A practical path from uncertainty to useful evidence.
Four connected phases keep the question, context, methods, interpretation, and next decision visible throughout the work.
The RIL method
Four phases. One useful evidence cycle.
The sequence creates discipline without pretending the work is perfectly linear. Teams can revisit earlier steps as context changes, evidence appears, or the decision becomes clearer.
Step 01 of 10
Orient / Phase 01
Frame the decision
We begin with who needs to act, what choice they face, and what would become possible if the uncertainty were reduced.
Phase 01
Orient
Understand the decision, context, and evidence already in the room.
- Guiding question
- What decision must this evidence improve?
- Phase outcome
- A shared decision frame, a grounded view of context, and a map of what is already known.
- 01
Frame the decision
We begin with who needs to act, what choice they face, and what would become possible if the uncertainty were reduced.
- 02
Understand context
We listen to the people closest to the work and examine the incentives, constraints, routines, and power relationships shaping the setting.
- 03
Map what is known
We bring together existing data, research, practitioner knowledge, and assumptions so the new work starts from evidence rather than from scratch.
Phase 02
Design
Choose a proportionate method and prepare the work carefully.
- Guiding question
- What would be proportionate, credible, and feasible here?
- Phase outcome
- Clear questions, suitable methods, practical tools, defined roles, and safeguards ready for use.
- 04
Define questions and success
We turn the decision into answerable questions, clarify intended outcomes, and agree what credible and useful evidence should look like.
- 05
Choose proportionate methods
We select qualitative, quantitative, experimental, measurement, or synthesis methods because they fit the question, not because one method is our default.
- 06
Prepare people, tools, and safeguards
We build instruments, protocols, roles, quality checks, ethics, and data safeguards with the teams who will make the research happen.
Phase 03
Learn
Generate and interpret evidence without losing sight of reality.
- Guiding question
- What is happening, for whom, and why?
- Phase outcome
- Evidence interpreted with implementation reality, alternative explanations, and limitations intact.
- 07
Generate evidence carefully
We stay close to implementation, document deviations, protect data quality, and adapt field processes without quietly changing the research question.
- 08
Analyse and interpret together
We test patterns, alternatives, and limitations, then interpret findings with people who understand the setting and will use the result.
Phase 04
Act
Turn what was learned into choices, systems, and the next test.
- Guiding question
- What should change because of what we learned?
- Phase outcome
- Clear choices, useful outputs, and a next learning loop that teams can carry forward.
- 09
Translate evidence into choices
We organise outputs around the real decision: what to continue, change, stop, fund, communicate, measure, or investigate next.
- 10
Learn, adapt, and scale
We help teams build feedback loops, protect essential features, and carry uncertainty forward honestly as the work evolves.
Methods
Method follows question.
We combine methods when that produces a more credible and useful answer. No technique is treated as a badge of rigour on its own.
The strongest design is the one that can answer the question well, within the realities of the setting.
- 01Impact and process evaluation
- 02Field experiments and causal inference
- 03Qualitative interviews and observation
- 04Surveys and measurement frameworks
- 05Evidence and literature synthesis
- 06Theory of change and learning agendas
- 07Curriculum and assessment design
- 08Implementation and systems research
Quality and collaboration
Rigour should be visible in the work.
Credibility comes from explicit choices, honest limits, and regular interpretation with the people closest to the programme and the decision.
Four non-negotiables
- 01Clear research logic and decision criteria
- 02Proportionate ethics and data safeguards
- 03Documented quality checks and deviations
- 04Transparent interpretation and limitations
Three collaboration moments
Frame together
At the start
Align the decision, users, constraints, existing evidence, and definition of a useful output.
Interpret together
During the work
Use evidence reviews and field check-ins to test emerging meaning against implementation reality.
Choose together
Before the finish
Stress-test interpretations and shape recommendations, tools, and next steps with the people who will use them.
A proportionate starting point
Bring us the decision before the method.
We can help clarify the question, identify what is already known, and shape a sensible first step.