If your organization implements a project with the support of the EU or UN agencies, external evaluation is not a formality, but a mandatory element of the project cycle with clear rules of the game. The Six OECD DAC Criteria adopted in 2019 have become the universal language used by all major donors and evaluators today. UNEG norms, AEA principles, BOND tools and ACAPS methodologies are added to them – each framework solves its own task, but together they form a single architecture of qualitative assessment. This article breaks down all the key standards, donor requirements, and reporting structures so you know what to expect from the evaluator and what the donor will expect from you.
Six OECD DAC criteria — the foundation of any assessment
In December 2019, the OECD Development Assistance Committee (DAC) approved updated evaluation criteria, replacing the classic five of 1991. A fundamental change was the addition of a new criterion — coherence (coherence). Six criteria form a normative framework for determining the value and significance of any development intervention.
Relevance (“Does the intervention do the right thing?”) measures how well the project’s goals and design are aligned with the needs of beneficiaries, national priorities, global goals and partners’ policies. Key point: relevance is assessed not only at the start — the criterion checks whether the project remains relevant under changing circumstances. Coherence (“How well does the intervention fit in?”) is a new criterion that analyzes the compatibility of the project with other interventions. Internal coherence checks whether the project does not contradict other programs of the same organization and international norms. External — whether it complements the efforts of other actors or does not duplicate them.
Effectiveness (“Does the intervention achieve its goals?”) measures the extent to which intended outcomes are achieved, including differentiated outcomes for different groups—a requirement that the updated version added in response to the “leave no one behind” agenda. Efficiency of resource use (efficiency – “How rationally are resources used?”) assesses whether results are achieved economically and on time, comparing costs with realistic alternatives.
Impact (“What difference does the intervention make?”) examines long-term, higher-level effects—positive and negative, planned and unintended, including transformative changes in systems and norms. Finally, sustainability (“Will the benefits last?”) examines whether the results will be sustained after the end of the funding, by analyzing the financial, institutional, social and environmental preconditions for this.
Two principles of using the OECD DAC criteria deserve special attention. First, the criteria should be applied thoughtfully, not mechanically — they should be contextualized for a specific assessment. Second, not all criteria have the same weight in each evaluation: priorities are determined by the purpose of the evaluation, the needs of stakeholders, and available resources.
DAC Quality Standards and Assessment Report Structure
DAC Quality Standards for Development Evaluation (2010) articulate five overarching principles: transparency and independence, integrity and respect for diversity, partnership and coordination, capacity building, and quality control at every stage. The classic triad — independence, credibility, usefulness — remains the basis of any quality assessment.
The structure of the DAC assessment report covers ten blocks. The report begins with the rationale, purpose, and goals of the evaluation, moving on to its scope—including intervention logic, theory of change, criteria, and evaluation questions. Next, the context (developmental, institutional, sociopolitical) is presented, followed by a detailed description of the methodology, explaining the methods, their validity, limitations, and sampling. A separate block is dedicated to information sources that require cross-validation. Sections on evaluator independence, ethical standards, and quality assurance are mandatory.
The evaluation results are drawn up with a clear distinction of four levels: findings → conclusions → recommendations → lessons learned. Each next level logically follows from the previous one. The report concludes with an executive summary that includes key findings, recommendations, and lessons.
UNEG: 14 norms for assessment in the UN system
The norms and standards of UNEG (United Nations Evaluation Group), adopted in the revised edition in April 2016, are a mandatory framework for all evaluations in the UN system and are widely used by the global evaluation community. In February 2025, a new Norm 11 on environmental and social impact was added.
UNEG defines ten general norms that must be followed in any assessment. These include utility (clear intent to use results), credibility (underpinned by independence and methodological rigor), independence (behavioral and organizational), impartiality (no conflict of interest), ethics (do no harm, informed consent, confidentiality), transparency (public access to results), integration of human rights and gender equality, professionalism and support of national evaluation capacities. Four institutional norms relate to the organizational environment: the requirement of an evaluation policy, an independent evaluation function with an adequate budget (from 0.5% to 3.0% of organizational costs), and mandatory use of results with management response.
The UNEG standard 4.9 defines the structure of the evaluation report in six blocks: what was evaluated and why; how the assessment was designed and conducted; what was discovered and on what evidence base; які висновки зроблено; що рекомендовано; які уроки засвоєно. Recommendations should be based on facts, not opinions, and designed with likely perpetrators in mind.
AEA and Joint Committee Standards: Ethics and Assessment Quality
The American Appraisal Association (AEA) offers two complementary documents. The Guidelines for Evaluators (updated in 2018) govern the ethical conduct of the evaluator, while the Standards for Program Evaluation of The Joint Committee (JCSEE, 3rd ed., 2011) define requirements for the quality of the evaluation itself.
AEA’s five guiding principles form the ethical framework. Systematic research requires a methodical, evidence-based approach with a transparent description of methods and limitations. Competence requires ensuring that the team is appropriately qualified, including cultural competence. Integrity involves disclosing conflicts of interest, transparent reporting, and refraining from evaluation if it could lead to misleading conclusions. Respect for persons encompasses informed consent, confidentiality and minimizing risks to participants. Common good and justice—a new focus for 2018—requires the evaluator to consider power imbalances and not exacerbate historical inequality.
The JCSEE standards contain 30 standards in five categories: usefulness (8 standards), feasibility (4), correctness (7), accuracy (8), and assessment accountability (3). The category of accountability added in the third edition requires documentation of the evaluation process and conducting a meta-evaluation — an assessment of the quality of the evaluation itself.
BOND: five evidence principles for NGOs
Bond – a network of over 400 UK international development organisations – has created a practical framework specifically tailored to the realities of NGOs with limited budgets and long causal chains. Bond’s five principles of evidence allow you to assess the quality of any evaluation or research report.
The Voice and Inclusion principle requires evidence to reflect the perspectives of beneficiaries, particularly marginalized groups, disaggregated by key demographic characteristics. “Appropriateness” checks whether the methodology is suitable for specific issues, context and type of change. Triangulation requires cross-checking from multiple sources, methods, and perspectives. “Contribution” (Contribution) – the central principle – asks whether a convincing causal argument is made about the effect of the intervention on change, not requiring strict attribution, but only plausible justification. “Transparency” (Transparency) obliges to openly describe methods, limitations and assumptions.
Bond does not impose specific methods, but provides a set of practical tools: an interactive checklist for evaluating the quality of evidence, a guide for choosing appropriate evaluation methods (from RCT to process tracing), a Health Check Tool for self-assessment of organizational capacity, and Impact Builder — an online database of proven indicators and data collection tools developed by more than a hundred NGOs.
ACAPS: Analytics for Humanitarian Solutions
ACAPS (Assessment Capacities Project) works in a different plane – humanitarian analysis and assessment of needs, but its approaches are valuable for any NGO working in crisis contexts. Founded in 2009 in Geneva, ACAPS monitors the situation in 150 countries and provides free analysis for NGOs, UN agencies and donors.
Three key principles of ACAPS are worth noting. “Make sense, not data” (Make sense, not data) – an emphasis on interpretation, not on the accumulation of information. Better to be roughly right than exactly wrong – in a crisis, a timely analysis with caveats is more useful than a perfect but late one. “Know what you need to know” – any analysis begins with the question of what decisions need to be made.
Among the ACAPS analytical frameworks are the INFORM severity index (31 indicators in three dimensions: impact, victim conditions, complexity), humanitarian access assessment methodology (three pillars, nine indicators), risk analysis framework (threat × vulnerability ÷ capacity) and scenario planning. In particular, The Good Enough Guide, a practical step-by-step guide to conducting a needs assessment in the early days after a disaster, contains the principle that an imperfect assessment is better than no assessment, as long as its limitations are clearly defined.
What key donors demand: EU, UN agencies and others
European Union
The EU applies all six OECD DAC criteria and adheres to the “Evaluation First” principle enshrined in the 2023 “Evaluation Matters” policy. The DG INTPA Evaluation Handbook (July 2024) describes a six-phase evaluation process: preparatory, introductory, interim, synthetic, dissemination and follow-up. The evaluation report should include an executive summary, description of the intervention with theory of change, methodology, findings according to the DAC criteria, conclusions, recommendations, lessons learned, and appendices. A certificate of financial statements becomes mandatory for cumulative reimbursement requests of €325,000 or more. For grants regulated by NDICI-Global Europe, a logical framework with clear indicators, baselines and targets is mandatory.
UN Agencies
All UN agencies follow UNEG norms and OECD DAC criteria. UNDP sets clear financial thresholds: projects with a budget of over $5 million (up to 4 years) require at least one evaluation, and projects lasting more than 4 years require both an interim and a final evaluation. Projects with a budget of $3-5 million require at least one estimate. GEF projects with a grant of more than $2 million require a mandatory interim review and final evaluation. Management response is submitted within 6 weeks.
UNICEF additionally emphasizes the rights of the child and the GEROS (Global Evaluation Reports Oversight System) quality assessment system. The United Nations High Commissioner for Refugees (UNHCR) requires regular evaluation of implementing partners according to pre-agreed criteria, and for local NGOs with projects over $100,000 — a mandatory audit certificate.
Common to all UN agencies are requirements for mainstreaming gender equality and human rights (UN-SWAP), public access to all evaluations, management response with an action plan, and independence of evaluators.
How It All Fits Together: Frames Comparison
| Framework | Focus | Key tool | For whom it is most relevant |
|---|---|---|---|
| OECD DAC | Evaluation criteria for interventions | 6 criteria + quality standards | All NGOs and donors |
| UNEG | Rules of evaluation activities in the UN system | 14 norms + 5 standards | NGOs working with UN agencies |
| AEA / JCSEE | Ethics of the evaluator and evaluation quality | 5 principles + 30 standards | Professional assessors |
| BOND | Quality of evidence in the NGO sector | 5 principles of evidence + checklist | British and international NGOs |
| ACAPS | Humanitarian analysis and needs assessment | Analytical frameworks (INFORM, Risk) | NGOs in the humanitarian sector |
Conclusions: what this means for Ukrainian NGOs
Understanding these standards is not an academic exercise, but a practical tool for organizations working with international donors. The OECD DAC criteria have become de facto mandatory: they are used by the EU, UNDP, UNICEF and most large foundations. The difference between donors is in the details: the EU emphasizes coherence with national strategies, while UN agencies, for example, focus on gender integration and human rights.
For an NGO preparing for an external evaluation, the practical minimum is as follows: integrate the evaluation into the project design from the very beginning, laying down the budget and basic data; to ensure the independence of the appraiser and the absence of conflicts of interest; require the evaluator to clearly delineate findings, conclusions, recommendations and lessons learned; check the quality of the report according to BOND principles (voice of beneficiaries, triangulation, transparency of methodology, contribution argument). ACAPS tools will be useful for those who work in a humanitarian context and need quick but structured analysis.
Ultimately, quality external evaluation is not an audit for the sake of reporting. It’s a mechanism that helps an organization understand what’s working and what’s not, and demonstrate that to donors in a language they understand and recognize.