What is a learning agenda?
A learning agenda is a plan for finding out what you do not yet know. It lists a few questions that matter to your work, says how you will gather evidence on each, and says when the team will meet to discuss the answers. It turns “we should learn from this” into dates and owners.
Most NGOs already collect data. The gap is the step from data to a decision. A learning agenda closes that gap on purpose.
How do you pick good learning questions?
A good question has three features. It is open (not yes or no), it is tied to a decision someone will make, and it can be answered with evidence you can get.
Weak: “Is our training good?” Better: “Which parts of our farmer training lead to changed practice after six months, and for which groups?”
Ask these prompts in a workshop with program, field and finance staff:
- Where do we repeatedly argue about what works?
- Which decision next year would be easier with better evidence?
- Where are we unsure that a new approach beats the old one?
- Which groups do we reach least, and why?
Write down ten ideas, then vote down to three to five.
What steps build the agenda?
- Choose the questions. Use the workshop above and the strategy of your organization.
- Link each question to a decision. Write who decides, what the options are and by when.
- Choose the evidence. List what you already have, such as indicator data, feedback records and evaluation reports, and what you must collect.
- Set review moments. Pick dates for learning meetings. Tie them to existing events, such as quarterly reviews, so they do not add extra work.
- Assign owners. One person owns each question and reports back.
- Record the result. Note what you learned, what you decided and what changed.
What can the agenda look like?
A single table is enough.
| Learning question | Decision it informs | Evidence | When | Owner |
|---|---|---|---|---|
| Which outreach channels reach women farmers? | Where to spend next year’s outreach budget | Attendance by sex, feedback records, group discussions | Quarterly review | Program manager |
| Why do some water points fail within a year? | Whether to change the maintenance model | Monitoring visits, repair records | Mid-year | Technical lead |
How do you make learning happen in practice?
Learning fails when it depends on goodwill. Build it into routines:
- Short, regular sessions. A one-hour review each quarter beats a long annual workshop.
- Mixed groups. Invite field staff, managers and, where possible, partners and community members. Different views reveal different patterns.
- Honest discussion. Include what did not work. Use a simple format, such as what we planned, what happened, why, and what we do next. Many teams run this as an After Action Review.
- A decision log. Record each decision that came from learning, with a date.
- Sharing. Send a one-page summary to the people who need it, not only to the donor.
The MEAL plan should name the learning meetings, so the two documents stay aligned.
How do you know it is working?
Ask three questions each year:
- Did we answer most of our questions?
- Did any decision change because of what we learned?
- Do staff say they use the findings?
If the answers are no, shrink the agenda or move the meetings. Remove answered questions and add new ones. A living agenda of three questions beats a thick one nobody reads.
What does a learning cycle look like over one year?
A simple yearly rhythm works for many teams:
- January. Agree the questions and owners; link them to the MEAL plan.
- Each quarter. Hold a one-hour review. Look at the evidence collected so far and note early signals.
- Mid-year. Pause and check which questions are still useful. Drop or reword weak ones.
- Year end. Write a one-page summary of what you learned and what changed. Share it with staff and partners.
Our guide to lessons learned that get used goes deeper on turning findings into action. Keep the summary short and plain, and name who will act on each finding. If a finding has no owner, nothing will happen.
Where EMPOBase fits
EMPOBase has a Risk & Learning area and an After Action Review for closing a project, so lessons can sit next to the project record. Monitoring holds indicator results, and Complaints & Feedback (CFM) holds feedback you can use as evidence. EMPOBase does not manage a learning agenda as a separate object, so keep the questions and owners in a shared document or in a project record. Field data counts records, not unique people, which affects what some questions can answer.