A clear evaluation plan signals rigor and sets up strong grant reporting. Learn how to write an evaluation plan, what funders want to see, and pitfalls to avoid.

2025–2026 STATUS UPDATE: The Institute of Education Sciences (IES), the federal education-research arm at the U.S. Department of Education, was largely dismantled in 2025. Staff dropped from ~175 to fewer than 20; ~$900M in research contracts were terminated. Lawsuits are pending. References to IES-operated resources (including the What Works Clearinghouse) and IES grant programs may be unavailable or significantly diminished. Verify at ies.ed.gov.

The evaluation plan is the section of a grant proposal that quietly tells the funder how serious you are. A vague evaluation plan signals “we’ll figure it out later.” A clear, well-designed one signals an organization that knows what success looks like and how to prove it.

Funders increasingly weight evaluation plans heavily in scoring, especially federal agencies like HRSA, CDC, ED, NSF/NIH, and DOJ. And once funded, the evaluation plan is the framework you’ll use to actually report and improve the work.

This guide covers how to write an evaluation plan that earns confidence and is actually usable.

TL;DR: Quick Answers

What is an evaluation plan? The proposal section that explains how you’ll measure and learn from the project, what you’ll measure, how, and what you’ll do with the results.
What does it need to include? Evaluation questions, indicators and measures, data collection methods, analysis approach, and how you’ll use findings.
What’s the difference between process and outcome evaluation? Process evaluation tracks implementation (what happened, how, with whom). Outcome evaluation tracks results (what changed).
How does it tie to the rest of the proposal? Tightly, your SMART goals, logic model, and methods all show up in the evaluation plan.

What an Evaluation Plan Needs to Cover

A solid evaluation plan answers five core questions:

1. What are the evaluation questions? What do you want to learn about the project? Common questions include: Did we implement what we planned? Did participants change? Why or why not? Who benefited most?

2. What will you measure? What specific indicators tell you whether the project worked? These should map directly to your SMART goals and logic model.

3. How will you collect data? Surveys, interviews, observations, administrative records, pre/post tests, focus groups. Match the method to the question.

4. How will you analyze it? Quantitative, qualitative, mixed methods, and what comparisons you’ll make.

5. How will you use findings? Mid-course adjustments, reporting to funders, dissemination, organizational learning.

Process vs. Outcome Evaluation

A common framework worth knowing:

Process (implementation) evaluation answers “Did we do what we said we’d do, and how?” It tracks reach (who was served), dosage (how much), fidelity (did we deliver the model as designed?), and quality.

Outcome evaluation answers “Did it work?” It measures changes in knowledge, skills, behavior, conditions, or status, the outcomes in your logic model.

Most grants want both. Process data tells the funder what you actually did; outcome data tells them what changed. Reporting only outputs (“we held 12 workshops”) without outcomes (“85% of participants demonstrated skill gain”) falls short of what funders increasingly expect.

Choose Defensible Measures

A few rules:

Use established measures where they exist. Validated instruments and standard indicators carry more credibility than homegrown ones. Cite them.
Set baselines. “30% improvement” only matters compared to a starting point. Show pre-intervention measurement.
Match measures to the change you’re claiming. If you promise improved health, measure health, not just attendance.
Be realistic about what you can measure. A small nonprofit can’t run a randomized controlled trial. A well-designed pre/post survey with comparison can still be powerful.
Disaggregate where appropriate. Equity-focused funders want to see whether outcomes vary across population subgroups.

Common Designs

Pre/post design. Measure participants before and after the intervention.
Pre/post with comparison group. Compare participants to a similar non-participant group, much stronger evidence.
Time-series. Multiple measurements over time to track trajectories.
Mixed methods. Pair quantitative outcomes with qualitative interviews to explain why.
Implementation studies. Track fidelity, reach, and adaptation.

You don’t need every design. You need the simplest design that credibly answers the questions you’re asking.

The Role of an External Evaluator

For larger grants, an external evaluator strengthens credibility and often performs the design and analysis. Federal grants frequently expect (or require) one. Smaller grants may not. Build evaluator costs into the budget where appropriate.

How to Avoid Common Mistakes

Vague measures. “Improved outcomes” is not a measure. Name the indicator, instrument, and target.
Disconnected from goals and logic model. Every SMART goal should have at least one measure in the evaluation plan, and every outcome in the logic model should be measurable.
Overpromised rigor. Claiming RCT-level evidence on a small grant is unrealistic. Match the design to the resources.
No use of findings. A plan that collects data but never describes how it will be used is a weak plan. Show how findings inform adaptation, reporting, and learning.
Skipping process measures. Funders care about whether you implemented as designed, not just whether outcomes shifted.

How Evaluation Connects to Reporting

A well-built evaluation plan is the structure of your future grant reports. Reviewers know this; they’re looking at the evaluation plan partly through the lens of “will this organization report meaningfully to us later?” Writing a strong evaluation plan is an investment that pays off across the entire grant lifecycle.

How Grantboost Helps With Evaluation Sections

Evaluation plans must align with goals, the logic model, and the budget. Drift kills credibility. Grantboost helps you develop the proposal as one coherent piece in your organization’s voice (see training AI on your past proposals), structured around the funder’s specific evaluation requirements, so the evaluation plan, goals, logic model, and budget tell one connected story.

Try Grantboost free and build proposals with evaluation plans funders trust.

How to Write an Evaluation Plan for Your Grant Proposal