Proposal Writing
Logic Model: Components, a Full Example, and How to Build One
Grant Writing Service Editorial Team
June 12, 2026 · 5 min read
Table of contents
Key takeaways
- Outputs count what you did; outcomes describe what changed for the people you served.
- Build the model backward from the outcome you want, then read it forward to test each link.
- Write down the assumptions and outside factors each arrow depends on; reviewers look for them.
- Every outcome in the model should reappear as an indicator in your evaluation plan.
A logic model is a one-page diagram that shows how a program is supposed to work: the inputs it uses, the activities it runs, the outputs those activities produce, the outcomes they create for participants, and the long-term impact they contribute to. Read left to right, it says "with these resources we do these things, which lead to these results." Funders ask for one because it exposes, at a glance, whether the activities you propose can plausibly produce the results you promise.
The five components, with the line that trips people up
| Component | Question it answers | Grant example |
|---|---|---|
| Inputs | What do we need? | Two part-time tutors, donated classroom, phonics curriculum, school partnership |
| Activities | What do we do? | Small-group tutoring, three sessions a week for 30 weeks |
| Outputs | How much did we do? | 120 students served, 90 tutoring hours delivered |
| Outcomes | What changed for participants? | Reading fluency scores rise; more students read at grade level |
| Impact | What larger change do we contribute to? | Higher third-grade reading proficiency across partner schools |
The line between outputs and outcomes is where most logic models break. "120 students tutored" is an output; it proves effort. "Reading fluency rose from fall to spring" is an outcome; it proves change. A model whose right-hand columns only restate activity counts tells a reviewer you will be busy, not that anything will improve.
Outcomes are usually split by time. Short-term outcomes are changes in knowledge, skills, or attitudes. Medium-term outcomes are changes in behavior or practice. Long-term outcomes are changes in condition or status. Keep impact modest: it is change you contribute to alongside many other forces, not change you can claim alone.
Most published templates, including the widely used W.K. Kellogg Foundation format, add three more elements around the five columns: the problem statement the program responds to, the assumptions behind each arrow (for example, that families can get children to the site), and external factors outside your control (a school schedule change, a new state assessment). Including them shows a reviewer you have thought about what could break the chain.
A complete logic model example
The example below is illustrative, built for a fictional nonprofit applying for a two-year grant to run an after-school literacy program.
Problem statement: In the two partner elementary schools, 62 percent of third graders scored below grade level on the state reading assessment last spring, and neither school offers structured reading support after school.
| Inputs | Activities | Outputs | Short-term outcomes | Medium-term outcomes | Long-term outcomes and impact |
|---|---|---|---|---|---|
| Half-time program coordinator | Recruit and screen second and third graders reading below grade level | 120 students enrolled each year | Students' oral reading fluency improves between fall and spring benchmarks | 55 percent of participants read at grade level by the end of the school year | Third-grade reading proficiency rises across both partner schools |
| Four trained part-time tutors | Small-group tutoring, groups of four, three times a week, 30 weeks | 90 sessions per group; 80 percent of students attend at least three times a week | Students report greater confidence reading aloud | Participants complete more independent reading at home | Fewer students referred for reading intervention in fourth grade |
| Classroom space donated by the school district | Monthly family reading nights | 8 family nights per year; 60 families attend at least one | Families learn three home-reading strategies | Families read with children at least three nights a week | |
| Phonics curriculum and leveled books | Coordinate with classroom teachers each quarter | 4 data-sharing meetings per year | Tutors adjust groupings using classroom data | ||
| Partnership agreement with two Title I schools |
Assumptions: students can stay after school because the district provides late buses; classroom teachers will share benchmark data; tutors stay for the full school year.
External factors: changes to the district calendar, a change in the state assessment, staff turnover at the partner schools.
Look at what the table makes checkable. The donated classroom appears as an input, so it must also appear in the budget as an in-kind contribution. Every activity leads to at least one outcome; the teacher coordination activity earns its place because it feeds tutors' grouping decisions. And every outcome is something an instrument can measure, which is exactly what the evaluation plan for the grant will need.
How to build one
- Start from the right. Write the long-term outcome first, then ask what has to change before it can happen. This backward pass is the same reasoning a theory of change makes explicit, and it stops you from listing activities you happen to run rather than activities the outcome requires.
- Write outcomes as changes. Each one should name who changes and in what direction: "participants' fluency scores rise," not "literacy support."
- Add the activities that produce each outcome. If an outcome has no activity feeding it, either add the activity or drop the outcome.
- List outputs as counts. Number of people, sessions, hours, products.
- List inputs last, and check them against the budget. Every resource in the model should be paid for or documented as in-kind; our guide to building the budget covers how to show both.
- Write the assumptions under the arrows. For each link, ask what would have to be true for it to hold.
- Read it forward, then backward. Forward: "with these inputs we run these activities, producing these outputs, leading to these outcomes." Backward: "to reach this outcome we need these outputs from these activities using these inputs." Any link that sounds like a leap needs a missing step.
If you want a blank structure to fill in, the free logic model template has the columns, assumption rows, and an example row already laid out.
How the logic model connects to the rest of the proposal
The model is a summary of arguments made in full elsewhere, so it has to agree with them:
- Objectives. Each medium-term outcome should match a measurable SMART objective with the same baseline, target, and date.
- Evaluation. Each outcome becomes an indicator, with a data source and collection schedule.
- Budget. Each input is a line item or a documented in-kind contribution.
- Narrative. The activities column is your methods section in compressed form.
Some federal programs score the model directly. The Department of Education's general selection criteria (34 CFR 75.210) include "the quality of the logic model or other conceptual framework underlying the proposed project, including how inputs are related to outcomes" as a factor a competition can use, so read the notice of funding opportunity to see whether yours does and what format it expects.
Logic models are one piece of a complete grant proposal; if you would rather have the model, narrative, and evaluation built together so they match, our nonprofit grant writers handle the full application.
