DatAgro: SMS-Based Agricultural Information for Rural Chile
How DatAgro delivered SMS agricultural information to cooperative members in Chile's VI Region.
Executive summary
Chile's mobile boom made DatAgro possible. According to the International Telecommunication Union (ITU), mobile subscriptions in Chile rose from about 14.8 million in 2008 to about 19.9 million in 2010, a gain of about a third in two years. Farmers could already receive a text message on their own basic phones, so the service needed no new equipment and no internet connection.
Between 2008 and 2010, DatAgro sent agricultural information by SMS to members of Coopeumo, a farmers' cooperative in Chile's VI Region (O'Higgins). By the end of the project, 238 members were participating, and the team had built a bank of more than 500 selected messages. DatAgro won recognition in Latin America and internationally in 2009 and 2010. Two evaluations documented how the program worked and how members took part.
This case study covers how DatAgro worked, what it delivered, and how members took part.
The problem: an information gap in rural agriculture
Farmers in Chile's VI Region (O'Higgins) needed timely information on market prices, weather and agronomic practice, and that information was hard to reach. Rural connectivity was very low. According to figures from the Fundación para la Innovación Agraria (FIA), connectivity ranged between 1 and 2% of the rural population.
What farmers lacked
- Real-time market prices for their crops
- Weather forecasts and seasonal alerts
- Agricultural innovations and techniques
- Pest and disease management guidance
- Access to cooperative announcements
What they had
- Basic SMS-capable phones (not smartphones)
- An organized cooperative – Coopeumo, about 400 associates
- Geographic concentration in a rural area
- A stated need for better information
Coopeumo, the cooperative behind the program, is based in Peumo, a farming town in Chile's O'Higgins region. It marked its 41st year in 2010 and has about 400 associates. Coopeumo members grow maize, potatoes, avocados, oranges, plums and melons, so one set of messages could not fit every farm.
The technology to reach farmers existed (SMS), but most Coopeumo members did not know how to use it at the time. Furthermore, the farmers were hungry for agricultural information because they viewed it as critical to improving the productivity of their farms. And, finally, one of the primary benefits of smallholder cooperatives was the sharing of information. Datagro was among the first agricultural technology projects to connect these three elements.
The solution: SMS-based information delivery
DatAgro sent short, useful information by SMS. No navigation was required, and it worked on any feature phone at a time when smartphones were largely absent in the Chilean countryside.
Technical architecture
DatAgro ran on DataDyne's Mobile Information Platform (MIP): a web-based administration interface, message scheduling and subscriber management, and two-way SMS, so that farmers could send messages back as well as receive them. Messages were drawn from weather, agronomic and market-price sources. Entel PCS is named as a partner in the program's 2010 proposal documents. The Zoltner Consulting Group ran the Chilean implementation locally.
Writing every message by hand
Every message was written and checked by people. No software could judge which information a farmer needed on a given day, so an agronomist and the cooperative's leadership worked with the team to choose and draft each one. Bernarda Labarca, who co-coordinated the project, was part of that work from the start.
Each message required human judgment about what farmers actually needed to know on that particular day. Content curation was never a technical problem – it was a domain-expertise problem, and it demanded agricultural knowledge at every step.
Implementation: five phases
DatAgro moved from concept to sustained operation over roughly eighteen months. The first launch carried much of the early learning.
DataDyne and its local partner, ZCG, designed the system, met with Coopeumo to understand farmer information needs, and partnered with FIA for funding and agricultural expertise.
Recruited an initial 30 volunteers, designed message templates, secured the Entel PCS partnership, and identified content sources across FIA, market databases and weather services.
Messages went to 30 volunteers, whose feedback shaped the content. The launch relied on volunteers recruiting others, which did not reach the cooperative as a whole. Sign-ups moved to the cooperative's extension workers, who visit members every month.
Participation reached 238 members, and the archive held more than 500 selected messages. The earthquake of 27 February 2010 paused recruitment for several months.
A mid-term review and a Final Report documented how the program worked, drawing on conversations with members, technicians and cooperative leaders.
The results
Three figures describe the program's reach and output: how many members took part, how large the message archive grew, and how a sample of messages read when the mid-term review coded it.
How the messages read
The mid-term review coded a sample of 123 messages. Most read clearly: 73.2% (90 messages) had no drafting problem. The other 26.8% (33 messages) did. Here, a drafting problem meant the wording made a simple idea hard to take in, even when each word was simple. Technical vocabulary was a separate issue. Only 21.1% of the sample (26 messages) used terms that farmers found hard to follow, and 78.9% (97 messages) did not. Some messages also repeated advice that had already gone out. That was deliberate. A practice that returns across a season has more chances to stick, which is the point of a message service.
Subscriptions and sustainability
Subscriptions rose in the months after the cooperative's extension workers began raising DatAgro with members on their regular monthly visits. The workers already called on each member to share advice about planting, pests and harvest timing, so the service arrived in a familiar setting. Alongside that, the team and an agronomist built a message bank that was meant to stay with the cooperative as a resource.
Awards & international recognition
DatAgro drew national press coverage and international recognition in 2009–2010.
Press coverage
Between April 2009 and December 2010, 48 items about the program appeared in Chilean national and regional outlets, business and technology sites, radio, and Latin American and Peruvian media. Coverage clustered around the May and June 2009 award announcements, the May and June 2010 items, and the November 2010 eLAC awards.
See the press coverage, 2009–2010.
The budget
The Knight Foundation grant was $325,000 over 2008 to 2010. By July 2010, $235,571 had been spent, about 72% of the grant, on program staff, programming, evaluation, travel and marketing.
What the evaluations measured
The two evaluations listened closely to the people the program served. The mid-term review drew on testimony and informal conversations with members, technicians and cooperative leaders, and it coded a sample of 123 SMS messages to see how the content read. The Final Report, completed in August 2010, documents the program in full.
| Coded sample of 123 messages | Result |
|---|---|
| Messages in plain, everyday language | 78.9% (97 messages) |
| Messages with clear wording | 73.2% (90 messages) |
Two external contractors carried out the evaluations, commissioned by DataDyne.org, the platform developer.
Where the model went next
- Peru, 2010. The same platform was used for SaludMóvil, a text-message reinforcement pilot for health workers in the Andes. See the SaludMóvil case study.
Key learnings
Five principles carried out of DatAgro that shape how we approach rural technology today.
Rural populations don't reject technology out of inability – they reject irrelevant technology. When information addresses real needs through trusted channels, adoption follows.
Short messages and no navigation forced discipline about content quality. A short message leaves room for only the most valuable point.
You cannot automate the judgment of what a farmer needs to know today. It requires understanding local crops, seasons, markets, and decision-making.
Vendor subscriptions and project grants both ran out. Embedding the service in the cooperative's own operations is the lesson we would carry forward.
One coordinator for 238 members would need about 42 coordinators to reach 10,000. That is an estimate.
Project at a glance
| Metric | Result |
|---|---|
| Participants | 238 cooperative members |
| Geographic focus | VI Region (O'Higgins), Chile |
| Project duration | 2+ years (2008–2010) |
| Messages | More than 500 selected messages |
| Recognition | Two prizes at the N2Y4 Mobile Challenge final (2009); FRIDA/eLAC 2010 award; a 2009 finalist place |
| Technology | SMS via Mobile Information Platform (MIP) |
| Product | DataDyne.org (its EpiSurveyor software was renamed Magpi in 2013) |
| Partners | Zoltner Consulting Group (local partner), Coopeumo, FIA, Entel PCS |
| Funding | Knight Foundation grant of $325,000 |
Appendix: the content-creation work
The work behind DatAgro's simplicity was the same each time: find something useful, check it, and fit it into a short message.
Sourcing
- Check market prices and weather
- Scan research and advisory material
- Read farmer feedback
- Consult agricultural advisors on relevance
Selection & sending
- Choose the most useful candidate
- Write it within the SMS length limit
- Check it for accuracy
- Send at a set time and answer farmer queries
The technology was relatively simple; the implementation was labor-intensive. The limit on scale wasn't the platform – it was the agricultural expertise and daily coordination required to curate short message content. Message quality mattered more than quantity, and local expertise more than centralized systems.
Every message began with a decision. Someone had to choose which market price, weather warning or farming tip a member most needed that week, then say it in 140 characters without losing the point. The team and the cooperative's staff did that work by hand, one message at a time, and it was the hardest part of the program.
In practice, the program could grow only as fast as the people who could write and check the messages. When sign-ups moved to the cooperative's extension workers, who already visited members every month, the service reached the cooperative as a whole in a way that volunteer recruitment had not. Trusted people did the reaching. The platform carried the message.
Case study first published July 2026 and corrected October 2026 · Project period 2008–2010 · DataDyne.org, with the Zoltner Consulting Group as local partner