For Humans

Scoping the Analysis

Before you start analyzing, define what you are analyzing and why. Be specific about the product (washed Arabica in Rwanda is a different analysis than Robusta in Vietnam), the geographic boundaries (national, regional, a specific supply corridor), the time period, and which stages of the chain you will cover. Most development-focused VCAs focus on farm to export.

Know who commissioned the analysis and why. A donor designing a program needs different outputs than a government writing a national coffee strategy or a private company evaluating a sourcing decision. The purpose shapes everything.

Before You Go

Understand the context before you start mapping the value chain. Read the political history. Understand the land tenure system. Know the major ethnic and regional dynamics. Learn what happened to the coffee sector during structural adjustment, during conflict (if applicable), during commodity booms and busts. Rwanda's coffee sector cannot be understood without the genocide and the government's top-down reconstruction. Ethiopia's cannot be understood without the ECX. A value chain analysis that ignores context will produce technically correct but practically useless recommendations.

Identify the right mix of stakeholders. Your interview list should not be limited to the obvious actors like farmers and exporters. Input suppliers know what farmers are buying and what they can afford. Transporters know the real costs of moving coffee from farm to mill to port. Government officials know the regulatory environment and where the political constraints are. NGOs and donor agencies know what interventions have been tried and which ones failed. Extension agents know what actually happens on the ground, as opposed to what headquarters thinks is happening. Cast a wide net.

Talk through hypotheses with experts early. Do not wait until the analysis is done to get feedback. Share your preliminary map, your early price data, and your emerging hypotheses with people who know the sector: country experts, industry veterans, experienced consultants. They will tell you what you are missing, where your numbers look wrong, and which assumptions are unrealistic. An hour of expert feedback early in the process can save weeks of wasted analysis.

In the Field

Get diverse opinions. Do not rely on a single perspective. A cooperative manager will tell you the cooperative channel works well. A private trader will tell you cooperatives are inefficient. A government official will tell you the policy environment is supportive. A farmer may tell you something different from all three. Triangulate. When three independent sources give you the same story, you are probably close to the truth. When they diverge, you have found something interesting worth investigating further.

Corroborate numbers with different sources. This is the single most important data quality practice. If two independent sources (say, a cooperative manager and an exporter) give you the same farm-gate price for cherry in a given region, you can use that number with confidence. If they disagree by 50%, you need to investigate further. In VCA work, a number that comes from only one source should be treated as an estimate, not a fact. Label it accordingly.

Make it easy for people to give you data. People in the coffee industry are busy. If you want data (production figures, price records, cost structures), make it easy to share. Have a simple template ready. Offer to sign an NDA if they are concerned about confidentiality. Do not show up to a meeting with a 40-question survey and expect someone to fill it out on the spot. Short, focused asks get better responses than comprehensive but burdensome ones.

Go both downstream and upstream. There are two ways to trace a value chain. Downstream: start at the farm and follow the product forward through processing, trading, and export. Upstream: start at the buyer (an importer, a roaster, a retailer) and trace backward to the source. Both approaches have value. Going downstream gives you the farmer's perspective and reveals the full sequence of transformations. Going upstream gives you the buyer's perspective and reveals what the market actually demands. The best analyses do both.

Back at the Desk

Cross-check official statistics. National production figures, area under cultivation, number of registered farmers: these numbers come from government agencies that may not have the budget, capacity, or incentive to maintain accurate records. Cross-check with industry sources: exporter associations, international organizations (ICO, World Bank), NGO reports, and academic studies. When official numbers and industry numbers diverge significantly, note the discrepancy and explain which source you trust more and why.

Segment, don't generalize. The "average farmer" is a statistical fiction. In any given country there are subsistence farmers growing coffee on tiny plots alongside food crops, semi-commercial farmers investing in quality improvements, and commercial operations running hundreds of hectares. Their cost structures, quality levels, market access, and risk profiles are fundamentally different. Your analysis and your recommendations should reflect this.

Reporting

Structure your report for the audience.

For donors and development agencies, lead with the problem statement and proposed interventions. Put methodology at the end. Emphasize impact potential and beneficiary numbers.

For government officials, lead with national-level implications (employment, export revenue, rural development). Connect recommendations to existing policy frameworks. Be realistic about institutional capacity.

For private sector, lead with the business case. Focus on sourcing risks, quality opportunities, and competitive positioning.

For all audiences: keep the main report concise (20-30 pages), put detailed data and interview notes in appendices, use visual summaries (the map, the waterfall, the benchmark chart, the matrix), state your assumptions explicitly, and acknowledge limitations and data gaps.