11,464 Smallholder Farmers Trained (114.6% of Target)
Under GIZ's Scaling Digital Agriculture Innovations through Start-ups (SAIS) programme, Zowasel deployed AgroServe to deliver GAP/BMP training, digital data capture, and Crop Centre infrastructure to 11,464 farmers across 13 Nigerian states.
11,464
Farmers Trained & Captured
114.6% of 10,000 contractual target
78
Cooperatives Reached
Organized grower clusters
13
States Covered
Across 5 geopolitical zones
43.0%
Female Participation
In-season verified gender inclusion
1
Crop Centre Upgraded
Ikom aggregation hub
GIZ SAIS-SUPPORTED AGROSERVE PROJECT | NIGERIA | 2022-2024 EVIDENCE WINDOW
Improve Income in African Agriculture and Food
Scaling seasonal agronomy, digital advisory, market access and finance pathways for smallholder farmers
Public review draft. Reported training reach exceeded the 10,000-farmer target. Income and productivity signals are promising, but inconsistent records, recall-based baselines, price inflation and the absence of a comparison group mean causal impact is not established.
“Reach is the strongest result; income is an evidence agenda, not yet a headline claim.”
Zowasel case-study evidence finding
| 11,464farmers reported trained114.6% of the 10,000 target | 78cooperatives reported reachedfinal impact report |
|---|---|
| 380farmers in 2024 follow-upthree states; three focus crops | EUR 49,998.80GIZ contract valuesigned service contract |
11,464
farmers reported trained
114.6% of the 10,000 target
78
cooperatives reported reached
final impact report
380
farmers in 2024 follow-up
three states; three focus crops
EUR 49,998.80
GIZ contract value
signed service contract
PROJECT PROFILE
1. Project at a glance#
- Project name
- Improve Income in African Agriculture and Food
- Public case-study title
- Scaling digital and last-mile agricultural services through Zowasel AgroServe
- Lead implementer
- Zowasel.com Limited
- Funding and contracting partner
- Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ) GmbH through Scaling Digital Agriculture Innovations through Start-ups (SAIS)
- Commissioning context
- SAIS is implemented by GIZ on behalf of Germany's Federal Ministry for Economic Cooperation and Development (BMZ)
- Community delivery actors
- Facilitators, agronomists, sustainability specialists, agents, trustees, farmer cooperatives, community leaders and participating smallholder farmers
- Target customers and beneficiaries
- Women and men smallholder farmers aged 25-60 producing food and cash crops; farmer cooperatives and clusters; buyers and financial-service actors using Zowasel's market and data infrastructure
- Country
- Nigeria
- States / regions
- Five reported regions and 13 states: Akwa Ibom, Benue, Cross River, Ekiti, Jigawa, Kaduna, Kano, Kwara, Nasarawa, Ondo, Oyo, Plateau and Taraba
- Department
- Value Chain Department / AgroServe delivery, working with Agronomy, Sustainability, Programmes, Marketplace and data teams
- Technology deployed
- Zowasel Marketplace, ACESS as a finance-enabling layer, Kobo/Traceability data capture, Crop Centre infrastructure, SMS and WhatsApp price information, and the Extension Advisory App. The *7199# Farmers Call Centre appears in 2024 continuity reporting, not as a verified 2023 contractual deliverable.
- Contract duration
- 23 December 2022-31 December 2023; final report contractually due 30 November 2023
- Outcome follow-up
- Impact assessment fieldwork in May 2024 and report dated 4 June 2024, outside the signed implementation period
- Budget
- EUR 49,998.80 fixed lump-sum GIZ service contract; the reviewed files do not establish Zowasel co-investment or a consolidated total project cost
- Program Manager
- To be confirmed by Zowasel management before public release
- Supervisor
- To be confirmed by Zowasel management before public release
- Last updated
- 29 August 2026
The signed contract is the controlling source for runtime and remuneration. The June 2024 impact assessment is treated as post-project outcome follow-up, not as an extension of the contractual delivery period.
Executive summary
Across rural Nigeria, smallholder farmers face a connected set of constraints: limited extension coverage, uneven access to improved inputs and mechanisation, weak farm records, constrained finance, fragmented markets and poor price visibility. GIZ's SAIS programme sought to scale proven digital innovations that could increase income in African agriculture and food systems. Under a EUR 49,998.80 service contract, Zowasel adapted its AgroServe model to combine field-based Good Agronomic Practice (GAP) and Best Management Practice (BMP) training with digital data capture, local facilitators, Crop Centres, price information, market linkage and alternative-finance pathways.
The contract ran from 23 December 2022 to 31 December 2023. Delivery reports state that 11,464 smallholder farmers were trained and captured across three seasonal phases in 13 states, exceeding the 10,000-farmer target by 14.6%. The final report also records 78 cooperatives, 51 agents and completion of the Ikom Crop Centre upgrade. In-season records provide the most detailed verified implementation snapshot: 5,027 farmers across 13 states, with 43.03% reported female.
Outcome evidence is useful but not causal. A 2023 early-harvest assessment used 1,189 farmer records and 16 focus group discussions on maize and rice. A May 2024 follow-up assessed 380 farmers in Nasarawa, Jigawa and Oyo on paddy rice, sorghum and wheat. Reports describe shifts toward higher harvest and price bands, 128 respondents selling through Zowasel, and 315 respondents assigned to positive earnings-growth bands. However, baseline estimates relied on recall, there was no comparison group, 2023 and 2024 prices were affected by severe inflation and currency volatility, bag weights were not standardised in the narrative, and several income, gender and geographic totals do not reconcile.
The project's strongest public result is therefore reach and delivery infrastructure, not a definitive income percentage. The case study finds high relevance, substantial effectiveness on training reach, partial effectiveness on wider service conversion, indeterminate efficiency beyond the contract value, promising but unverified impact, and plausible continuity through Crop Centres, Marketplace, ACESS and the Farmers Call Centre. Future reporting should use unique farmer IDs, matched seasonal panels, standard units, verified sales and costs, comparison methods, and explicit environmental and employment indicators.
Zowasel demonstrated that a distributed facilitator-and-platform model could reach farmers at national scale. The available evidence does not yet isolate how much of observed yield, price or earnings change was caused by the intervention. Public communications should say the project contributed to reported improvements and should not claim that it doubled farmer income.
CONTEXT AND DESIGN
2. Context#
Digital agriculture can lower the cost of delivering information, market and financial services, but a digital product alone does not resolve last-mile barriers. Many farmers operate in remote locations, use basic phones, have limited formal records and depend on community networks for trust and mobilisation. The project therefore combined technology with local facilitators, cooperatives, Crop Centres and seasonal field training.
The SAIS tender aimed to identify digital solutions with an end-user focus and a measurable income pathway. It offered up to EUR 50,000 per selected bid and required proof of impact, milestone reporting and evidence that the solution remained operational after full delivery. Zowasel's response connected agronomy, farmer data, price discovery, market access and finance rather than treating training as a standalone activity.
GIZ's current SAIS project page describes the programme's wider purpose and official SDGs. It does not publish a standalone result page for this specific 2022-2023 Zowasel contract. Project-level results in this report therefore come from the supplied contract and implementation records.
3. Theory of change#
Organise local delivery
Recruit and train agronomists, sustainability specialists, facilitators and agents linked to farmer groups and Crop Centres.
Build farmer capability
Deliver practical pre-season, in-season and post-season GAP/BMP modules adapted to priority food and cash crops.
Create usable data
Capture farmer, farm, input, mechanisation, finance, market and training information through digital tools.
Connect services
Use Crop Centres, Marketplace, price information and ACESS-related pathways to link farmers with inputs, buyers and suitable finance.
Support adoption
Provide continued field and digital advisory so farmers can apply improved practices through the crop cycle.
Improve farm economics
Contribute to better yields, lower avoidable losses, stronger price negotiation, more efficient production and higher net earnings.
Sustain the model
Keep the service active beyond the contract through local networks, repeat transactions and low-bandwidth digital channels.
| CRITICAL ASSUMPTIONS Farmers must be able to access suitable inputs and machinery, apply the advice, complete a full crop cycle, sell at transparent prices, and manage climate, security and liquidity shocks. Outcome measurement must also distinguish intervention effects from weather and price inflation. Several of these assumptions were only partly met. |
|---|
4. Objectives#
Implement a digital and field-based solution with a positive income pathway for smallholder farmers in Nigeria.
Provide GAP and agronomy services to 10,000 women and men smallholder farmers aged 25-60 across food and cash-crop value chains.
Improve productivity, sustainability and profitability through seasonal GAP/BMP training and better input, post-harvest and farm-management practices.
Upgrade selected remote Crop Centres to improve access to agronomy, inputs, financial inclusion and markets.
Onboard cooperatives and cluster farmers for more efficient training, distribution, data capture and transactions.
Measure reach and income-related effects and demonstrate that the solution continued operating after implementation milestones.
The official project-reporting target was 10,000 farmers. A separate internal operating sheet used a 15,000-farmer annual target and 640 cooperatives. This report uses the contractual 10,000 target for effectiveness and treats the larger sheet as an internal stretch plan.
DELIVERY MODEL
5. Our approach#
| Workstream | Delivery approach | Observed implication |
|---|---|---|
| Local facilitator network | Recruit community-based facilitators with training experience, farmer-group relationships and digital capability; train them through virtual and physical ToT sessions. | 127 applications were screened; 20 facilitators were initially selected and 27 were listed by Q3. |
| Seasonal learning | Deliver pre-season, in-season and post-season modules on crop production, farm business, sustainable practice and post-harvest management. | Reports describe three phases and multiple crop-specific modules rather than a single event. |
| Farmer mobilisation | Use cooperatives, community leaders, agents, town criers and Crop Centres to schedule training at farmer-accessible times and locations. | Extended reach but created scheduling, access, compensation and coordination challenges. |
| Digital data capture | Collect farmer profile, farm, input, mechanisation, finance, market and training data through Kobo and Zowasel tools. | Enabled detailed reach analysis but exposed survey length, consent, record-quality and reconciliation issues. |
| Price and market access | Share price information through SMS/WhatsApp and connect groups to buyers through Marketplace. | 128 of 380 follow-up respondents reported selling through Zowasel; most did not. |
| Finance pathway | Use ACESS-related farmer information to support access to improved inputs and mechanisation finance where qualified. | Reports describe an average USD 150 input/mechanisation loan, but borrower count, disbursement and repayment were not supplied. |
| Crop Centre access | Upgrade remote infrastructure for quality checks, aggregation and last-mile service access. | The financial offer planned five improved centres; the final report confirms the Ikom centre upgrade. |
| Continuity channels | Develop the Extension Advisory App and later the *7199# Farmers Call Centre for low-bandwidth advisory. | The call-centre channel appears in 2024 follow-up and should be reported as continuity, not a 2023 deliverable. |
6. Technology and service stack#
| Component | Project role | Evidence boundary |
|---|---|---|
| Zowasel Marketplace | Connect farmer groups with buyers and support price discovery, quality and sales pathways. | A third of the follow-up sample reported selling through it; transaction values and attributable price effects were not reconciled. |
| ACESS | Alternative credit evaluation layer using farm and transaction data to support finance eligibility. | A supporting layer, not the project itself. The reported average loan lacks a verified beneficiary count and repayment record. |
| Kobo / Zowasel data tools | Capture farmer profiles, farms, inputs, machinery, finance, market access and training responses. | 5,027 in-season data points were reported; uniqueness and consent audit were not independently reviewed. |
| SMS and WhatsApp | Distribute price information and support selling decisions. | Distribution is documented; use, comprehension and price effects were not measured with a comparison group. |
| Extension Advisory App | Provide facilitators and crop advisers with research-backed modules, maps, forms and weather-related tools. | Described in 2024 follow-up; user analytics were not supplied. |
| Crop Centres | Provide physical access to agronomy, inputs, aggregation, quality checks, finance and market services. | One completed upgrade is evidenced against five centres in the financial plan. |
| Farmers Call Centre / *7199# | Low-bandwidth USSD, IVR, SMS and messaging access to advisory and other services. | Presented as a post-project continuity response in 2024, not counted in contractual results. |
7. Implementation timeline#
| Period | Milestone | Evidence status |
|---|---|---|
| Oct-Nov 2022 | GIZ issued the Call for Solutions and assessed technical, impact, outreach, scalability and value-for-money criteria. | Tender terms and submission records. |
| 21-23 Dec 2022 | Contract executed; performance period commenced on 23 December. | Signed service contract. |
| Jan-Mar 2023 | Facilitator recruitment, screening, internal ToT, facilitator ToT, agent acquisition and first farmer training began. | March progress report; 127 applicants, 20 selected facilitators and 47 agents acquired at that stage. |
| Mar-Jun 2023 | Pre-season GAP/BMP training and farmer data capture expanded across operating regions. | Project briefing later reported 6,340 pre-season farmers. |
| Jul-Sep 2023 | In-season training and data collection reached 5,027 farmers; Ikom Crop Centre upgrade continued. | Detailed Q3 report with region and state tables. |
| Oct-Dec 2023 | Post-season, market and impact work continued; the service contract ended 31 December. | Impact and closure materials; final report was contractually due 30 November. |
| May-Jun 2024 | Post-project follow-up assessed 380 farmers in three states and three focus crops. | Impact assessment report dated 4 June 2024. |
| 2024 onward | Farmers Call Centre and later SAIS bootcamp activity show continued relationship and service development. | Continuity context; not counted as 2023 project output. |
PERFORMANCE AND EVIDENCE
8. Results#
| Results level | Defensible finding | Evidence status |
|---|---|---|
| Inputs | EUR 49,998.80 contract, four budgeted specialist roles, facilitator/agent network, training modules, digital platforms and Crop Centre infrastructure. | Contract, financial offer and implementation records. |
| Activities | Recruitment, ToT, three seasonal training phases, digital data capture, price information, cooperative mobilisation, field support and impact assessment. | Documented across progress and impact reports. |
| Reach output | 11,464 farmers reported trained and captured across 13 states, 14.6% above the 10,000 target. | Project-reported; uniqueness was not independently audited. |
| Institutional output | 78 cooperatives and 51 agents reported reached or onboarded. | Final impact report; role duration and activity levels are not reconciled. |
| Infrastructure output | Ikom Crop Centre reported upgraded for quality checks and aggregation. | One completed site is evidenced; the financial offer planned five improved centres. |
| Inclusion output | Final report states 40.2% female participation; in-season report states 43.03% female among 5,027 farmers. | Final gender counts do not reconcile with the total and stated percentages; correction required. |
| Market-use outcome | 128 of 380 follow-up respondents reported selling through Zowasel; 252 did not. | Self-reported; no verified transaction values or comparison group. |
| Productivity and earnings outcome | Reports show movement toward higher harvest, sale-price and earnings bands and illustrative maize/rice farm models. | Promising but observational; affected by recall, inflation, unit and reconciliation problems. |
| Long-term impact | No causal result is established for household income, food security, resilience, gender equality or environmental condition. | Requires longitudinal and comparative evaluation. |
9. Reach and inclusion#
| Reach measure | Reported figure | Interpretation |
|---|---|---|
| Contractual farmer target | 10,000 | Primary effectiveness denominator. |
| Final farmers trained and captured | 11,464 | Reported total across three phases; 114.6% of target. Unique-person audit not provided. |
| Pre-season | 6,340 | Project briefing figure; phase totals are not fully reconciled in the final report. |
| In-season | 5,027 | Detailed July-September dataset across 13 states. |
| In-season gender | 56.97% male; 43.03% female | Internally consistent percentages; underlying sex counts were not supplied in extracted text. |
| Final gender | 6,359 male; 5,069 female; 59.8% / 40.2% stated | Counts sum to 11,428, 36 below the reported total, and the percentages do not match the counts. |
| Cooperatives | 78 | Organisations reached; cooperative membership must not be added to direct farmer reach. |
| Agents | 51 | Onboarded support network; agents are not automatically employees or beneficiaries. |
| States | 13 | Akwa Ibom, Benue, Cross River, Ekiti, Jigawa, Kaduna, Kano, Kwara, Nasarawa, Ondo, Oyo, Plateau and Taraba. |
| 2024 outcome sample | 380 | Three states and three crops; not representative of every state and crop without weighting evidence. |
| REACH REPORTING RULE Report unique farmers, training attendances, cooperatives, agents, data records and digital users as separate units. The 11,464 figure should remain 'project-reported farmers trained and captured' until a deduplication and consent audit is complete. |
|---|
10. Budget and value for money#
| Financial item | Amount / design | Interpretation |
|---|---|---|
| GIZ service contract | EUR 49,998.80 | Fixed lump-sum remuneration for the contracted services. |
| Milestone 1 | 30% | Kick-off, solution presentation, implementation plan and agreed KPIs/data methods. |
| Milestone 2 | 30% | Evidence of 50% of impact reached. |
| Milestone 3 | 30% | Evidence of 100% of impact reached, full report and participation in an event. |
| Milestone 4 | 10% | Proof that the solution remained operational five months after milestone 3. |
| Budgeted specialist roles | 2 agronomists and 2 sustainability specialists | Four planned field-specialist positions; actual contracts, duration and payroll were not reconciled. |
| Crop Centre equipment | EUR 19,510.50 in the financial offer | Planned for five improved remote Crop Centres; final evidence confirms one Ikom upgrade. |
| Motorcycles | EUR 4,076.00 for four units in the financial offer | Planned field-mobility input; procurement and asset status were not confirmed in the final report. |
| Cost per reported trained farmer | Approximately EUR 4.36 | Contract value divided by 11,464 reported farmers; not a full economic cost and not an impact cost-effectiveness ratio. |
The contract value is not the same as total project cost. The reviewed evidence does not include an audited expenditure statement, Zowasel co-financing, farmer time, community contributions or the value of pre-existing technology and infrastructure.
11. Monitoring and evaluation results#
| M&E layer | Available evidence | Assessment and next requirement |
|---|---|---|
| Starting position | Baseline form reported Zowasel had worked with 9,617 farmers by December 2022, including 5,672 women. | Company-level pre-project reach, not a project baseline and not to be added to project reach. |
| Implementation monitoring | Kobo/Zowasel data on profile, farms, inputs, mechanisation, finance, market access and training; 5,027 in-season records. | Strong operational breadth; verify unique IDs, consent, completeness, enumerator quality and phase matching. |
| 2023 early-harvest assessment | 1,189 farmers, including 797 maize and 609 rice responses; 16 FGDs in northern states. | Crop selections overlap; recall-based and concentrated among early harvesters. |
| Illustrative maize model | Production 4,000 to 4,800 kg; cost NGN 630,000 to NGN 784,000; earnings fields conflict across files. | Yield signal aligns with the 20% target, but income percentage is withheld pending reconciliation. |
| Illustrative rice model | Production 4,300 to 6,500 kg; cost NGN 690,000 to NGN 800,000; endline earnings differ across sources. | Promising model, not a representative farmer estimate; spreadsheet contains conflicting revenue cells. |
| 2024 follow-up design | 380 respondents against a planned 372, with FGDs and a questionnaire in Nasarawa, Jigawa and Oyo. | Covers 3 of 13 states and paddy rice, sorghum and wheat; no documented weighting or comparison group. |
| Marketplace use | 128 respondents sold through Zowasel; 252 did not. | Measure verified transaction date, volume, price, quality, fee, counterfactual channel and payment timing. |
| Earnings bands | 239 respondents in NGN 10,000-199,000 growth band; 60 in NGN 200,000-399,000; 15 in NGN 400,000-599,000; 1 at NGN 1 million+. | Listed bands account for 315 of 380; status of 65 respondents is not explained. |
| Attribution | Reports acknowledge recall limits and weather, inflation and other PESTEL factors. | Use contribution language; adopt matched panels, comparison groups or credible quasi-experimental methods. |
| Planet indicators | Training included soil conservation, agroforestry, intercropping, organic practice and IPM. | No adoption, soil, water, biodiversity, input-efficiency or emissions outcome was measured. |
12. Employment opportunities and employment created#
The project created or supported roles across agronomy, sustainability, facilitation, data capture, agent services and cooperative mobilisation. The files document opportunities and assignments, but they do not contain a reconciled employment register showing which roles were net new, paid, full-time, retained or still active after the contract. Public reporting must therefore distinguish employment opportunities from verified jobs created.
| Employment category | Evidence found | Public-reporting decision |
|---|---|---|
| Agronomists | Two positions budgeted in the financial offer. | Report as budgeted specialist roles until contracts and workdays are verified. |
| Sustainability specialists | Two positions budgeted in the financial offer. | Report as budgeted specialist roles until contracts and workdays are verified. |
| Facilitators | 127 applicants; 20 initially selected; 27 facilitators listed in Q3. | Facilitator assignments evidenced. Verify unique people, contract status, paid days, sex, age and retention before calling them jobs created. |
| Agents | 47 acquired by March; 51 reported onboarded by the final report. | Support roles evidenced; employment form, compensation and active status are not established. |
| Trustees | 16 reported in a January-September briefing. | Community roles documented; compensation and duration not reconciled. |
| Data-capture agents | 2024 plan expected two per state in three sample states. | Short-term evaluation assignments; actual paid headcount not confirmed. |
| Farmer employment effects | Not measured. | Track family and hired labour days, wage cost, seasonal jobs and net changes by sex and age. |
| EMPLOYMENT HEADLINE FOR PUBLIC USE Four specialist positions were budgeted and a network of 27 facilitators and 51 agents was reported active or onboarded. The number of net new and sustained jobs created remains to be verified. |
|---|
PEOPLE, COMMUNITY AND PLANET
13. Impacts on people, communities and the planet#
| Dimension | Credible contribution pathway | What the evidence establishes |
|---|---|---|
| People - knowledge | Practical seasonal learning can improve farmer decisions on soil, seed, fertiliser, pests, records, harvest and marketing. | Training reach and content are evidenced; knowledge retention and adoption rates are not consistently measured. |
| People - income | Higher yields, lower losses, better timing, transparent prices and direct markets can improve gross margin. | Reports show positive signals but not a causally verified average income effect. |
| Women | Local facilitators, cooperatives and accessible schedules can reduce participation barriers. | Women were 43.03% of in-season reach and 40.2% in the final reported share; final totals need correction and empowerment outcomes were not measured. |
| Communities | Crop Centres, agents and cooperatives can create durable local access to advisory, aggregation, inputs, finance and buyers. | Thirteen-state coverage, 78 cooperatives and new activity in Taraba and Plateau are reported; community welfare and spillovers were not evaluated. |
| Digital inclusion | SMS, WhatsApp, advisory apps and USSD/IVR can reach farmers beyond smartphone-only services. | Channels were built or used; unique active users, repeat use and service outcomes were not supplied. |
| Planet - opportunity | Soil conservation, IPM, agroforestry, intercropping and better input use can improve resilience and reduce harm. | These practices were in curricula; adoption, soil condition, biodiversity, water and emissions were not measured. |
| Planet - risk | More fertiliser, agrochemicals and mechanisation can raise environmental burdens without safeguards. | The farm models show higher input spending after training; environmental trade-offs were not quantified. |
| IMPACT BOUNDARY This report uses 'contributed to', 'reported' and 'promising signal' for outcome claims. It does not state that the project caused a specific percentage increase in income, yield or environmental performance. |
|---|
14. Evaluation against OECD-DAC criteria#
Relevance: High. The project addressed clear last-mile constraints in advisory, records, inputs, markets and finance and aligned with SAIS's income and scaling objective.
Coherence: High. Facilitators, cooperatives, Crop Centres, Marketplace, data tools and finance pathways formed a plausible value-chain service model.
Effectiveness: Substantial for reach; partial overall. Reported training exceeded target, but one of five planned Crop Centre upgrades is evidenced and service-use outcomes were uneven.
Efficiency: Indeterminate. The contract value and reach imply a low partial cost per trained farmer, but audited expenditure, co-investment, service quality and impact cost-effectiveness are unavailable.
Impact: Promising but not verified. Yield, price and earnings signals are observational and affected by recall, inflation, sample coverage and reconciliation problems.
Sustainability: Promising / partly evidenced. The service continued through Crop Centres, Marketplace, ACESS and later low-bandwidth advisory, but active-use, revenue and outcome evidence is incomplete.
15. Failures, constraints and adaptations#
| Failure or constraint | What happened | Response / implication |
|---|---|---|
| Youth and facilitator recruitment | Agriculture was unattractive to some rural youth; some candidates declined agreements because of literacy, trust or contractual concerns. | Use plain-language agreements, local references, realistic role economics and retention support. |
| Survey burden and data reluctance | Some farmers considered questionnaires too long or were reluctant to share information; record-keeping was weak. | Use data minimisation, informed consent, shorter modules, progressive profiling and farmer-visible value. |
| Scheduling and remote access | Training calendars conflicted with farmer availability; roads and dispersed communities reduced turnout and increased cost. | Co-design schedules, cluster sessions, provide travel plans and combine physical with low-bandwidth follow-up. |
| Facilitator disengagement | Remote work and time-consuming data capture led some facilitators to leave abruptly. | Track workload, pay, supervision, retention and replacement; separate training from long survey administration where appropriate. |
| Elections and cash crisis | The 2023 elections, cash scarcity, failed transfers and high cash-access charges disrupted mobilisation and logistics. | Build payment redundancy, liquidity plans and flexible delivery windows into risk registers. |
| Weather and climate | Heavy rain, early cessation or insufficient rainfall affected activities and threatened yields. | Use weather-informed calendars, resilient varieties and climate-risk measures; avoid attributing seasonal results without controls. |
| Conflict and insecurity | Communal crises and insecurity delayed or prevented activity in some locations. | Apply security, safeguarding, conflict-sensitivity and stop-work protocols. |
| Cooperative compensation requests | Some cooperatives sought payment before data capture. | Clarify participation benefits and reimbursement policy; avoid incentives that distort consent or reported reach. |
| Infrastructure target gap | The financial offer planned five improved Crop Centres; final documentation confirms one Ikom upgrade. | Verify the status of the remaining four and publish the result against the approved milestone. |
| Marketplace conversion | 66.3% of the 2024 sample reported not selling through Zowasel. | Measure and address timing, crop, quality, aggregation, buyer demand, price, logistics and trust barriers. |
| Impact-measurement mismatch | Agriculture needs a full seasonal cycle, while milestone pressure sought rapid proof of income impact. | Design measurement before planting, follow matched farmers through sale and budget for outcome follow-up. |
| Data inconsistency | Gender, income, sample, geographic and spreadsheet totals conflict across files. | Create one controlled data dictionary, analysis dataset and signed results annex before publication. |
16. Learnings#
Last-mile digital agriculture works best as a hybrid system: technology, local people, trusted organisations and physical access points reinforce one another.
Reach is necessary but not sufficient. Training attendance should be followed by adoption, service use, crop performance, verified sales and net-income indicators.
A single unique farmer ID and season-by-season panel would resolve much of the current duplication and attribution problem.
Income measurement must use standard units, verified quantities and prices, all production costs, crop-specific timing and inflation-adjusted comparisons.
Marketplace access should be measured as a conversion funnel: eligible farmer, listed volume, quality-checked volume, matched volume, delivered volume, paid value and repeat sale.
ACESS should be measured separately through applications, approvals, disbursement, purpose, cost, repayment, default and farmer gross margin; a stated average loan is not enough.
Gender-responsive delivery needs targets beyond attendance, including adoption, control over revenue, access to finance, paid roles and leadership in cooperatives and facilitator networks.
Environmental content must be paired with adoption and outcome indicators; training on soil or climate-smart practice is not itself a planet impact.
Crop Centre scale-up should be milestone-based, with operating readiness, service quality, utilisation, revenue and community access verified per site.
Public storytelling is strongest when the limitations are explicit: the project can credibly claim scale, learning and a promising service model without overstating causal impact.
PARTNER MODEL AND CONTINUITY
17. Partner roles and public confirmation#
| Actor | Role supported by evidence | Attribution boundary |
|---|---|---|
| GIZ / SAIS | Issued and managed the call, contracted Zowasel, financed the service and set milestone and impact requirements. | GIZ's current SAIS page describes the wider programme; project-level outputs here are from supplied records. |
| BMZ | Commissioning context and primary public funder of SAIS at the time of the tender. | Do not imply BMZ directly managed field delivery. |
| Zowasel | Designed and implemented the AgroServe delivery model, training, data capture, Crop Centre, market and finance linkages and reporting. | Current corporate reach or later projects must not be backcast as 2023 results. |
| Facilitators and agents | Mobilised farmers, delivered or supported training, captured data and maintained local follow-up. | Their employment status and contribution volumes require reconciliation. |
| Cooperatives and community leaders | Supported trusted access, scheduling, clustering, participation and local administration. | Membership is not the same as direct reach or outcome. |
| Farmers | Participated in training, data collection, FGDs, production, sales and follow-up assessment. | Stories and images require informed public-use consent. |
| OFFICIAL PUBLIC CONFIRMATION GIZ's SAIS project page includes Nigeria and states that proven digital innovations should be disseminated in selected markets and generate income. The SAIS portfolio profile lists Zowasel as a Nigerian AgriFinTech participant and describes its marketplace. These sources confirm programme relationship and solution context, not the 11,464-farmer result. |
|---|
18. Sustainability and post-project continuity#
The reviewed records describe continued operation beyond the main 2023 milestones through the Ikom Crop Centre, Marketplace, ACESS, the Extension Advisory App and the *7199# Farmers Call Centre. Zowasel also appeared in a later SAIS bootcamp communication. These are credible signs of organisational and technology continuity, but they do not prove that every trained farmer remained active or that the 2023 service became financially self-sustaining.
| Scale condition | Decision needed | Evidence to collect |
|---|---|---|
| Active use | Define active farmer, facilitator, cooperative and digital user by service and season. | Unique monthly and seasonal users, repeat use and completion. |
| Outcome conversion | Follow farmers from training through adoption, harvest and sale. | Matched panels, verified inputs, yield, loss, quality, price, cost and net margin. |
| Crop Centre viability | Approve site-level minimum service and financial standards before additional upgrades. | Open days, transactions, users, service quality, revenue, cost and working capital. |
| Marketplace conversion | Align crops, quality, aggregation, buyer demand, logistics and payment timing. | Funnel conversion, delivered value, price comparison, payment days and repeat sales. |
| Responsible finance | Use ACESS with affordability, consent, insurance and repayment safeguards. | Approval, disbursement, effective cost, repayment, default and borrower margin. |
| Facilitator economics | Define compensation, workload, supervision and progression. | Paid days, earnings, retention, farmer load, quality and gender. |
| Low-bandwidth inclusion | Measure whether USSD/IVR/SMS reaches farmers excluded by smartphone and data costs. | Unique users, completion, language, comprehension, action and outcome. |
| Environmental integrity | Set a practice and outcome framework before making climate claims. | Adoption, input rate, soil, water, biodiversity, fuel and emissions indicators. |
19. Contribution to the Sustainable Development Goals#
| SDG | Contribution pathway | Evidence status |
|---|---|---|
| SDG 1 - No Poverty | Agronomy, market and finance services can contribute to higher and more stable smallholder earnings. | Income signals are promising but not causally verified. |
| SDG 2 - Zero Hunger | Better practices, inputs and post-harvest management can improve production and food availability. | Training and reported production signals are evidenced; food-security outcomes were not measured. |
| SDG 5 - Gender Equality | Women-focused reach and local service networks can improve access to knowledge, markets and finance. | Female participation is reported, but final totals conflict and agency or income control was not assessed. |
| SDG 8 - Decent Work and Economic Growth | Specialist, facilitator and agent roles can create rural work while improved farm economics support local growth. | Opportunities are evidenced; net new, decent and sustained jobs are not yet verified. |
| SDG 9 - Industry, Innovation and Infrastructure | Marketplace, ACESS, digital data tools, advisory channels and Crop Centres strengthen rural service infrastructure. | Technology and one completed centre upgrade are evidenced; wider adoption and viability need measurement. |
| SDG 13 - Climate Action | Soil conservation, agroforestry, IPM and climate advisory can support resilience and lower-impact production. | Curriculum content is evidenced; adoption and environmental outcomes were not measured. |
PUBLIC NARRATIVE AND EVIDENCE
20. Public visibility and media insights#
The in-depth public search found two strong official anchors. GIZ's current SAIS page identifies project number 2018.0121.6, includes Nigeria, names BMZ as client and describes a programme that scales digital agriculture innovations and links startups to capability and capital. The SAIS accelerator site profiles Zowasel as a Nigerian AgriFinTech participant, with 2020 as its year of participation and a marketplace connecting growers and buyers.
Zowasel's LinkedIn page confirms its current positioning around agriculture, alternative finance and data. A later LinkedIn post describes participation in a SAIS bootcamp, which demonstrates continuity of the relationship but postdates this contract. Zowasel's Medium publication was reviewed; no directly visible story specifically reporting the 2022-2023 Improve Income project was located. Social posts on Facebook and Instagram corroborate field training and participant stories, but they are Zowasel-authored communications rather than independent evaluation.
A 2025 FarmFit Insights Hub case study provides useful post-project context on ACESS, data validation and finance infrastructure. Because it was published after the GIZ contract, it is used to explain system continuity and learning, not to validate 2023 results. The public search did not locate a standalone GIZ article announcing Zowasel's implementation results under the exact 'Improve income in African agriculture and food' title.
| PARTICIPANT STORY - CONSENT AND EVIDENCE CHECK REQUIRED In a November 2023 public communication, Zowasel described Florence, a farmer from Wannune in Tarka LGA, Benue State, returning after a successful white-maize harvest following project training. The story is a valuable human illustration, but it should be published in this case study only after confirming identity spelling, informed consent, harvest data and permission for any image or quotation. |
|---|
GIZ: Scaling digital agriculture innovations through start-ups (SAIS) — official programme, project number, geography, runtime, partner and SDGs
SAIS Accelerator: Zowasel start-up profile — official portfolio relationship and solution description
Zowasel LinkedIn company page — company positioning and public communications reviewed
Zowasel LinkedIn: SAIS bootcamp — later programme-continuity communication; not 2023 project result
Zowasel Medium publication — publication reviewed; no specific visible 2022-2023 project report located
FarmFit Insights Hub: Zowasel and agricultural data for finance — 2025 ACESS, data-governance and finance-continuity context
Zowasel Facebook: Improve Income field update — Zowasel-authored field communication
Zowasel Facebook: Florence participant story — participant story; consent and outcome verification required
21. Claim reconciliation for public release#
| Claim area | Evidence issue | Case-study decision |
|---|---|---|
| Project title | Files use 'Improve income in African agriculture and food', 'Improve Income for Farmers...' and 'Income Improvement of Actors...'. | Use the signed contract title and explain AgroServe as Zowasel's delivery model. |
| Runtime | Contract ended 31 December 2023; impact assessment was completed in June 2024. | Separate implementation period from outcome follow-up. |
| Farmer target | Contract and reports use 10,000; an internal sheet uses 15,000. | Evaluate against the contractual 10,000 target and label 15,000 as an internal stretch plan. |
| Reach | 11,464 is reported but no deduplication audit was supplied. | Publish as project-reported trained and captured reach, not independently verified unique beneficiaries. |
| Gender | Counts sum to 11,428 and do not match the stated 59.8%/40.2% shares. | Retain the reported share with a visible reconciliation caveat until corrected. |
| Geography | Reports vary between nine, 10, 13 states and four or five regions; phase lists differ. | Use the final 13-state list and show that detailed in-season reach covered the same 13 states. |
| Crop Centres | Financial offer planned five; final report confirms Ikom. | Publish one confirmed upgrade and request status of the other four. |
| Maize income | Spreadsheet baseline earnings are NGN 201,600, while a query response uses NGN 255,500; percentage claims therefore conflict. | Withhold the percentage income increase; publish the model only as illustrative and unverified. |
| Rice income | Endline earnings appear as NGN 320,750 in one cell and NGN 305,975 in narrative; another revenue cell is inconsistent. | Withhold the percentage income increase pending workbook correction and source documentation. |
| 2024 earnings | Growth bands listed account for 315 of 380 respondents. | Report the bands and disclose the unexplained 65 respondents; do not calculate a mean. |
| Price change | 2023-2024 Nigeria inflation and currency volatility affected nominal prices. | Do not attribute nominal price-band change to training without inflation-adjusted comparison. |
| Employment | Roles and networks are documented, but net new paid jobs and duration are not. | Report employment opportunities; verify jobs created before using a jobs headline. |
| Planet impact | Sustainable practices were taught but not measured in fields. | Report training content and contribution pathway, not environmental impact. |
| ACESS / USSD | Supporting technology evolved during and after the project. | Distinguish 2023 project use from later system-continuity evidence. |
22. Priority actions before publication and scale-up#
Confirm the official public title, Program Manager, Supervisor, department wording and GIZ-approved partner attribution.
Deduplicate the 11,464 farmer records and reconcile phase, state, sex and age totals to a signed reach annex.
Correct the maize and rice baseline/endline workbooks, define every unit and provide the farmer-level source and calculation method.
Reanalyse nominal sales and earnings using comparable dates, standard bag weights, complete costs and appropriate inflation adjustment.
Explain the 65 follow-up respondents not represented in published earnings-growth bands and verify whether zero or negative changes occurred.
Verify the status of all five planned Crop Centre upgrades, four planned motorcycles and the four budgeted specialist roles.
Build an employment register with unique worker ID, role, contract type, start/end dates, paid days, earnings, sex, age and retention.
Create a Marketplace conversion and transaction dataset and an ACESS loan dataset with consent, disbursement, pricing, repayment and outcome fields.
Commission a matched two-season outcome study with a comparison strategy and transparent attrition, missing-data and sensitivity analysis.
Add adoption and planet indicators covering soil management, input rates, water, biodiversity, fuel, emissions and climate resilience.
Confirm informed consent and image/testimonial permissions for Florence and all other public participant stories.
Obtain Zowasel and GIZ communications, legal, data-protection and finance approval before releasing the public PDF.
Management questions to close
| Question | Why it matters |
|---|---|
| Who should be named publicly as Program Manager and Supervisor? | The governance fields cannot be inferred from contract signatories or current job titles. |
| Was 11,464 a deduplicated unique-farmer total, and what explains the 36-person gender gap? | This determines the public reach and gender headline. |
| What is the approved final count by phase, state, sex and age? | Reports use different geographic and phase summaries. |
| Were all four budgeted specialist positions filled, and which facilitators and agents were paid roles? | This determines employment created rather than employment opportunities. |
| Were five Crop Centres upgraded, or only Ikom? What happened to the remaining budgeted locations? | Infrastructure effectiveness must be reported against the financial plan. |
| Which maize and rice income figures are correct, and can the farmer-level dataset be reproduced? | Public income claims are currently inconsistent. |
| What happened to the 65 farmers outside the listed 2024 earnings-growth bands? | The follow-up result is incomplete without zero, negative and missing outcomes. |
| How many farmers received ACESS-supported finance, at what total value, cost and repayment performance? | The stated average USD 150 loan is insufficient for a finance result. |
| What Marketplace volumes, prices and payment terms were verified for the 128 sellers? | This is required to substantiate a market-access and income pathway. |
| Did GIZ formally accept every milestone and the June 2024 follow-up? | Partner acceptance and closure status should be confirmed before public release. |
| May Florence's name, story, image and outcome be published, and is there a consent record? | Public beneficiary storytelling requires explicit, informed permission. |
Evidence base and source register#
Evidence hierarchy used for this case study: signed agreements and approved budgets; dated project M&E/final reports; implementation records; then public communications. Where figures conflict, the most project-specific and most recent verified source is used and the conflict is disclosed.
Internal project records reviewed
81292385 service contract — legal parties, title, runtime, contract value, milestones, reporting and payment structure.
81292385 / Tender 81290635 Terms of Reference — SAIS context, tender objective, EUR 50,000 ceiling, impact criteria and required milestones.
81292385 financial offer — budgeted specialist roles, Crop Centre equipment, motorcycles, training and certification activities.
Technical bid submission form — proposal framework and evaluation fields; extracted text did not expose completed form-field detail.
Baseline for Improve income contract — company-level pre-project reach and Zowasel's stated income-measurement approach.
Zowasel AgroServe Progress Report, 31 March 2023 — facilitator applications, selection, agent acquisition, ToT and initial activities.
Zowasel GAP 2023 Targets workbook — internal 15,000-farmer stretch plan, agent and cooperative planning.
Project Briefing January-September 2023 — objectives, pre-season reach, agents, trustees, cooperatives, challenges and next steps.
Third Quarter Progress Report, July-September 2023 — 5,027 in-season farmer records, states, gender, age, modules and implementation detail.
GIZ GAP Project Impact Assessment 2023 — early-harvest sample, FGDs, maize and rice models, Marketplace and ACESS narratives.
Response to Additional Queries on Impact Report — income-calculation questions, cost and earnings ratio response and recall limitations.
Zowasel Baseline and Endline Project Analysis 2023 workbook — illustrative maize and rice cost, production, revenue and earnings calculations.
Zowasel GAP Impact Assessment Report, 4 June 2024 — final reach, cooperatives, agents, Crop Centre, 380-person follow-up, findings, challenges and recommendations.
Top-level GIZ GAP Project Analysis of Impact Assessment 2023 — condensed outcomes, methodology, sample statements and contribution caveats.
Implementation, training, facilitator and media folders — training calendars, modules, contracts, field evidence and operational documentation reviewed at folder level.
Restricted Google Drive source folder supplied by Zowasel — master evidence repository; private links are intentionally omitted from the public report.
Public sources and media reviewed
GIZ: Scaling digital agriculture innovations through start-ups (SAIS) — official programme, project number, geography, runtime, partner and SDGs
SAIS Accelerator: Zowasel start-up profile — official portfolio relationship and solution description
SAIS Accelerator programme site — programme context and portfolio information
VC4A: GIZ SAIS profile — programme implementation and accelerator context
Zowasel LinkedIn company page — company positioning and public communications reviewed
Zowasel LinkedIn: SAIS bootcamp — later programme-continuity communication; not 2023 project result
Zowasel Medium publication — publication reviewed; no specific visible 2022-2023 project report located
FarmFit Insights Hub: Zowasel and agricultural data for finance — 2025 ACESS, data-governance and finance-continuity context
Zowasel: What we do — current corporate and service-stack context; not backcast as project result
Zowasel contact page — public contact information
Zowasel Facebook: Improve Income field update — Zowasel-authored field communication
Zowasel Facebook: Florence participant story — participant story; consent and outcome verification required
Zowasel Instagram: Ologogi farmer training — Zowasel-authored field activity communication
Zowasel Instagram: 2023 precision-agriculture update — Zowasel-authored programme update
Further project information#
| Contact | Public information | ||
|---|---|---|---|
| Zowasel | sales@zowasel.com | marketplace@zowasel.com | https://www.zowasel.com/ |
| Zowasel Nigeria | +234 201 342 4403 | WhatsApp +234 702 506 5664 | |
| GIZ SAIS | Michel Bernhardt via the official GIZ SAIS project page: https://www.giz.de/en/projects/scaling-digital-agriculture-innovations-through-start-ups-sais | ||
| Project geography | Nigeria: 13 reported states across five regions |
Zowasel Nigeria
+234 201 342 4403 | WhatsApp +234 702 506 5664
GIZ SAIS
Michel Bernhardt via the official GIZ SAIS project page: https://www.giz.de/en/projects/scaling-digital-agriculture-innovations-through-start-ups-sais
Project geography
Nigeria: 13 reported states across five regions
Publication approval checklist
| Control | Approval required before public release |
|---|---|
| Governance | Confirm official title, manager, supervisor, department, runtime and closure wording. |
| Partner attribution | Obtain GIZ and Zowasel approval for programme, funding and implementation descriptions. |
| Reach and inclusion | Approve deduplicated total and reconciled phase, state, sex and age tables. |
| Finance | Approve contract value, expenditure treatment, co-investment wording and infrastructure-delivery status. |
| Outcomes | Approve corrected income, yield, price, Marketplace and ACESS analysis and retain attribution caveats. |
| Employment | Approve verified net new jobs, paid assignments, duration, earnings and disaggregation. |
| Data protection | Confirm lawful basis, consent, data minimisation, retention, rights and removal of personal data. |
| Stories and media | Verify Florence and other participant consent, spelling, quotations, images and supporting outcome evidence. |
| Final publication | Approve website PDF, download form, privacy notice, follow-up consent and named release owner. |
| PUBLICATION CONTROL Do not publish a 'doubled income' claim. Before public release, reconcile reach and gender totals, correct the farm-economics workbooks, explain missing follow-up outcomes, verify employment and infrastructure delivery, confirm partner attribution, and secure participant consent. |
|---|
Suggested website download gate: first and last name, organisation, job title, work email, country, sector, reason for download and required acknowledgement of the privacy notice. Use a separate optional checkbox for follow-up or marketing contact, state purpose and retention period, and provide an easy opt-out route.
