Sopact is a technology based social enterprise committed to helping organizations measure impact by directly involving their stakeholders.
Copyright 2015-2026 © sopact. All rights reserved.
What a logframe is, the 4x4 logical framework matrix explained cell by cell, how to write OVIs that survive evaluation, and how to run a living logframe.
A logframe (logical framework) is a four-row matrix — goal, purpose, outputs, and activities — with columns for indicators, means of verification, and assumptions, used to plan and appraise a program. It is the appraisal format most widely used in international development, required in some form by USAID, the UN agencies, and the World Bank.
The logframe answers two questions in one grid: what will this program achieve at each level, and how will anyone know. The columns turn each row into something measurable — an indicator, a source that verifies it, and the external condition it depends on. That last column is where Sopact focuses, because it is the part that decides whether the plan holds and the part teams almost never revisit.
Used by: M&E and MEL leads, program managers writing donor proposals, foundation appraisal officers, and evaluators who inherit a logframe pasted into a grant agreement and have to make its indicators produce real data.
Here is the quiet failure inside almost every logframe. The four rows and the indicator column get careful attention; the assumptions column gets filled in the last hour before the proposal is due and is never opened again. Yet the assumptions column is the one that decides whether the vertical logic actually works — it names every external condition that has to hold for an activity to produce an output, an output a purpose, a purpose the goal.
When an assumption breaks — a partner withdraws, a policy shifts, demand collapses — a filed logframe says nothing. The indicators still read as planned because no one is watching the condition beneath them, and the program discovers the broken assumption only in the final evaluation, when the money is spent. A logframe that lists assumptions but never monitors them is an appraisal document, not a management tool.
That is the shift Sopact is built around: keep the logframe live. Every indicator is wired to data on a persistent stakeholder ID, and every assumption becomes a monitored condition with a signal that fires while there is still time to act — instead of a matrix pasted into a proposal and filed. For the broader practice this sits inside, see monitoring and evaluation and impact measurement & management.
A logframe reads as a grid. The four rows are a hierarchy of objectives. Goal is the long-term change the program contributes to. Purpose (or outcome) is the change the program is directly accountable for. Outputs are the deliverables the program produces. Activities are the tasks that produce those outputs. Read bottom to top, each level is the means to the one above it.
Three columns make each row testable. The indicator (objectively verifiable indicator) states what changes, in whom, by how much, by when. The means of verification names where the evidence comes from — the survey, record, or instrument that produces the indicator's value. The assumption names the external condition that must hold for that row to lead to the row above it. A fourth column, inputs or resources, appears in some variants at the activity row.
Two logics run through the grid. The vertical logic is the if-then chain: if the activities are done and the assumptions hold, then the outputs follow; if the outputs are delivered and their assumptions hold, then the purpose is achieved, and so on up to the goal. The horizontal logic is the measurement: for each row, the indicator plus its means of verification prove whether that level was reached. Sopact keeps both live — the vertical logic monitored through the assumptions column, the horizontal logic wired to real data on one ID.
Watch — the logical framework matrix explained. How the four rows and three columns fit together, why the assumptions column carries the vertical logic, and how to keep a logframe connected to participant data instead of filing it with the proposal. Presented by Unmesh Sheth.
These four formats are constantly confused, and treating them as rivals leads to rebuilding the same content four times. They are layers, not alternatives. A theory of change supplies the causal reasoning — the mechanism on every arrow and the assumption behind every link, the why a program should work. The logframe is the appraisal matrix — it takes that reasoning and lays it out as goal, purpose, outputs, and activities with indicators, verification, and assumptions, the format a donor appraises.
A logic model is the operational cousin: inputs, activities, outputs, outcomes, impact, read left to right, describing what a program does without the assumptions column that makes a logframe an appraisal tool. A results framework orders the results hierarchy — goal, intermediate results, sub-results — and is the format USAID pairs with a logframe. The clean rule: build the theory of change first for the reasoning, derive the logframe as the appraisal matrix, and switch to a results framework or logic model when a funder asks for that view. The detailed contrast between the first two layers is in the theory of change vs logic model guide, and a filled reference grid is in the logframe example guide.
Because all four derive from the same underlying logic, Sopact builds it once and presents it four ways rather than maintaining four documents by hand. Indicators can then map to a funder taxonomy like IRIS+ as a translation layer, not a rebuild.
A logframe earns its keep at four moments — building the goal-to-activities matrix, attaching an indicator and a means of verification to each row, monitoring every assumption on one persistent ID, and keeping it live for the funder. The animation below runs the loop; the four prompts under it are the ones behind each job.
1 · Build the matrix. Draft goal, purpose, outputs, and activities from the program document, with an assumption named under every level. The walkthrough is in how to build a logframe.
Academy walkthrough → How to build a logframe
Build a logframe from this program description: [PROGRAM URL OR DOC]. Fill the four rows — goal, purpose, outputs, and activities — and for every row write the objectively verifiable indicator, the means of verification, and the assumption that must hold for that level to lead to the one above it. Flag any row with a missing assumption or an unverifiable indicator.
2 · Review the logic. Pressure-test the vertical if-then chain and the horizontal indicator-verification pairing before a donor does.
Academy walkthrough → How to review a logframe
Review this logframe: [PASTE OR LINK]. Check the vertical logic (does each level plus its assumption actually lead to the level above?) and the horizontal logic (does every indicator have a real means of verification?). Flag every unmeasurable indicator, every missing means of verification, and every assumption that is stated but has no way to be monitored.
3 · Add the causal reasoning. A logframe appraises; derive it from a theory of change so the mechanism behind each level is on record.
Academy walkthrough → How to build a theory of change
From this logframe: [PASTE OR LINK], derive the theory of change beneath it: name the mechanism on each vertical link (why this level should produce the one above) and expand the assumptions column into the conditions each link depends on. Return the causal version the logframe summarizes.
4 · Convert to the funder's hierarchy. Turn the same logic into the results framework a donor like USAID asks for alongside the logframe.
Academy walkthrough → How to build a results framework
From this logframe: [PASTE OR LINK], produce a results framework: order the goal, intermediate results, and sub-results as a hierarchy, keeping every indicator and assumption consistent with the logframe rather than rebuilding from scratch. Note where the two formats disagree.
The sections above are the argument; the Academy articles are the practice — each a hands-on companion written to run on your own data.
A logframe (logical framework) is a four-row matrix — goal, purpose, outputs, and activities — with columns for indicators, means of verification, and assumptions, used to plan and appraise a program. It is the standard appraisal format in international development, required in some form by USAID, the UN, and the World Bank. In Sopact a logframe is kept live — every indicator wired to data on a persistent ID and every assumption monitored — rather than pasted into a proposal and filed.
A logic model reads left to right — inputs, activities, outputs, outcomes, impact — and describes what a program does. A logframe reads as a matrix with a hierarchy of objectives (goal, purpose, outputs, activities) and adds two things a logic model lacks: a means-of-verification column and an assumptions column, which make it an appraisal tool rather than a description. Sopact builds both from the same underlying logic so you switch views instead of maintaining two documents.
The four rows are goal, purpose, outputs, and activities. The goal is the long-term change the program contributes to; the purpose (or outcome) is the change the program is directly accountable for; outputs are the deliverables it produces; activities are the tasks that produce those outputs. Read bottom to top, each row is the means to the one above it — the vertical logic that Sopact keeps monitored through the assumptions column.
The three standard columns are the indicator, the means of verification, and the assumption. The indicator (objectively verifiable indicator) states what changes, in whom, by how much, by when. The means of verification names the source of the evidence — the survey, record, or instrument that produces the indicator's value. The assumption names the external condition that must hold. In Sopact the means of verification is a live data source on one persistent ID, not a promise to collect data later.
Assumptions are the external conditions that must hold for one level of the logframe to lead to the next — for the activities to produce the outputs, the outputs to deliver the purpose, and so on up to the goal. They carry the vertical logic of the whole matrix, yet most teams fill the assumptions column once and never revisit it. Sopact turns each assumption into a monitored condition that fires a signal when it breaks, while there is still time to act, instead of surfacing in the final evaluation.
A theory of change supplies the causal reasoning — the mechanism on every arrow and the assumption behind every link, the why a program should work. A logframe is the appraisal matrix that takes that reasoning and lays it out as goal, purpose, outputs, and activities with indicators, verification, and assumptions. They are layers, not rivals: build the theory of change first for the reasoning, then derive the logframe as the format a donor appraises. Sopact derives both from one source.
Start from the program's objectives: state the goal and purpose, then the outputs that deliver the purpose, then the activities that produce the outputs. For each row, write the objectively verifiable indicator, name the means of verification, and state the assumption the level depends on. Check the vertical if-then logic and the horizontal indicator-verification pairing. Sopact drafts the four rows from a program document and attaches an indicator and a monitored assumption to each, so the logframe starts live rather than filed.
A logframe is a matrix that appraises a single program level by level, with an assumptions column. A results framework orders a hierarchy of results — goal, intermediate results, sub-results — and is often used across a portfolio; USAID pairs the two, using the results framework for the hierarchy and the logframe for the appraisal detail. Because both derive from the same logic, Sopact generates each from one source rather than rebuilding by hand.