Three worked logframe examples - education, public health, and environment - each a complete 4x4 logical framework matrix with every cell filled in.
Reading about the logframe is easy; building one that holds is harder. Three worked logframe examples follow — education, public health, and environment — each a complete 4×4 matrix with the goal, purpose, outputs, activities, indicators, means of verification, and assumptions filled in. Copy the structure. But an example is a starting point, not a finish: what makes any logframe defensible is wiring its indicators to real data — covered in the full logframe guide.
By Unmesh Sheth · Founder & CEO, Sopact · Updated May 2026
A logframe example is a completed logical framework matrix from a real program type, with every cell filled. It is the fastest way to learn the tool: seeing how another program defined its purpose, wrote its indicators, and named its assumptions makes the structure concrete. The three examples below span the most-searched sectors — education, public health, and environment.
Logframe example, logframe sample, logical framework example — the same thing: a filled matrix to learn from and adapt. For the framework behind them, see the logframe guide; to draft your own, the logframe template and AI wizard.
A nonprofit runs a structured-literacy program across primary schools, aiming to lift Grade 4 reading proficiency in under-served districts.
| Level | Narrative summary | Indicator (OVI) | Means of verification | Assumptions |
|---|---|---|---|---|
| Goal | Improved literacy outcomes for children in under-served districts | Grade 4 reading proficiency in target schools rises 20% within 4 years | National standardized reading assessment | Schools stay open and staffed across the period |
| Purpose | Enrolled students read at grade level by year-end | 75% of enrolled students reach grade-appropriate fluency by program-year end, verified baseline to exit | Oral reading fluency assessment on persistent student IDs | Students attend regularly; home support is stable |
| Outputs | Students and teachers complete the literacy program | 600 students baselined and enrolled; 40 teachers certified in structured literacy | Enrollment records and teacher certification logs | Teachers stay in post through the school year |
| Activities | Deliver instruction, train teachers, assess progress | Structured-literacy sessions delivered weekly; teacher training completed by Month 3 | Session logs and training attendance | Materials and trained staff available on schedule |
A program improves antenatal care and skilled birth attendance to reduce preventable complications during pregnancy and childbirth.
| Level | Narrative summary | Indicator (OVI) | Means of verification | Assumptions |
|---|---|---|---|---|
| Goal | Reduced maternal and newborn complications in target districts | Under-5 stunting prevalence in target districts falls 15% within 4 years | District health management information system data | No major outbreak diverts clinic capacity |
| Purpose | Enrolled mothers adopt safe antenatal and feeding practices | 85% of enrolled mothers demonstrate correct complementary feeding at a 12-month follow-up | Direct observation against a standardized behavioral rubric | Household food supply is sufficient; mothers have decision authority |
| Outputs | Health workers trained; mothers enrolled and monitored | 200 community health workers certified by Month 6; 2,000 mothers enrolled with baseline | Training records and the prenatal monitoring register | Clinics remain functional and supplied |
| Activities | Train workers, run outreach, track appointments | Community outreach delivered; appointment tracking running by Month 2 | Outreach logs and the appointment system | Community trust and access to clinics hold |
Communities restore degraded land, pairing forest recovery with local livelihoods through nurseries, planting, and monitoring.
| Level | Narrative summary | Indicator (OVI) | Means of verification | Assumptions |
|---|---|---|---|---|
| Goal | Restored ecosystems and increased forest cover on community land | Forest cover in community-managed areas rises 25% within 5 years | Satellite GIS imagery, baseline to endline | Weather patterns stay within the normal range |
| Purpose | Planted areas survive and communities sustain them | 500 hectares reforested with a 70% seedling survival rate at 2 years | Field survival surveys on mapped plots | Local participation in maintenance is sustained |
| Outputs | Seedlings distributed; nurseries and rangers established | 100,000 seedlings distributed; 12 nurseries and 30 trained rangers in place | Distribution records and nursery logs | Seedlings sourced sustainably and on time |
| Activities | Map sites, set up nurseries, plant, and monitor | Site mapping completed; quarterly community monitoring running | Mapping records and quarterly monitoring reports | Land-use agreements with communities hold |
The three examples span different sectors, but the same three things make each one hold up. Copy these — not just the layout.
Who, by how much, by when, verified how. "Improved literacy" is not an indicator; "75% of enrolled students reach grade-level fluency by year-end" is. Each example's purpose-level OVI passes that test.
Not a phrase in a cell — a named, real source: a fluency assessment, a monitoring register, satellite imagery. If the instrument does not exist yet, the row is not verifiable.
Each assumption is something outside the project's control that must hold — and specific enough to track, like "clinics remain supplied" rather than "conditions stay favorable."
A copied example is still a static matrix. What makes a logframe defensible at evaluation is not the wording — it is whether each indicator is wired to live data. Take the structure from these examples, build yours as a living logframe, and draft it fast with the logframe template wizard.
A short masterclass on the Indicator Gap — why even a well-written logframe produces weak evaluation evidence when the matrix is filed instead of run. The reason an example is a starting point, not a finish.
Your Logical Framework Is Broken — the masterclass. The full argument is in the logframe guide.
A logframe example is a completed logical framework matrix from a real program type, with every cell filled — the goal, purpose, outputs, and activities, plus the indicators, means of verification, and assumptions for each. It is the fastest way to learn the tool, because seeing how another program defined its purpose and wrote its indicators makes the abstract 4×4 structure concrete. The terms logframe example, logframe sample, and logical framework example all describe the same thing.
This page gives three complete worked examples. In the education example, the goal is improved literacy, the purpose is students reading at grade level, the indicators name a 20% proficiency rise and a 75% fluency rate, and the means of verification are a national assessment and an oral fluency test. The public-health and environment examples follow the same pattern with different content. Each is a full 4×4 matrix you can read top to bottom.
An education logframe maps an instructional intervention to learner outcomes. In the example above, a school literacy program sets its goal as improved district literacy, verified by a national reading assessment; its purpose as enrolled students reading at grade level, verified by an oral fluency assessment on persistent student IDs; its outputs as certified teachers and baselined students; and its activities as weekly structured-literacy sessions. The assumptions cluster around attendance, teacher retention, and home support.
A public-health logframe maps a health intervention to behavior and outcome change. In the example above, a maternal health program sets its goal as reduced complications, verified by district health system data; its purpose as mothers adopting safe feeding practices, verified by direct observation against a behavioral rubric; its outputs as trained health workers and enrolled mothers; and its activities as outreach and appointment tracking. The killer assumption is whether household food supply and decision authority hold.
A good logframe sample is not the one with the neatest formatting. It is the one where every objectively verifiable indicator names a population, a magnitude, a timeframe, and an instrument; where every means of verification points to a real data source rather than a phrase; and where every assumption is external to the project and specific enough to monitor. A sample that fails those tests teaches the wrong habit, however tidy the table looks.
Use the example for its structure, not its content. Copy the four levels and four columns, then replace every cell with your own program's specifics — your population, your magnitudes, your instruments. Do not copy an indicator wholesale; an indicator that fits a literacy program will not fit yours. The fastest route is to start from a draft: the logframe template wizard builds the matrix from a description of your program, so you adapt rather than transcribe.
The three worked examples on this page — education, public health, and environment — cover the most common sectors. Donor agencies such as the EU, the World Bank, and UN bodies also publish sector logframe examples in their project guidelines. Whichever you study, read it against the test above: a real example earns its place by being verifiable, not by being well formatted.
This guide is educational and based on publicly available methodology from the logical framework approach. The examples are illustrative composites. Information current as of May 2026. To suggest a correction, email unmesh@sopact.com.
The full guide — the 4×4 matrix cell by cell, writing OVIs that survive evaluation, and how to run a living logframe.
The blank 4×4 template with a prompt for every cell — and an AI wizard that drafts the whole matrix from a description of your program.
The planning tool that feeds the logframe — how and why change happens, before it is compressed into the matrix.
A working session, not a demo. Describe your program and we will draft your 4×4 matrix live — modeled on the examples above, with every indicator wired to a real instrument. You leave with a logframe that is yours, not a copied template, and that is answerable at evaluation.
Live walkthrough · with Unmesh Sheth, Founder & CEO · bring a program to describe
