three categories: (1) potato processing companies; (2) government and non-government organizations providing support to potato farmers in the form of loans, micro-credit, advisory support etc. (3) crop insurance companies.

Our goals for the next 5 years may be put in the following phases:

Phase 1: Expand the current GeoPotato service to other regions of Bangladesh

Phase 2: We will develop late blight disease control advisory services for other tuber crops in Bangladesh, such as tomatoes.   

Phase 3: Based on the experience with late blight disease prediction advisory, we will develop similar geo-data based services for other crops

Phase 4: We will take the service to other countries like India, which have similar cropping and climate patterns.

To be able to achieve the vision, we need to address the following barriers:

Market:

The first barrier is to convince the market (farmers, company and government representatives) that prediction of late blight disease to a fair degree of accuracy is possible. Our positive preliminary results based on RCT trials in limited geographies, should be circulated through mass communication campaign which is possible if we conduct RCT at boarder scale in diverse agro-ecology.

Technical:

We need to constantly fine-tune the algorithms and the model for further improvement in accuracy in other geographic locations. Being an intelligent system GEOPOTATO needs to constantly learn and evolve from more data from as diverse locations. Hence we need to expand to new geographic locations which is a barrier. Setting up Automated Weather Stations in new geographic locations is subject to high import taxes. 

Financial:

We need more financial strength to address the above two barriers. The Netherlands Space Office provided seed funding for developing GEOPOTATO. We are investing ourselves to keep the service operational in next season. Within the next 5 years we plan to make this service self-sustaining. For that we need to reach a critical mass of beneficiaries which is approximately 350,000 farmers.

Legal:

There are no legal barriers as such since our advice and recommendations are around spraying of fungicides, which is already a well-accepted practice globally. In addition, the government has not put any legal barriers in providing the service. 

For further scaling up our impact studies and deployments to new geographic locations, in addition to Boston-based Precision Agriculture for Development (PAD), we are also partnering with Connecticut-based Yale Research Initiative on Innovation and Scale (Y-RISE) and Innovations for Poverty Action (IPA) in Bangladesh. These will help us create further evidence to convince the market players regarding the reliability of GeoPotato and its impact on livelihoods and food security. We are also planning some extensive communication strategies around GeoPotato – both through traditional events and marketing channels, and also through digital marketing.

Technical:

To further fine-tune the model and expand the prediction service to other crops and new geographies, mPower has been investing heavily to build up the capacity of the GeoPotato technical team. We are also partnering with geo-data processing companies and artificial intelligent / machine learning experts and companies. We are also developing capacity through partnerships to develop a locally produced Automated Weather Station so that we do not have to import expensive ones from abroad. A working prototype ready to be tested in the field will be ready by November 2019.

Financial:

We are engaged in conversations, which are at various degrees of maturity, with potential impact funders. We also applied for several innovation funds in collaboration with Boston-based IPA, PAD and Connecticut-based YRISE.

GEOPOTATO is a service owned by mPower Social Enterprises Ltd.

7 people is involved directly in GEOPOTATO solution. 

The owner of the solution, mPower Social Enterprises Ltd., is a well-established and recognized ICT solution and service provider in the development sectors having 10+ years’ experience in designing, developing and implementing hundreds of ICT based development projects. mPower also has an incubation wing through which innovative ICT-based solutions are later turn into sustainable and profitable business as spin-offs. So the organizational experience of the mPower is a great asset to for the GeoPotato team.

Another unique aspect of the success behind The Geopotato is its extremely multi-disciplinary team, brief profiles of key members are given below: 

Mridul Chowdhury, founder and CEO of mPower, a graduate from Kennedy School of Government, Harvard University, is the Chief Executive of the solution. He has 20+ years of experience in innovating ICT4D solutions in various development domains.

Shah Md. Mushfiqur Rahman, Team Leader, has been working with mPower for the last 3 years and has been involved in leading teams of ICT based agriculture and livestock advisory service solutions. Before joining mPower, he worked with various local and multinational companies for many years in various managerial capacities.

Md. Nazrul Islam, e-Agriculture Specialist, is a senior retired government officer from the Ministry of Agriculture, with specialization in leading innovative ICT-based solutions. Badal Kumar Dash, GIS & Remote Sensing Expert, has many years of experience in managing GIS and remote sensing data driven projects

Devashish Chowdhury, Software Development Lead the tech lead for GeoPotato, has many years of experience in Software Architecture Designing and Development.

For putting together the service of GEOPOTATO and operationalize it, we partnered with Wageningen Plant Research of Wageningen University & Research as the research partner. For satellite image processing, we partnered with TerraSphere Imaging & GIS B.V.

For piloting business models we also partnered with agribusinesses including Bayer Crop Sciences Ltd., Lalteer Seeds and Kellogg’s.

Full time staff: 2 - Team Lead - GIS & RS Specialist

Part time staff: 5 - Advisor - Plant Pathologist - Satellite Image Processing Specialist - Agronomist - Software Development Lead  

Your business model describes how you provide value to the populations you serve, both in terms of impact and in terms of revenue. You can summarize it through who your key customers and beneficiaries are, and what product or service you provide and how you provide it.

The GeoPotato business model is based on a service subscription mode. The service is extended directly towards the farmers, while the input suppliers and potato processing companies subscribing to the service for promoting their product.

We will expand through both B2B (organizational farmer-centric) and B2C (individual farmer-centric) approaches but primarily through the B2B model as this provides us with scale to reach a large segment of farmers. Potato is the second largest crop of country and most important cash crop for smallholders. The return of investment from potato farming is substantial hence farmers are willing to cultivate potato in ample quantities. But potato is an input dependent crop thus fungicide companies earn sizeable revenue. The market competition for increase the sales is the reason why GEOPOTATO has commercial potential among input companies. By subscribing to the service they will be able to promote their product among farmers.

Boston based NGO- PAD have concluded that, branded messages in the GEOPOTATO service resulted in an 18% increase in the usage of brand recommended products. Based on this finding PAD suggest that the partnership model between the GEOPOTATO service and private input suppliers may provide a win-win-win solution.

The main source of revenue for the service is revenue from Branded Messaging Alert Service. Input companies with a view to promoting their products, will purchase the service at specific costs per service message. These messages will be the medium to reach the farmers, as well as a medium for promoting the input companies.

Having a full-functioning GEOPOTATO platform in place, all we need now is to increase the number of beneficiary farmers to expand our services among more farmers. This will increase the number of cases we can serve in a day without significant additional investment for supporting staff and logistics cost which will in turn increase our revenue stream.

The service will gain its revenue from branded messaging service cost, charged to subscribe input companies. Field agents of the input companies will subscribe to the messaging service on a B2B mode for promoting their brand through the Geopotato service sent to farmers. They will be charged at 0.52 USD per message sent to each farmer. By reaching target farmer count of 50,000 in the 1st season, total revenue from 10 messages to each farmer will be 26,000 USD. As we expect the farmer numbers to grow up to 350,000 farmers within season 3, total accumulated revenue will be 313,500 USD.

These revenue will cover the service expenses and other fixed costs, totalling to 300,000 USD within season 3. This will incur an accumulated operational profit of 10,000 USD within season 3. Thus, the breakeven point will be achieved in mid-season 3.

We are applying to Tiger Challenge because we believe that this will help us with covering some of the financial constraints we currently have with respect to expanding rapidly, particularly with respect to creating evidence of impact and acceptance of the service at a broader scale. But more importantly, this would greatly help us with gaining the credibility needed to raise Series A equity investment to rapidly expand our business. In order for our data analytics and targeted advertisement side of the revenue base to kick in, we need a significant enough number of subscribers. We need the initial investment to reach that tipping point to ensure that a competitor cannot easily overtake the market that we have been creating over the past few years.

Besides the financial aspect, we believe that a strong affiliation with the MIT Solve Team will help us with fine-tuning our business model, networking for expansion and replication of the model to other countries. Since mPower, the mother organization, was founded with funding from the MIT 100K Business Competition in 2008, there is already a connection with the MIT Media Lab. Winning this Challenge can strengthen that long-standing relationship further.  

Also, MIT Solve Team’s particular focus on advancing lasting solutions from tech entrepreneurs to address the world’s most pressing problems is the reason GEOPOTATO wants to partner up with Solve.

We would like to partner with the following kinds of organizations either as clients, knowledge/ capacity partners or with investors/ funders: 

Research/ Impact Study:

To conduct impact assessment and measuring the willingness to pay of beneficiary GEOPOTATO has been in the process of collaborating with Yale Research Initiative on Innovation and Scale (Y-RISE) and Innovations for Poverty Action (IPA) Bangladesh .

Investors:

We want to partner up with equity investor to scale the solution up.

Organizational Clients:

We want to partner up with agribusiness, agro-input companies for commercializing the service. For example, we have been working with Bayer Crop Science Bangladesh to build a commercial version of GEOPOTATO. We have piloted Bayer branded recommendation in selective upazila in Bangladesh.

Technical Capacity/ Knowledge:

We want to partner with AI (Artificial Intelligence), ML (Machine Learning) and geo-data processing companies around the world to develop new application of the GEOPOTATO technology. We believe innovation is a continuous journey and more innovative angle is possible to explore in partnership with specialized deep techs.

Design of Solutions with Go-to-Market Strategy:

mPower want to partner up with renowned consulting firms to receive consultation about integrating business and technology strategy in go-to-market product and service design compatible with market dynamics to capture market value, and change organizational capabilities to reflect evolving markets and technological dynamics.