A nonprofit evaluation plan connects the question you need answered to the evidence you will collect and the decision you will make. It specifies measures, design, responsibilities, timing, and limitations before everyone starts collecting different versions of the same information.
Start with use: who will read the findings, and what will they do with them? A program manager deciding how to improve delivery may need different evidence from a funder deciding whether to support expansion.
Free resource: Open the printable evaluation-plan example and blank worksheet, or download the HTML worksheet. No email required. Open the file in a browser to print it or save it as a PDF.
Describe who the program serves, what it provides, and how those activities are expected to lead to change. A logic model can help make that connection explicit. Explain the assumptions behind it, including conditions outside the program that may affect outcomes.
Write a concrete decision alongside that description: for example, whether to revise the follow-up component before the next cohort. Avoid promising an evaluation that answers every question about the organization.
Include delivery questions when implementation is uncertain and outcome questions when you need to understand participant results. Treat a claim that the program caused an outcome as a separate question requiring an appropriate design.
Discuss priorities with staff, participants or their representatives, and the people who will use the report. Ask what they need to know and what collection would burden participants. Select a small, useful set of questions rather than collecting everything that might be interesting.
For each question, specify the measure, source, collection date, and meaning of its values. Define denominators: completion among enrolled participants is different from response among people contacted for follow-up.
Choose measures that fit the population and purpose. Satisfaction describes an experience; it does not establish that an employment or learning outcome improved. Our guide to choosing evaluation outcomes covers this decision in more detail.
Monitoring records can describe reach and delivery. Baseline and follow-up measures can describe observed change. A causal claim needs a design that addresses plausible alternative explanations.
If there is no credible comparison, explain what remains uncertain: outside events, selection, or other services may contribute to changes. If you need evidence of impact, assess comparison options early. See when a program needs a control group.
Name who collects each measure, who checks it, and who can access it. Agree on participant identifiers, consent where applicable, secure storage, retention, and transfers. Specify how potentially identifying small groups and quotations will be handled in reports.
Plan follow-up before participants leave. Keep enrollment, service participation, and outcome follow-up separate: stopping attendance does not necessarily mean losing outcome information. Record missingness reasons where feasible and plan how they will affect interpretation.
Budget for your team's work as well as external support. Our nonprofit evaluation-cost guide explains the items to include and the assumptions to state.
Describe planned summaries and comparisons in plain language. Include missing-data handling, uncertainty, and any subgroup analyses. State the reporting products, audience, and dates. Identify who will discuss the findings and assign next actions.
Treat the plan as a working agreement. Document changes in implementation, measures, or analysis and explain how they affect the conclusions.
This example is invented for planning purposes. It contains no client results, validated instrument, or sample-size recommendation. Enrollment and timing are illustrative.
An organization plans an eight-week employment-support course for about 60 adults. Staff want to improve delivery and understand participants' employment status three months later. They have no comparison group, so this evaluation will describe implementation and outcomes without estimating the causal effect of the course.
The follow-up lead contacts everyone enrolled, including people who stopped attending. The report distinguishes enrollment, attendance, contact attempts, and response. Employment transitions are described for people with both measurements, alongside available baseline information about those missing follow-up.
The team will interpret those findings with their limitations. Differences between respondents and nonrespondents may indicate bias; similar baseline characteristics do not prove that missing outcomes are harmless. Follow-up changes also cannot be attributed to the course without addressing what would have happened otherwise.
Staff use the delivery and feedback findings to revise the next cohort. If the funder later needs evidence of causal impact, the organization commissions a separate design and feasibility assessment.
The printable example and worksheet includes a question-to-measure matrix, data responsibilities, analysis and reporting prompts, and a budget worksheet. Adapt the questions and measures to your program rather than copying the example's enrollment or schedule.
If you need a review of whether the evidence matches your funder's expectations, DASS can help develop or review your evaluation plan.
One email whenever we publish something new. No spam, unsubscribe anytime.
{{ subscribeErrorMsg }}