How to Write a Log Frame That Donors Actually Approve

The logical framework — almost universally known as the log frame — is simultaneously the most important and most misunderstood tool in development project design. Every major donor, from USAID and the EU to the World Bank and bilateral funders like FCDO and GIZ, requires a log frame as part of any project proposal. Yet the quality of log frames submitted by Kenyan NGOs and government agencies remains astonishingly variable — and the weak ones are costing organisations funding they desperately need.

At the Centre of Corporate Studies, our Monitoring and Evaluation trainers review hundreds of log frames every year through our online M&E courses and consulting work. The patterns of failure are remarkably consistent. This article distils everything we know about what separates log frames that win funding from those that do not — and gives you a step-by-step guide to building one that donors will actually approve.

What is a Log Frame and Why Does It Matter?

The Logical Framework Approach (LFA) was developed by USAID in the 1970s as a tool for project planning, management and evaluation. It has since been adopted, adapted and refined by virtually every major development organisation in the world. The log frame itself is a 4×4 matrix — four rows (Goal, Purpose/Outcome, Outputs, Activities) and four columns (Description, Indicators, Means of Verification, Assumptions) — that tells the story of your project in a structured, logical way.

The central logic of the log frame is vertical: if you implement the Activities, you will produce the Outputs; if you produce the Outputs, you will achieve the Purpose (immediate outcome); if you achieve the Purpose, you will contribute to the Goal (long-term impact). This is the results chain. Everything else in your proposal — your budget, your implementation plan, your risk framework — should be consistent with and traceable to your log frame.

Donors review log frames because they are the most efficient way to assess whether a project has been designed with intellectual rigour, whether the team understands the problem they are trying to solve, and whether the proposed intervention is likely to work. A strong log frame signals to a donor that you know what you are doing. A weak log frame — vague, inconsistent, with unmeasurable indicators — signals the opposite, regardless of how compelling your narrative proposal might be.

The Seven Most Common Log Frame Mistakes

Before getting to the step-by-step guide, it is worth identifying the mistakes that most commonly cause log frames to fail. Understanding the failure modes makes the success criteria clearer.

1. Activity-level thinking at the outcome level

The single most common mistake is writing activities when the log frame calls for outcomes. "Training 200 women in financial literacy" is an activity. "200 women demonstrate improved financial management practices as measured by a validated financial behaviour assessment" is an output. "Increased household economic resilience among target women" is a purpose/outcome. Donors immediately spot this confusion — and it signals that the project team has not thought clearly about what change they are trying to create and why their activities will create it.

2. Unmeasurable indicators

Indicators must be SMART: Specific, Measurable, Achievable, Relevant and Time-bound. "Improved nutrition" is not a SMART indicator. "Percentage of children under 5 with a mid-upper arm circumference above 12.5cm, as measured by community health volunteers using standardised tools, increased from 62% at baseline to 78% by project end" is a SMART indicator. The difference is the difference between a donor understanding what you will measure and not understanding it.

3. Circular or inconsistent logic

The vertical logic of a log frame must be coherent. Each level must logically lead to the next. Where log frames often break down is at the Output-to-Purpose link: the outputs do not plausibly add up to the stated purpose. If your purpose is "reduced maternal mortality" but your outputs are all about training healthcare workers with no mention of access to emergency obstetric care, equipment, or community referral systems, the logic is incomplete — and a sharp donor reviewer will flag it.

4. Assumptions that undermine the project

Assumptions are the external conditions that must hold true for your results chain to work. Many proposal writers either leave assumptions blank (a red flag) or state assumptions so unlikely to hold that they undermine the entire project design. "Government will pass enabling legislation within 6 months" as a critical assumption in a 12-month project is a project-killer — unless you have specific evidence that this is likely to happen.

5. Means of verification that are not credible

The Means of Verification (MoV) column must specify exactly how you will collect the data to verify each indicator. "Government statistics" as the MoV for a nutrition indicator in a remote county with no functioning health information system is not credible. "Monthly community health volunteer reports compiled into a district database, verified quarterly by an independent data quality assessor" is credible.

6. Goal confusion — contribution vs attribution

The Goal level of a log frame should be a long-term, high-level impact to which the project contributes — not something the project alone will achieve. "Eliminated poverty in Kwale County" is not a realistic project goal. "Contributed to improved livelihood resilience among 5,000 smallholder farmers in Kwale County" is appropriate — it acknowledges that your project contributes to this goal alongside other actors and enabling conditions.

7. Log frame inconsistency with other proposal sections

Your budget must be traceable to your activities. Your implementation timeline must align with your activities. Your monitoring plan must explain how you will collect data against your indicators. When a donor reviewer finds inconsistencies between the log frame and other sections, it signals poor project management and reduces confidence in your ability to implement.

Step-by-Step: Building a Winning Log Frame

Now to the practical guide. This is the approach our M&E experts at CCS use in our training and consulting work — adapted for Kenyan and East African contexts.

Step 1: Start with the problem analysis

Before you open the log frame matrix, you need a clear and evidence-based understanding of the problem you are addressing. The standard tool for this is a problem tree — a visual representation of a core problem, its causes (roots) and its effects (branches). From the problem tree, you develop an objective tree by converting problems into positive conditions.

This step is where most proposals fail before the log frame is even started. Teams jump to solutions without thoroughly diagnosing problems. The result is a log frame built on assumptions about causality that are not supported by evidence — and donors, increasingly, are sophisticated enough to spot this.

Step 2: Define your Goal (Impact)

The Goal is the long-term development impact to which your project will contribute. It should be broad enough to reflect why the project ultimately matters, but specific enough to be meaningful. Typical Goal statements reference reduced mortality, increased income, improved governance, enhanced food security — at a population level, over a multi-year horizon.

Good Goal: "Improved food and nutrition security among 50,000 smallholder farming households in Isiolo County by 2030."

Indicators at the Goal level should come from credible secondary sources — national surveys, sector data — since you are unlikely to have primary data collection systems that can measure impact at this level. State this clearly in your MoV column.

Step 3: Define your Purpose (Outcome)

The Purpose is the change in behaviour, practice or condition among your target group that your project will directly produce. It should be achievable within your project timeframe and directly attributable to your intervention.

Good Purpose: "1,200 smallholder farmers in Isiolo County adopt improved, climate-resilient agricultural practices by December 2027."

The Purpose indicator is the most important indicator in your log frame — it is the primary measure of whether your project worked. Invest the most time here. Make it SMART, make it measurable with your own data systems, and make sure it represents the change you are actually trying to create.

Step 4: Define your Outputs

Outputs are the concrete, tangible products of your project activities — the things you will directly produce and can most directly control. Think: number of people trained, number of tools distributed, number of farmer groups established, number of policies developed.

A well-structured project typically has three to five outputs. Each output should be directly and plausibly connected to the Purpose — you should be able to explain clearly how producing that output contributes to the purpose-level change.

Step 5: Define your Activities

Activities are the specific tasks and interventions the project will carry out to produce each output. They should be actionable, specific and budgetable. In most log frames, activities are listed under each output to maintain the logical connection.

Step 6: Develop SMART Indicators for each level

For each row of your log frame, you need at least one — ideally two or three — SMART indicators. The key formula for a strong indicator is:

[Number/Percentage] of [who] [does/achieves what] [by when] [as measured by what tool]

Example: "Percentage of trained farmers (n=1,200) who report adopting at least three improved agricultural practices in the most recent cropping season, as measured by a standardised post-harvest survey administered by field officers, increased from 0% at baseline to 70% by December 2027."

Step 7: Complete the Assumptions column

For each level of your log frame, identify the external conditions that must hold for the vertical logic to work. Assumptions should be:

• Specific enough to be monitored
• Realistic — likely enough to hold that the project is worth funding
• Outside your project's direct control (if you can control it, it should be in your activities)
• Important enough that if they do not hold, the project will fail

The Kenya-Specific Context: Common Donor Requirements

Kenyan NGOs most commonly work with the following major donors, each of whom has specific log frame preferences and requirements:

USAID: Requires results frameworks aligned to their standard indicators (standard foreign assistance indicators or SFAIs). Strong emphasis on gender disaggregation of all indicators. Increasingly requires open-source data collection tools (KoboToolbox, ODK).

EU: Uses the EU's specific LFA format with separate logical framework and indicators tables. Strong emphasis on cross-cutting issues (gender, environment, governance) integrated at every level. Detailed means of verification requirements.

DFAT (Australia) and FCDO (UK): Both emphasise adaptive management — log frames that can be revised based on learning — and increasingly require theories of change alongside log frames.

GIZ (Germany): Has its own DAC-compatible log frame format with strong emphasis on the causal logic between levels and a separate risk and assumption management plan.

From Log Frame to Monitoring System

A log frame is only useful if it becomes the foundation of your project's monitoring system. Every indicator in your log frame should have a corresponding data collection tool, a responsible person, a collection frequency and an analysis and reporting process. If you cannot describe how you will collect data against an indicator, that indicator does not belong in your log frame.

The tools most commonly used by Kenyan NGOs for primary data collection — KoboToolbox, ODK, CommCare — have made mobile data collection much more accessible. If you are not using digital data collection tools, you are missing an opportunity to strengthen your MoV credibility with donors who increasingly expect it.