If you've ever responded to a USAID, EU or UN call for proposals, you've come across these six words without always knowing exactly what they mean: relevance, coherence, effectiveness, efficiency, impact, sustainability. These are the OECD's six evaluation criteria, the reference framework used by nearly every international donor to judge a development programme. Understanding them isn't an academic exercise. It's what determines whether your evaluation report is judged serious or superficial.
Where does this framework come from?
The OECD's Development Assistance Committee defined these criteria to give a common language to the evaluation of international aid. The current version, revised in 2019, has six criteria. Coherence was added to the five historical criteria to better capture how an intervention fits with other actions in the same context.
1. Relevance: are we doing the right thing?
Relevance asks whether the intervention matches real needs. A programme can be perfectly executed and still be irrelevant if it addresses a need nobody expressed, or was designed without consulting the people concerned. In practice, an evaluator will check whether the programme's objectives match beneficiary priorities, national policies in the country of intervention, and donor priorities.
2. Coherence: does it fit with everything else?
This most recent of the six criteria looks at both internal coherence (do the programme's different components reinforce each other, or contradict one another?) and external coherence (does this intervention complement or duplicate what other actors are doing in the same area?). Two NGOs distributing the same health kit in the same village without coordinating: that's a typical external coherence failure.
3. Effectiveness: are objectives being achieved?
Effectiveness measures the extent to which set objectives were actually achieved. This is where the logframe and SMART indicators come in directly: without measurable objectives defined upfront, judging effectiveness becomes a matter of impression rather than evidence.
4. Efficiency: at what cost?
Efficiency compares results achieved to resources mobilised: time, money, staff. A programme can be effective (it achieves its objectives) without being efficient (it gets there spending three times the planned budget, or twice the announced time). This is the criterion donors scrutinise most closely during financial audits.
5. Impact: what difference does it really make?
Impact looks beyond the programme's immediate results, toward its larger-scale effects, including unintended effects, positive or negative. This is the hardest criterion to measure, because it often requires isolating the programme's effect from other factors evolving in parallel. This is precisely the role of impact evaluation methods like randomised controlled trials or difference-in-differences.
6. Sustainability: will the benefits last?
Sustainability asks whether the programme's benefits will continue after funding ends. A well built but never maintained because nobody locally was trained to maintain it: that's a sustainability failure, even if the programme was effective and efficient at the time of implementation.
How to actually use this
In an evaluation report, the six criteria aren't six cosmetic boxes to tick. They're six analytical angles that should structure data collection itself, from the design of the monitoring and evaluation plan onward. A donor reading a report where sustainability or coherence are absent or covered in a single sentence understands immediately that the evaluation was rushed.
The six criteria don't replace a well-built logframe. They exist to verify, after the fact, that the logframe actually delivered on its promises.