Monitoring, learning and evaluation

In complex systems, differences between expected and observed results are often some of the most useful information available. MLE connects monitoring to learning and learning to changed decisions.

Monitoring reads responseObserve what the system does after foments are released, including absence, feedback and surprise.

Three indicator typesCompliance, movement and learning answer different questions and should not be collapsed into one dashboard.

The proposition log is the learning historyRecord what was believed, what happened, what was learned and what changes next.

Use distinct indicators for distinct purposes.Compliance protects accountability; movement tests direction; learning shows whether the team’s read and actions are improving.

Grasp · Three audiences

The team, donor and landscape need different things.

A useful MLE architecture makes evidence travel between real-time decisions, public accountability and the institutional memory that must survive the project.

Three indicator families

Track activity, direction and learning.

Explore · Keep a proposition log

Make the learning chain visible.

For each foment, preserve the original belief and the evidence that changed it. A successor team should be able to see not only what was done, but why the intervention evolved.

1 · Proposition

If we do X, we believe Y may happen because of Z. What signals and risks will we watch?

2 · Response

What actually happened? What moved, failed to move, surprised us or affected people differently?

3 · Learning

What does the response reveal about the maps, feedback loops, power or practice configuration?

4 · Next decision

What will we pursue, modify, reinforce, abandon or investigate next – and why?

If maps change but decisions do not, learning is decorative rather than operational.

Evaluation approaches fit for complexity

Look backwards from change and alongside action.

Outcome harvesting

Start from observed changes, verify them and assess how the intervention plausibly contributed alongside testing progress against pre-specified outcomes.

Most Significant Change

Collect stories of consequential change and use structured deliberation to interpret why participants consider them significant.

Developmental evaluation

Embed evaluation close enough to the intervention to inform real-time sensemaking and adjustment alongside retrospective judgement.

Deep dive · Report for learning and accountability

Tell the causal story without pretending certainty.

What was released?

Describe the foment, the proposition, the target interaction space and any changes from the approved plan.

What did the system do?

Report intended and unintended responses, absence, dampening, ripple and cascade signals.

What was learned?

Explain what the response changed in the team’s maps, causal belief or understanding of heading.

What changes next?

State the next decision, its evidence and any required change control, budget or authorization.

Compliance remains necessary. A systems-oriented MLE architecture adds movement and learning so that each evidence stream does the job it is best suited to.