Create a Results Framework
Build a multi-level results chain showing how inputs lead to activities, outputs, outcomes, and impact.
On this page
- What You Need
- Provide the program design, activities, intended changes, population, timeframe, and any existing theory of change or indicators.
- Choose This Prompt When
- Use this to structure intended results; choose theory of change for deeper causal pathways or a MEL plan for data operations.
Build a results framework from the supplied program design. If activities or intended results are missing, ask for them before asserting a results chain. Show proposed links and assumptions as hypotheses until the program team validates them.
Provide a compact text chain from inputs and activities through outputs and outcomes to the longer-term goal, using only result levels that fit the intervention. For each link, explain the expected mechanism, key external assumption, and what evidence could test it. Add a small indicator table with linked result, definition, feasible source and method, frequency, responsible role, and disaggregation where useful. Mark unsupplied baselines, targets, partners, and policy alignments as unverified. Distinguish the program's contribution from changes it can prove it caused. Finish with missing links, risks, capacity needs, and decisions to validate. Do not build a full MEL plan unless requested.
Identify stakeholder roles in validating the chain and using it. If baseline data are absent, give a feasible plan for establishing them rather than a value.
Retain conflicting program claims and mark their implications for the chain unresolved.
Worked Example
Illustration only. The setting and figures are invented to show the pattern.
Scenario: A women's savings and enterprise program in two regions has a list of activities and a general aim, but no agreed chain from activities to change. The team wants a framework before drafting indicators.
What You Give the AI: The activities (savings group formation, business skills training, linking groups to a microfinance provider), the intended changes, the target population of rural women, a three-year timeframe, and a loose theory of change paragraph.
What the AI Might Return: A chain from inputs to impact. One link reads: activity "business skills training", output "women trained", outcome "women start or grow enterprises", with a note on why the link might hold, that training matched to local markets gives women usable skills. Assumptions and risks follow, such as "markets can absorb new products". An indicator crosswalk offers one indicator per result, a list says baselines must be set, and open decisions include who validates the chain with the women.
When to Choose This Prompt
- You need to set out intended results from inputs through activities to outcomes and impact.
- Someone must validate the chain before indicators or a logframe are drafted.
- You want assumptions, risks and open decisions listed next to each link.
When Not to Choose It
- You need to explain why and how change happens, with the full causal pathway. Create a Theory of Change
- You need data collection, schedule and responsibilities. Create a MEL Plan
- The result structure is agreed and needs to go into a logframe matrix. Generate a Logframe from an Existing Theory of Change
Checking the Output
- Take the activities from your own documents and see whether each one leads somewhere in the chain. Orphan activities and results with no activity both need an answer.
- At each link ask whether the result above could plausibly follow, given the activities and assumptions. A jump from training to income change with nothing in between is a gap.
- Output and outcome rows should be at the right level. Counts of things delivered belong in outputs, and changes in behavior or condition in outcomes.
- Indicator crosswalk entries should be kept few, and baselines marked as needing collection where you gave none.
- Assumptions should be specific to your setting. Generic lines such as "political stability" need replacing.
- The open decisions should name people, such as who validates the chain.
Common Problems and Fixes
- The chain is a list of your activities relabeled as outcomes. Fix: Each outcome should read as a change in people or conditions, and any that only restates an activity gets cut.
- Impact-level statements promise more than the program can influence. Fix: What is within your control sets the ceiling. Give impact wording to match it.
- Far too many indicators appear. Fix: The smallest set that tells you whether each result is happening is enough.
- Risks and assumptions are mixed up in one list. Fix: Two lists work: assumptions as conditions the chain relies on, risks as events that could break it.
Adapting It to Your Context
- Existing theory of change: Paste it and ask the framework to follow its wording, with differences listed.
- Multi-country program: Ask for a shared chain, then a note on where each country differs.
- Donor-prescribed levels: Paste the level names and ask for the chain to use them. Check the mapping yourself.
- Early design stage: Say that activities may change, and ask for the chain to separate fixed parts from options.