When Monitoring Loses Its Meaning
The Political Economy of Weak MEL in Climate Finance - and What It Would Take to Change It
By Helge Rieper - CMC’s Senior Impact and Fragility/Climate Advisor
Abstract
Monitoring, Evaluation, and Learning (MEL) systems are a central pillar of climate finance architecture, repeatedly invoked as safeguards for accountability, effectiveness, and impact. Yet despite increasingly elaborate results-based management frameworks, independent reviews, academic literature,and practitioner experience suggest that MEL systems in climate finance remain procedurally compliant but strategically weak.
This paper argues that the persistent fragility of MEL is not primarily a technical problem, but a structural and political one. Drawing on institutional reviews of major climate funds, climate evaluation literature,and first-hand experience with MEL design and implementation, it identifies systemic weaknesses at fund, Accredited Entity, and project design levels -with particular attention to the additional challenges posed by fragile,conflict-affected, and climate-stressed contexts. It concludes that current MEL practices often function more as instruments of reassurance than of learning and proposes a set of structural and practical reforms. Without a rebalancing of incentives, capacity, and decision-making power - and without alignment to the mainstream frameworks that move capital across the wider economy - climate finance risks continuing to measure activity rather than transformation.
Executive Summary
MEL in climate finance has grown procedurally elaborate but strategically weak. Despite increasingly sophisticated results frameworks, MEL rarely shapes decisions where they matter most: at the boards of climate funds,at the level of Accredited Entities designing programs, and - critically -across the mainstream finance system that ultimately allocates the capital required for transformation.
This paper makes four arguments:
§ MEL is structurally constrained, not just technically weak. Incentives reward upward reporting over learning, aggregation over meaning, and reassurance over honest evidence.
§ MEL must extend beyond climate funds into mainstream finance.
Strategic frameworks already in use by decision-makers - PESTLE, the UK5 Case Model - and reporting standards already adopted by capital markets -IFRS S1/S2, ISSB, TCFD, GRI, SASB - are the channels through which climate risk becomes material to real-world decisions. Climate finance MEL should align with and feed into these, not run in parallel to them.
§ Systemic change is the missing impact metric. Climate funds have built bespoke logframes and indicator libraries instead of embedding climate risk and resilience into themanagement and reporting systems that govern trillions in mainstream capital.This is the single biggest barrier to demonstrating systemic impact.
§ Without evidence of systemic change, replenishment is at risk. Contributors increasingly question relevance. If climate funds cannot show that their interventions shift mainstream incentives, standards, and capital flows - not just deliver project outputs - the case for the next replenishment cycle will weaken.
The paper proposes a set of structural reforms and concludes with defined actionables for funds, Accredited Entities, consultants, and contributors.
1. Introduction: The MEL Paradox in Climate Finance
Few domains have embraced the language of monitoring, evaluation, andlearning as enthusiastically as climate finance. Over the past decade,multilateral climate funds and implementing organisations have adopted increasingly sophisticated Results Management Frameworks, indicator libraries,and evaluation policies. The Green Climate Fund (GCF), for instance, has moved from initial performance measurement frameworks to an Integrated ResultsManagement Framework (IRMF) - now evolving into the Harmonised IRMF (H-IRMF) -explicitly intended to strengthen consistency, comparability, and learningacross its portfolio.
At the same time, a growing body of evidence highlights a persistent disconnect between MEL policy ambition and actual influence on learning or decision-making. Reviews undertaken under the UNFCCC, OECD, and independent evaluation offices consistently note challenges related to indicator relevance,institutional capacity, feedback loops, and the dominance of accountability over learning. Practitioners echo these concerns daily, while formal documentation often stops short of interrogating the institutional incentives that sustain them.
This paper refers to this tension as the MEL paradox: MEL has never been more prominent in climate finance discourse, yet its practical influence on project quality, adaptive management, and strategic decision-making remains limited.
The paradox is not unique to climate finance. International development has grappled with similar dynamics for decades. But climate finance - with its scale, urgency, and transformational ambition - makes the stakes uniquely high.A climate adaptation project that reports activity without understanding whether resilience has actually improved is not merely a missed learning opportunity; it is a missed chance to protect vulnerable populations in a closing window.
2. Fund-Level MEL: Architecture Without Leverage
2.1 Indicator Systems That Privilege Aggregation Over Meaning
At fund level, MEL is dominated by standardised indicator systems designed to enable aggregation across highly diverse projects and country contexts. This logic is understandable: multilateral funds must report portfolio-wide results to governing bodies, contributors, and conventions. But the pursuit of aggregation frequently comes at the expense of analytical depth.
OECD and UNFCCC guidance documents explicitly acknowledge the challenges of measuring adaptation, resilience, and transformational change, noting the limitations of purely quantitative indicators and the need for context-sensitive approaches. In practice, however, portfolio-level reporting continues to rely on proxy indicators that are weakly linked to actual climate outcomes - especially in adaptation finance.
Example: A GCF-funded school resilience programme might report the number of schools retrofitted with cyclone-resistantroofing. This is useful as an output, but says nothing about whether children actually return to school faster after a cyclone, whether teachers are trainedto use early warning systems, or whether school safety plans have changed community preparedness behaviour. The indicator satisfies the template; thelearning question goes unanswered.
Similarly, adaptation funds routinely report the number of beneficiaries reached - a figure that is easily aggregated but reveals little about the quality, depth, or durability of the benefit. Two projects reporting 50,000 beneficiaries each may differ radically in their actual contribution to resilience.
The result is a system capable of answering "What was delivered?" but far less capable of answering "What changed, why, and for whom?" What cannot be standardised is systematically marginalised -even when it is central to climate effectiveness.
What would help: Funds could complement aggregated indicators with a mandatory set of context-specific outcome narratives - structured accounts of change that follow a common template but allow projects to describe the causal pathway in their own terms. The Climate Investment Funds' Transformational Change Learning Partnership has experimented with this approach, producing portfolio-level insights that pure indicator data could not.
2.2 Underinvestment in MEL Capacity
While climate funds routinely invest heavily in fiduciary standards,safeguards, and financial oversight, MEL capacity receives comparatively modest attention. Readiness and support programmes often mention MEL in broad terms,but systematic investment in analytical skills, learning systems, and theory-based evaluation is limited.
Example: The GCF's Readiness and Preparatory Support Programme has disbursed hundreds of millions of dollars to strengthen country capacity. Yet readiness funding overwhelmingly targets National Designated Authorities' institutional setup, stakeholder engagement,and pipeline development. MEL capacity - the ability to design credible baselines, construct testable theories of change, manage data quality in the field, and commission rigorous evaluations - receives a fraction of the attention and budget.
This asymmetry is implicitly recognised in repeated calls for improved coherence, learning, and indicator quality across the multilateral climate fund architecture. The persistence of these calls suggests that the problem is not a lack of awareness, but a lack of prioritisation. MEL is treated as a necessary compliance function rather than a core strategic capability.
What would help: Ring-fenced MEL capacity funding within readiness programmes - not as an optional annex, but as a required component. The GCF's recent MEL capacity-building initiative for 140+Accredited Entities represents a step in the right direction, but sustained investment must extend beyond one-off training to long-term systems strengthening.
2.3 Decision-Makers Without MEL Literacy
Perhaps most critically, MEL remains institutionally peripheral to climate finance decision-making. Boards and senior management teams are typically composed of diplomats, economists, financiers, and sector specialists. While these perspectives are essential, they are often not complemented by MEL expertise capable of interrogating project logic.
As a result, funding decisions are made on the basis of proposal narratives, risk assessments, and strategic fit, with limited capacity to question theories of change, indicator logic, or evaluability. MEL frameworks exist on paper, but their interpretation and use are delegated downward.Learning, where it occurs, struggles to travel upward to where strategic choices are made.
Example: A board reviewing a $50 million adaptation programme is well equipped to assess fiduciary risk, country ownership, and alignment with national priorities. But it is less equipped to ask: Is this theory of change testable? Are the proposed indicators actually measuring resilience, or measuring proxies that may have no causal relationship to resilience? Will the MEL budget - typically 2–3% of total project cost - be sufficient to answer whether this intervention worked?
What would help: Independent MEL advisory capacity at board level - not as a veto, but as a standing function thatreviews proposal logic, evaluability, and the credibility of proposed results frameworks before approval. This mirrors the role of independent evaluation offices, but positioned earlier in the cycle where it can influence design rather than merely judge implementation.
3. Accredited Entities:Compliance Without Rigor
3.1 The Post-Approval Construction of MEL Systems
Accredited Entities (AEs) sit at the operational heart of climate finance delivery. Under frameworks such as the GCF's accreditation and monitoring and accountability policies, AEs are responsible for establishing project-level monitoring systems and commissioning evaluations.
In practice, however, MEL systems are frequently finalised only after project approval. Although funding proposals include monitoring plans, these are often generic, minimally costed, and weakly integrated into project logic.Detailed system design - indicator operationalisation, baseline planning,sampling frames, data management architecture - is deferred until implementation, at which point incentives shift to disbursement and delivery rather than analytical rigor.
Example: A multi-country adaptation programme approved with a logframe containing 30+ indicators may arrive at implementation with no baseline data, no sampling strategy, no data management system, and a MEL budget insufficient to address any of these gaps. The implementing team inherits a framework that was designed to pass review, not to generate evidence. The first year is then spent retrofitting a MEL system under operational pressure - with predictable consequences for quality.
This pattern is not occasional; it is systemic. It reflects a design-stage incentive structure where approval is the primary milestone, and post-approval MEL construction is treated as an administrative afterthought.
3.2 Complexity as Explanation, Not Alibi
Climate interventions are undeniably complex. Adaptation outcomes are context-dependent, causal pathways are uncertain, and attribution is difficult.However, evaluation literature - and decades of development practice -demonstrate that complexity requires more rigorous theory-based and adaptive MEL approaches, not less.
The challenge is real, and it should be named honestly. Measuring whether a community's resilience to drought has improved over five years is genuinely harder than measuring whether a road was built. But difficulty is not impossibility. Methods exist: contribution analysis, process tracing, outcome harvesting, realist evaluation, qualitative comparative analysis, andquasi-experimental designs have all been applied successfully in complex climate and development settings.
The problem arises when complexity is invoked to justify vague indicators, under-specified theories of change, and absent baselines - not as a methodological challenge to be addressed, but as a reason to lower the bar.
Example: A forest conservation programme reports "improved ecosystem resilience" as an outcome indicator without defining what resilience means in the specific context, what baseline condition is being compared, or what evidence would demonstrate improvement versus natural variation. Challenged on this, the response is often:"Ecosystem resilience is inherently complex and difficult to measure." This is true - but it is not a justification for measuring nothing meaningful.
What would help: Mandatory evaluability assessments at the design stage. Before approval, every project above a threshold size should demonstrate that its proposed outcomes are measurable with the proposed MEL budget and methods. If they are not, the design should be revised - not the standards lowered.
3.3 Accountability Upward, Learning Nowhere
Participatory and learning-oriented MEL is formally encouraged across climate finance, including within GCF guidance. Yet analyses of participatory monitoring approaches note limited clarity, weak enforcement, and inconsistent implementation across projects.
The dominant accountability relationship remains upward: from project toAE, from AE to fund, from fund to contributors. This architecture is necessary -public funds require accountability. But when upward reporting becomes the only function MEL serves, learning collapses.
Example: An AE submits Annual Performance Reports that meticulously track output delivery against milestones. The reports pass review. But the same AE has no structured mechanism for feeding evaluation findings back into the design of new projects. Lessons "learned" in a final evaluation of one programme are not systematically available to the team designing the next one - because the learning function has no institutional home, no budget, and no authority.
What would help: Learning mandates with teeth.AEs should be required to demonstrate - at re-accreditation or portfolio review- how evaluation findings from completed projects have been integrated into the design of new ones. This creates a feedback loop that currently exists in policy but rarely in practice.
4. Design Failures Upstream:Where MEL Loses Before It Begins
One of the most persistent weaknesses of MEL systems lies not in implementation, but in project design. Climate finance proposals frequently display ambitious objectives supported by fragile theories of change and indicators selected to satisfy templates rather than analytical logic.
From a practitioner perspective, this is where MEL failure is often pre-determined. Project concepts are shaped by eligibility criteria, investment priorities, and political considerations. Design timelines are compressed -sometimes to a few months for programmes worth tens of millions of dollars.Technical reviewers may flag weaknesses, but final decisions rest with bodies that often lack the expertise or incentive to reject proposals on MEL grounds.
Example: A $30 million livelihoods and climate adaptation programme in a fragile state is designed with a six-month concept development window. The theory of change assumes linear causation:training leads to behaviour change, behaviour change leads to income diversification,income diversification leads to resilience. Each step is plausible inisolation, but the chain is never tested, the assumptions are never made explicit, and no mechanism is built in to detect when the chain breaks - as it inevitably will in a conflict-affected context where displacement, market disruption, or institutional collapse can invalidate the entire logic.
The programme is approved. Three years later, an evaluation finds that training was delivered (outputs achieved), but behaviour change was negligible because the economic context shifted. The evaluation recommends "stronger theories of change in future programme design." The recommendation is noted. The next programme repeats the pattern.
What would help: Institutionalised design-stagequality gates. Projects with under-specified theories of change, absent baselines, or MEL budgets below a credible threshold (at minimum 5% of total project cost for complex interventions) should not proceed to approval without revision. This is not a call for bureaucratic obstruction - it is a call for the same rigour applied to fiduciary assessment being applied to results logic.
5. MEL in Fragile,Conflict-Affected, and Climate-Stressed Contexts
5.1 The Compounding Challenge
A growing share of climate finance flows to countries that are not merely climate-vulnerable but simultaneously fragile, conflict-affected, or inpost-crisis transition. The Sahel, the Horn of Africa, the Pacific small island states, South Sudan, Afghanistan, Myanmar - these are contexts where climate adaptation is most urgent and where MEL is most difficult.
Fragility compounds every MEL challenge identified in this paper.Theories of change must account for non-linear disruptions - coups,displacement, armed conflict, economic collapse - that can invalidate programme logic overnight. Baselines are harder to establish when populations are mobile,institutions are weak, and data systems are absent or unreliable. Field access is constrained by insecurity, seasonal flooding, or remotely managed operations. Indicator targets set at design stage may become meaningless within a year if context shifts radically.
Example: A food security and climate resilience programme in Sudan establishes baseline data across 90 schools in six counties. Within twelve months, armed conflict displaces half the target population. Schools close. Enumerators cannot access field sites. The logframe -designed for stability - cannot accommodate the new reality. The project continues to report against original indicators, producing numbers that are technically accurate but contextually meaningless.
5.2 What Fragile Contexts Demand
MEL in fragile and conflict-affected settings requires a fundamentally different operating model - not lower standards, but different methods and greater flexibility. The core requirements include:
Adaptive results frameworks. Logframes and theories of change must be designed for iteration. This means building in structured review points where assumptions are re-examined,indicators are adjusted, and targets are recalibrated based on contextual change - not as a sign of failure, but as a sign of analytical honesty.
Conflict-sensitive MEL. Data collection in fragile contexts can itself cause harm. Surveys that identify beneficiaries by ethnicity, location, or political affiliation can expose them to risk. Enumerators can be targeted. Community engagement can raise expectations that cannot be met. MEL design must integrate do-no-harm principles from the outset - not as a checklist item, but as a genuine analytical lens.
Remote and hybrid data collection. When field access is constrained, MEL systems must be able to operate through remote monitoring, third-party verification, satellite and geospatial data, phone-based surveys, and structured community feedback mechanisms. This is not a compromise - it is a capability that must be designed, tested, andresourced.
Example: Conflict Management Consulting(CMC's) Third-Party Monitoring of the GIZ Sudan portfolio operates across five projects in Eastern and Western Sudan under conditions of active conflict and remote management. The system integrates ACLED conflict event data, geospatial risk mapping, ad hoc location assessments, partner background checks, spotchecks, and end-of-project assessments - delivering 130+ field monitoring activities and biannual reports. This is not a simplified version of"normal" MEL. It is a purpose-built system that accepts operational constraints and works within them to produce credible, usable evidence.
Local partnerships and national ownership. In fragile contexts, international consultants parachuting in for two-week field missions produce limited value. Sustained MEL quality depends on embedded local partners with contextual knowledge, language capacity, community trust, and the ability to operate when international access is restricted.Building and investing in these partnerships is not a cost to be minimised - itis the foundation of credible evidence in difficult settings.
Higher MEL budgets, not lower. There is a persistent and counterproductive assumption that MEL in fragile contexts should cost less because data collection is"simpler" or "smaller scale." The opposite is true. Remotedata collection, third-party monitoring, security protocols, enumerator training in conflict-sensitive methods, adaptive framework management, and multi-modal verification all cost more than standard survey work in stable settings. MEL budgets in fragile contexts should be proportionally higher- typically 7–10% of total project cost - not lower.
5.3 The Risk of Fragility-Blind Climate Finance
Climate finance that flows to fragile states without adapting its MELexpectations to fragile realities produces a specific and damaging outcome: it creates a fiction of evidence. Reports are filed. Indicators are populated.Annual Performance Reports are submitted. But the connection between what is reported and what is happening on the ground becomes increasingly tenuous.
This is not a marginal concern. As climate impacts intensify and fragility spreads, a growing proportion of the global climate finance portfolio will operate in exactly these conditions. If the MEL architecture is not designed for fragility, it will not merely underperform - it will actively mislead.
6. From Bespoke Logframes to Systemic Integration: Aligning MEL with Mainstream Frameworks
6.1 The Parallel-Universe Problem
Climate funds have built their own results architectures - the GCF's IRMF and now H-IRMF, the GEF's core indicators, the Adaptation Fund's results framework, the CIF's transformational change indicators. Each is internally coherent. Collectively, they form a parallel universe of metrics that does not connect to the frameworks already shaping decisions in the wider economy.
This is a strategic failure. Climate change is not a sector - it is a systemic risk and opportunity that cuts across every balance sheet, regulator,supply chain, and public investment decision. MEL that lives only insideclimate funds cannot drive systemic change. To be commercially and politically relevant, MEL must speak the language of the systems it seeks to shift.
6.2 Strategic Frameworks That Already Guide Decision-Makers
Two frameworks already structure how public and private decision-makers think about risk and investment. Climate finance MEL should engage with them directly.
PESTLE analysis (Political, Economic, Social,Technological, Legal, Environmental) is widely used in corporate strategy,public policy, and investment due diligence to assess external risk and opportunity. Climate finance interventions can and should be assessed through aPESTLE lens - not as an academic exercise, but to surface how a program shifts the political feasibility, economic incentives, social acceptance,technological readiness, legal frameworks, and environmental conditions that determine whether change is durable. A program that improves a single environmental indicator but leaves the legal and economic incentives unchanged will not produce systemic impact. PESTLE makes that visible.
The UK Government's 5 Case Model (Strategic, Economic, Commercial, Financial, Management) is the standard framework for public investment appraisal in the UK and is increasingly used internationally. It asks the questions climate finance often skips: Is the strategic case compelling beyond climate-specific objectives? Is the economic case robust on whole-life costs and benefits? Is the commercialc ase viable for delivery partners? Is the financial case affordable and sustainable? Is the management case realistic? A climate adaptation program that cannot answer these five questions will not attract co-financing from mainstream public finance ministries, which is where transformational scale ultimately lies.
What would help: Climate funds should require, atdesign stage, that programs above a threshold size include a PESTLE assessment of the systems they aim to shift and a 5 Case Model business case adapted for climate outcomes. Both can be used alongside, not instead of, theories of change - PESTLE clarifies what system change requires; the 5 Case Model clarifies whether the intervention is investable by mainstream actors.
6.3 Reporting and Management Standards That Already Move Capital
Mainstream capital is increasingly governed by a set of standards that make climate risk material to conventional reporting and decision-making:
§ IFRS S1 and S2 (ISSB): the global baseline for sustainability and climate-related financial disclosures, now being adopted by major jurisdictions.
§ TCFD recommendations: the architecture underlying IFRS S2 and most regulatory disclosure regimes.
§ GRI standards: the dominant framework for sustainability reporting on impact.
§ SASB standards: industry-specific materiality metrics integrated into ISSB.
§ EU CSRD and ESRS: mandatory sustainability reporting for large firms operating in the EU.
§ ICMA Green Bond Principles and Climate Resilience Principles: standards for labelled debt instruments.
§ ISO 14001(environmental management) and ISO 14097 (climate finance assessment).
These are the systems through which climate risk becomes material to conventional reporting, investment decisions, and capital allocation. The climate fund MEL, which does not align with them, produces evidence that mainstreamactors cannot use.
The GCF's decision to build its own logframe architecture - rather than to require Accredited Entities and recipient countries to use, strengthen, andextend IFRS, GRI, SASB, TCFD, or ISO-aligned reporting - illustrates the problem. It generates internally consistent reporting that contributors canread, but it does not embed climate risk into the management systems of the banks, ministries, corporates, and regulators whose decisions actually determine whether transformation happens.
What would help:
§ Climate funds should require, where applicable, that Accredited Entities report against the relevant international standard (IFRSS2, GRI, SASB, TCFD, ICMA Principles, ISO 14001/14097) as part of project MEL -not as an additional burden, but as the primary reporting channel, withfund-specific indicators mapped onto these standards.
§ Readiness funding should be redirected toward buildingnational and entity-level capacity to apply IFRS S1/S2, ISSB, and TCFD-aligned reporting - embedding climate risk into the management systems that govern domestic finance and corporate decisions.
§ Evaluations should assess systemic change explicitly:whether the program shifted regulatory standards, disclosure practices,management systems, or capital allocation patterns - not only whether project outputs were delivered.
6.4 The Replenishment Argument: Why This Matters Now
This is not a methodological refinement. It is a strategic survival question for climate finance.
Contributors to the GCF, GEF, CIF, and Adaptation Fund face increasing fiscal pressure, competing priorities, and political scrutiny. Each replenishment cycle is harder than the last. The question contributors increasingly ask is not "did the fund deliver outputs?" but "did the fund produce systemic change at a scale that justifies continued contribution?"
A fund that can demonstrate it shifted mainstream disclosure standards,embedded climate risk into national budgeting systems, mobilised co-financing under standard public investment appraisal frameworks, and changed how privatecapital prices climate risk - that fund has a replenishment case. A fund that can demonstrate only that it retrofitted schools, trained farmers, and reported beneficiary numbers does not.
Without demonstrating systemic change, climate funds will struggle to demonstrate relevance - and replenishment will become increasingly difficult to secure. This is not a hypothetical risk.It is the central strategic challenge facing the climate finance architecture today.
7. The Political Economy of Weak MEL
The persistence of weak MEL systems cannot be fully explained by capacity constraints or methodological challenges. It reflects a deeper political economy.
Robust MEL systems create institutional risk. They can reveal under performance, unintended effects, or misalignment with stated objectives.They can challenge narratives of success and question strategic priorities. In highly politicised funding environments - where contributor confidence, board dynamics, and reputational management are constant considerations - this creates structural disincentives for genuine transparency.
Evaluation community reflections increasingly acknowledge this dynamic:MEL in climate finance has historically been oriented toward donor assurance rather than adaptive learning. While rhetoric has shifted toward learning,institutional practice remains cautious. Evaluation findings that challenge programme narratives are often softened, delayed, or buried in annexes.
Example: A mid-term evaluation of a large adaptation programme finds that a key component - community-based early warning systems - is not functioning as designed because local government structures lack the capacity and incentive to maintain them. The finding is clear and evidence-based. But the evaluation report, after multiple rounds of stakeholder review, reframes the finding as "early warning systems are at an early stage of maturity and require continued support." The institutional response is to extend the project, not to redesign the component. The evaluation has produced evidence; the system has neutralised its implications.
Weak MEL, therefore, represents a form of institutional equilibrium. Itallows funds to demonstrate procedural responsibility without exposing themselves to disruptive insights. It satisfies reporting requirements without demanding strategic change. This equilibrium is stable - and it will not be disrupted by better indicators alone.
What would help: Creating protected space for honest evaluation. This means evaluation independence that is not merely formal(a separate office on the org chart) but functional: independent evaluation units with their own budgets, direct board reporting lines, and the authority to publish findings without management approval. Some institutions - notably the GEF's Independent Evaluation Office - have moved in this direction. The principle needs to be standard, not exceptional.
8. Learning Without Power
Climate finance institutions increasingly speak of learning agendas,communities of practice, and knowledge products. Initiatives such as the Climate Investment Funds' Transformational Change Learning Partnership and theGCF's own learning-oriented work demonstrate the potential of structured learning efforts at the portfolio level.
Yet even well-designed learning initiatives often operate parallel to,rather than within, core decision-making processes. Learning outputs are produced - case studies, synthesis reports, good-practice briefs - but the pathway from insight to decision remains unclear.
Example: A climate fund produces a comprehensive synthesis of adaptation evaluation findings across 40 projects,identifying five recurring design weaknesses. The synthesis is published,presented at a board meeting, and cited in the annual report. But the next funding cycle's project approval criteria are not updated to reflect the findings. The next cohort of projects repeats four of the five weaknesses. The learning product exists; the learning does not.
Learning without institutional power remains marginal. Insights generated through evaluations and studies rarely influence investment criteria,approval thresholds, or strategic orientation. MEL outputs exist, but they do not reliably shape future choices.
What would help: Formal"learning-to-decision" protocols. Each major evaluation or synthesis should be accompanied by a management response with time-bound actions, tracked by the independent evaluation function and reported to the board. The GEF and several UN agencies already use this model. In climate finance, it should be the norm.
9. What Would Serious MEL Require?
Strengthening MEL in climate finance does not require more indicators orlonger reporting templates. It requires structural change - and a willingness to prioritise evidence with the same seriousness applied to fiduciary compliance.
9.1 MEL Literacy as a Governance Requirement
Decision-makers at board and senior management level should be equipped to interrogate project logic and evaluability - or supported by independent MEL expertise with genuine authority. This means MEL advisory capacity that is present at the table when funding decisions are made, not consulted afterward.
9.2 Design-Stage Quality Gates
Projects with weak theories of change, absent baselines, or MEL budgets below a credible threshold should not proceed to approval with the expectation that weaknesses will be corrected during implementation. Design-stage evaluability assessments should be a standard requirement, not an optional annex.
9.3 Ring-Fenced and Proportionate MEL Investment
MEL budgets should reflect the complexity and ambition of the intervention. For standard climate projects, a minimum of 5% of total project cost. For complex, multi-country, or fragile-context interventions, 7–10%. This investment should cover baseline establishment, data management systems,mid-term and final evaluations, and - critically - learning functions that operate throughout implementation, not only at milestone points.
9.4 Fragility-Adapted MEL Standards
Climate funds operating in fragile and conflict-affected contexts should adopt adapted MEL guidance that recognises the distinct requirements of these settings: adaptive frameworks, remote monitoring capability, conflict-sensitive data collection, local partnership investment, and higher MEL budgets. These should be standard operating requirements for fragile-context portfolios, not ad-hoc accommodations.
9.5 Empowered Learning Functions
Learning functions must be institutionally empowered - with pathways that link evaluation insights directly to strategy, not just reporting. This means management response protocols, tracked follow-up, and board-level accountability for acting on evidence. Learning that does not influence decisions is not learning - it is documentation.
9.6 Protected Evaluation Independence
Independent evaluation offices must have functional - not merely formal -independence: their own budgets, direct reporting lines to the governing body,and the authority to publish findings without management approval or editorial softening. The quality of evaluation is only as good as the freedom to report uncomfortable truths.
9.7 The Independence Problem: Consultants, Contracts, and the Cost of Honesty
The discussion of evaluation independence typically focuses on institutional structures - independent evaluation offices, direct board reporting lines, protected budgets. These are essential. But there is a more uncomfortable dimension of the independence question that is rarely addressed in policy documents: the position of external MEL consultants and evaluation specialists whose livelihoods depend on being rehired.
Climate finance MEL is delivered, in large part, by a global consultancy market. Accredited Entities commission baselines, mid-term reviews, and finalevaluations from consulting firms and independent specialists. These consultants are expected to produce rigorous, honest assessments. They are also- unavoidably - dependent on the commissioning organisation for future work.
This creates a structural tension that is well understood by practitioners but seldom acknowledged formally.
The unspoken calculus. An evaluator conducting a mid-term review discovers that a flagship programme component is fundamentally misconceived - not merely underperforming, but based on assumptions that field evidence has invalidated. Reporting this finding honestly may be professionally correct, but it is also professionally risky.The commissioning AE may not appreciate having its design logic questioned. The donor may not welcome evidence that challenges the narrative presented to its board. The evaluator knows - from experience, from colleagues' experiences,from the structure of the market - that being "difficult" can mean not being invited to the next tender.
The result is predictable. Findings are softened. Language is hedged."The programme faces significant implementation challenges" replaces"The programme design is flawed." "Adaptive management is recommended" substitutes for "This component should be discontinued." The evaluation fulfils its procedural function. The learning function is compromised.
Example: A consultant team conducts a final evaluation of a multi-country climate resilience programme. The evidence clearly shows that one country's implementation was negligent - funds were misspent, activities were not delivered, reported results were fabricated. Theteam documents this in the draft report. During the stakeholder review process,the AE requests that the finding be "contextualised" and the language"balanced." The team is reminded that the AE is tendering a major new evaluation contract next quarter. The final report describes"implementation challenges related to partner capacity." The fabrication is not named. The consultant team is invited to bid on the next contract.
This is not hypothetical. Every experienced MEL practitioner has encountered some version of this dynamic. It is not universal - many commissioners genuinely value honest findings - but it is common enough to shape behaviour across the market.
What would help:
Structural protections for evaluator independence. Evaluation contracts should include explicit provisions protecting evaluators from retaliation for honest findings. This could include clauses specifying that evaluation findings cannot be grounds for exclusion from future tenders, or requirements that disputes over findings be adjudicated by an independent third party (such as the fund's Independent Evaluation Office) rather than resolved through stakeholder"negotiation."
Separation of commissioning and rehiring decisions. Where possible, the decision to commission an evaluation and the decision to rehire a consultant for future work should be made by different institutional actors. This reduces the direct leverage that a dissatisfied programme team can exercise over an evaluator's future employment.
Publication of unedited evaluation reports. Requiring that evaluation reports be published in their original form -before stakeholder review edits - would create transparency about the gap between what evaluators find and what institutions are comfortable publishing.This is not about shaming institutions; it is about creating accountability for the editing process itself.
A cultural shift in how "difficult" findings are received. Ultimately, the independence problem reflects institutional culture. Organisations that genuinely value learning will treat evaluators who surface uncomfortable truths as assets, not threats. This requires leadership - from fund boards, from AE management, from donor representatives - that visibly rewards honesty and visibly protects those who deliver it.
None of this is easy. The consultancy market is competitive, and consultants are rarely in a position to demand structural protections. But climate finance institutions have the power to create these protections if they choose to. The question is whether they will.
9.8 Alignment with International Standards and Mainstream Frameworks
Climate finance MEL cannot drive systemic change if it operates in isolation from the frameworks that govern mainstream finance and management.Funds should:
§ Require alignment of project-level reporting with IFRSS1/S2, TCFD, GRI, SASB, ICMA Principles, and ISO 14001/14097 wherever applicable.
§ Embed PESTLE analysis and 5 Case Model appraisal into design-stage requirements for programs above a threshold size - alongside theories of change.
§ Map fund-specific indicators (IRMF/H-IRMF, GEF core indicators, AF results framework, CIF transformational change indicators) onto international standards so that data is usable by mainstream actors and decisions.
§ Reorient readiness funding toward building IFRS/ISSB/TCFD reporting capacity in recipient countries and Accredited Entities.
These are not additional reporting burdens. They are the channels through which climate finance MEL becomes commercially relevant and can demonstratesystemic change.
10. A Vision: What Would Genuinely Learning-Oriented MEL Look Like?
This paper has catalogued systemic weaknesses. But critique without vision is incomplete. If the reforms proposed here were implemented - if MEL were taken seriously as a strategic function rather than a compliance requirement - what would a genuinely learning-oriented climate finance MEL system actually look like?
The following is not a utopian fantasy. It draws on existing good practice from leading evaluation offices, innovative climate programmes, and decades of development learning. It is achievable - but it requires intentional institutional design.
10.1 At the Fund Level: MEL as a Strategic Function
A learning-oriented fund would treat MEL not as a reporting burden butas a core capability for adaptive portfolio management.
Board-level MEL literacy. Board members would receive regular briefings on MEL findings - notjust "results achieved" but "what we are learning about whatworks." At least one board seat or standing advisory role would be held by a MEL/evaluation specialist with authority to question proposal logic and evaluability.
Design-stage evaluability review. No project above a threshold size would proceed to approval without an independent evaluability assessment confirming that the theory of change is testable, the indicators are meaningful, the baseline plan is credible, and theMEL budget is sufficient. Projects that fail this review would be returned forrevision, not approved with caveats.
Learning-integrated approval criteria. Approval criteria would include explicit requirements for how the project will generate and use evidence - not just what it will deliver.Projects that cannot explain how they will learn and adapt would be deprioritised.
Mainstream-aligned reporting architecture. Fund-level indicators would be mapped onto IFRSS1/S2, TCFD, GRI, SASB, and ICMA Principles. Reporting would speak the languageof capital markets, regulators, and finance ministries - not only of climate convention secretariats.
Portfolio-level learning synthesis. The fund would invest in regular cross-project synthesis - not just aggregating indicators, but analysing patterns of success and failure,identifying recurring design weaknesses, and feeding these findings into updated guidance for new proposals. This function would be resourced, staffed,and empowered.
Public evaluation registry. All evaluations commissioned with fund resources would be published in an accessible, searchable registry - including management responses and follow-up actions. Transparency would be the default, not the exception.
10.2 At the Accredited Entity Level: MEL as Institutional Memory
A learning-oriented AE would treat MEL as the mechanism through which the organisation remembers what it has learned.
Dedicated learning function. The AE would maintain a dedicated learning unit or function - not merged with communications, not subordinated to programme delivery - responsible for synthesising evaluation findings, maintaining institutional knowledge, and feeding lessons into new programme design.
Pre-design learning briefs. Before any new programme is designed, the learning function would produce a brief synthesising what previous evaluations have found about similar interventions, contexts, or approaches. Design teams would be required to demonstrate how they have incorporated these lessons.
Standards-aligned project reporting. AEs would report against IFRS S2, GRI, SASB, TCFD, or ICMA Principles as the primary MEL channel, with fund-specific indicators mapped onto these standards.
Evaluation follow-up tracking. Every evaluation would generate a management response with time-bound commitments. Implementation of these commitments would be tracked and reported -to senior management, to the board, and to the fund.
Feedback loops to field teams. Learning would flow downward as well as upward. Field teams would receive synthesised insights from evaluations of comparable programmes - not as compliance requirements, but as practical support for adaptive management.
Honest internal culture. Senior leadership would visibly reward honesty and protect staff whosurface uncomfortable findings. Programme managers who acknowledge implementation challenges and propose course corrections would be valued more highly than those who maintain fictions of success.
10.3 At the Project Level: MEL as Adaptive Management
A learning-oriented project would treat MEL not as a reporting obligation but as the nervous system of adaptive implementation.
Testable theory of change. The theory of change would be explicit, specific, and testable - with clearly articulated assumptions and identified points at which evidence will be gathered to test them. It would be a living document, updated as implementation generates new evidence.
PESTLE and 5 Case Model integration. Project design would include a PESTLE analysis of the systems the intervention aims to shift and a 5 Case Model business case showing investability for mainstream actors. These would sit alongside the theory of change.
Real-time monitoring for decision-making. Monitoring data would be collected not for annual reports but for real-time decision-making. Dashboards, field feedback loops,and regular data review meetings would enable programme teams to detect problems early and adjust course.
Structured reflection points. Implementation would include structured "pause and reflect"moments - quarterly or semi-annually - where teams review monitoring data,assess progress against assumptions, and make documented decisions about adaptation. These would be built into the workplan, not added as afterthoughts.
Community feedback integration. Beneficiary and community feedback would be systematically collected and integrated into programme management - not as a box-ticking exercise, butas a genuine source of insight about whether the programme is working as intended.
Honest reporting. Progress reports would describe not just achievements but challenges, adaptations, and lessons. The incentive structure would reward honest reporting over narrative management.
10.4 For Fragile and Conflict-Affected Contexts: MEL as Adaptive Resilience
In fragile contexts, a learning-oriented MEL system would be designed for uncertainty from the outset.
Adaptive frameworks as standard. Logframes and results frameworks would be explicitly designed for iteration, with built-in review points, indicator adjustment protocols, and scenario-based contingency plans.
Conflict-sensitive by design. MEL tools and processes would be reviewed for conflict sensitivity before deployment. Data collection would bed esigned to minimise risk to respondents, enumerators, and communities.
Remote and hybrid capability. MEL systems would be designed to function under access constraints, with remote monitoring, third-party verification, and geospatial data integrated as core capabilities - not as emergency fallbacks.
Local partnership investment. A significant portion of MELbudget would be dedicated to building and sustaining local partner capacity -not as a cost to be minimised, but as the foundation of credible evidence in difficult settings.
Higher MEL budgets.Fragile-context projects would allocate 7–10% of total budget to MEL -reflecting the additional cost and complexity of generating credible evidence in these settings.
10.5 For the Consultancy Market: Protected Independence
A learning-oriented system would protect the independence of external evaluators.
Contractual protections. Evaluation contracts would include clauses protecting evaluators from exclusion for honest findings, with disputes adjudicated by independent third parties.
Publication of original reports. Evaluation reports would be published in their original submitted form,with any stakeholder-requested changes documented transparently.
Separation of commissioning and rehiring. Decisions to commission evaluations and decisions to award future contracts would be made by different institutional actors where possible.
Recognition of honest evaluation. Funds and AEs would actively recognise and reward consultants who deliver rigorous, honest work - including work that surfaces uncomfortable findings. Reputation in the market would be built on integrity, not accommodativeness.
10.6 The Cultural Foundation: Valuing Learning Over Narrative
Ultimately, all of these structural reforms depend on a cultural foundation: a genuine institutional commitment to learning over narrativemanagement.
This means:
§ Accepting that honest evidence sometimes challenges preferred stories.
§ Treating evaluation findings that reveal problems as opportunities for improvement, not threats to be neutralised.
§ Holding leaders accountable for acting on evidence,not just producing it.
§ Recognising that the purpose of MEL is not to demonstrate success, but to understand what is actually happening - and to use that understanding to do better.
This cultural shift cannot be mandated by policy alone. It requires leadership - from board chairs, from executive directors, from senior managers -that visibly models and rewards honest learning.
But culture follows structure. Institutions that design their MEL systems for learning - with protected independence, empowered learning functions, mainstream-aligned reporting, and genuine feedback loops - will,over time, develop cultures that value learning. Institutions that design their MEL systems for reassurance will continue to produce elaborate compliance without insight.
11. Conclusion: From Compliance to Systemic Change - and Why Replenishment Depends on It
Climate finance exists to drive transformation. Transformation is not the sum of project outputs. It is systemic change: shifts in the standards,management systems, regulatory frameworks, and capital allocation patterns that govern how the wider economy treats climate risk and resilience.
The MEL systems currently in place across climate finance are not designed to evidence systemic change. They are designed for upward accountability - reporting outputs to contributors, indicators to boards,narratives to conventions. This was adequate when the question was whether climate finance could deliver projects. It is now inadequate to ask whether climate finance can shift the system.
The strategic reforms required are clear. MEL must extend beyond bespoke fund logframes into the frameworks that already shape mainstream decisions:PESTLE and the 5 Case Model for strategic and investment appraisal; IFRS S1/S2,TCFD, GRI, SASB, ICMA Principles, and ISO 14001/14097 for reporting and management. Climate funds must stop building parallel universes of measurement and start embedding climate risk into the systems that move trillions, not millions.
The replenishment case follows directly. Contributors will not indefinitely fund an architecture that cannot evidence systemic impact. The next replenishment cycles for the GCF, GEF, CIF, and Adaptation Fund will be judged less on activity reports and more on whether climate finance has produced demonstrable shifts in how the broader economy values, prices, andmanages climate risk. Without that evidence, relevance erodes. Without relevance, replenishment falters. Without replenishment, the architecture itself is at risk.
The reforms proposed in this paper are not radical. They build on existing good practice. They require resources, but not impossibly large ones.What they require, above all, is political will: the willingness to treat evidence with the same seriousness as financial compliance, to align MEL with mainstream frameworks rather than around them, and to accept that honest learning sometimes produces uncomfortable findings.
The climate crisis does not have time for monitoring that produces reports but not insight, or for funds that produce projects but not systemic change. It demands MEL systems that align with mainstream standards, evidence systemic impact, and give contributors a credible answer to the question they are increasingly asking: are these funds still relevant?
The challenge is not to perfect the rituals of measurement. It is tomake monitoring and learning matter - by connecting them to the systems where transformation actually happens.
12. Key Actionable
For climate fund boards and secretariats
1. Adopt a systemic change impact lens. Require every program above a threshold size to articulate the system it aims to shift (using PESTLE) and the investment case it offers mainstream public finance (using the 5 Case Model or equivalent).
2. Map fund indicators to international standards. Map IRMF/H-IRMF, GEF core indicators, and Adaptation Fund metrics onto IFRS S2, TCFD, GRI, SASB, and ICMA Principles. Where indicators do not map, ask whether they should exist.
3. Embed systemic change in evaluation criteria. Require independent evaluations to assess whether the program shifted standards, regulations, management systems, or capital flows - not only whether outputs were delivered.
4. Make the replenishment case explicit. Commission a portfolio-level synthesis ahead of each replenishment cycle that evidences systemic change in standards alignment,mainstream finance mobilisation, and regulatory shifts.
5. Establish design-stage quality gates. No approval without a credible theory of change,evaluability assessment, baseline plan, and MEL budget (≥5% standard; 7–10%fragile contexts).
For Accredited Entities
6. Use the standards capital already uses. Report against IFRS S2, TCFD, GRI, SASB, or ICMA Principles as the primary MEL channel; map fund-specific indicators onto them.
7. Build a learning function with authority. A dedicated unit with budget and access to senior decision-making - not merged with communications or programme delivery.
8. Apply 5 Case Model logic to design. Ensure programs are investable by mainstream public finance - not only fundable by climate windows.
9. Track evaluation follow-up. Every evaluation produces a time-bound management response, tracked and reported.
For consultants and the evaluation market
10. Push for contractual protections. Include independence clauses; advocate for separation of commissioning and rehiring decisions.
11. Publish unedited findings where contracts allow. Build market reputation on integrity, not accommodation.
For contributors and oversight bodies
12. Condition replenishment on systemic change evidence. Make replenishment cases dependent on demonstrated shifts in standards, regulations, and capital flows - not only output delivery.
13. Fund mainstream alignment work. Direct readiness and capacity funding toward buildingIFRS, TCFD, and ISSB-aligned reporting capacity in recipient countries and Accredited Entities.
For the wider climate finance architecture
14. Stop building parallel universes. The next iteration of climate finance MEL should integrate with - and strengthen - international reporting and management standards, not replicate them.
15. Make systemic change the headline metric. Report it, evidence it, and let it carry the replenishment case.
About the author (for end of document or website):
Helge Rieper is a Senior Impact and Fragility/Climate Advisor at Conflict Management Consulting (CMC). He has over 25 years' experience designing MEL systems and leading evaluations across multilateral trust and climate funds and UN portfolios, with a focus on fragile, conflict-affected,and climate-stressed contexts. His work spans evaluability assessments,theory-based and mixed-methods designs, third-party monitoring during active conflict), and portfolio-level learning on adaptation and resilience. He specialises in making MEL matter - turning evidence into management decisions.
📧 helge.rieper@cmc-consulting.org 🔗linkedin.com/in/helgerieper
