Skip to main content
Nonprofits

Impact Data for Nonprofits: How to Measure What Actually Matters to Funders

Funders are increasingly sophisticated about what 'impact' means — and counting outputs isn't enough. Here's how to build an impact measurement framework that satisfies both mission and funder requirements.

Key Takeaways
  • Outputs (meals served) are no longer enough — funders increasingly require short-term outcomes
  • A theory of change connects every metric you collect to a specific program element
  • Use validated instruments from your field rather than creating custom surveys from scratch
  • Collect contact information for longitudinal follow-up at enrollment, not at program completion
  • Combine data with narrative — data validates the story, narrative makes it human

The Output vs. Outcome Distinction

Most nonprofits report outputs: meals served, people trained, hours of programming delivered. These are countable, verifiable, and important. But funders increasingly want outcomes: did the meals address food insecurity? Did the training lead to employment? Did the programming change a behavior?

Understanding this distinction and building measurement systems for outcomes — not just outputs — is the most important shift in nonprofit data strategy of the last decade.

What Funders Want to See

Tier 1 (baseline expectation): Outputs with verified counts How many people did you serve? This must be verifiable — not estimated. Attendance sheets, enrollment records, service logs.

Tier 2 (increasingly expected): Short-term outcomes Did something change for participants as a direct result of your program? Pre/post surveys, skill assessments, knowledge tests.

Tier 3 (increasingly valued): Long-term outcomes Did the change persist? Follow-up surveys at 6 or 12 months, employment verification, housing stability records. This is the hardest data to collect and the most compelling to show.

Donor data you can trust — Sohovi validates contact records and flags duplicates automatically — try Sohovi free.

Tier 4 (rare but powerful): Comparative or control group data Did outcomes differ from a comparison group? This is the gold standard of impact evidence and typically requires research partnerships or significant investment.

Building a Theory of Change

Before you can measure impact, you need a theory of change: if we do X, it will lead to Y, which will lead to Z. Map out your:

  • Inputs (resources, funding, staff)
  • Activities (what you do)
  • Outputs (immediate products of those activities)
  • Short-term outcomes (what changes immediately)
  • Long-term outcomes (what changes over time)

Every metric you collect should tie back to a specific element of this model. If you're collecting data that doesn't connect to any part of your theory of change, ask why.

Practical Impact Data Collection

Pre/post surveys: The simplest way to measure short-term outcomes. Administer the same instrument before the program starts and immediately after. The difference is your measured outcome.

Validated instruments: Rather than creating your own survey, use a validated instrument from your field (PHQ-9 for depression, TABE for adult literacy, etc.). Validated instruments have established benchmarks so you can compare your results to broader populations.

Sohovi lets you set up validation rules for any column and instantly see which rows fall outside them — no code or SQL required.

Longitudinal follow-up: Email or phone surveys at 6 and 12 months post-program. Response rates are typically 15–30% — plan for attrition in your sample size calculations. Collect contact information for follow-up at enrollment, not at graduation.

Telling the Impact Story With Data

Data without narrative is a spreadsheet. Narrative without data is an anecdote. Combine them:

"87% of participants reported reduced food insecurity at 3 months (n=214 of 247 who completed the survey). Maria's story illustrates what that number represents in a person's daily life…"

The data validates the anecdote. The anecdote makes the data human.

Frequently Asked Questions

How do I do a pre/post survey if participants don't want to be surveyed?

Keep surveys short (5 questions or under), explain why you collect data and how it's used, make completion easy (paper, phone, QR code link), and consider a small incentive for participation. Response rates of 80%+ are achievable for in-person programs.

What's a validated instrument and where do I find one?

A validated instrument is a survey tool tested for reliability and validity in research settings. Sources: Annie E. Casey Foundation measurement library, Urban Institute's National Center for Charitable Statistics, your field's national associations, and academic journals in your sector.

How many people do I need to survey for results to be meaningful?

For descriptive reporting (what % of participants showed improvement), 30+ completions is a useful threshold. For statistical significance claims, consult a statistician. Many nonprofit program sizes (100–300 participants) are sufficient for credible descriptive outcome reporting.

Selva Santosh

Data quality, for people who ship

Selva writes practical guides on data quality, profiling, and governance to help teams ship better data.

Start for free

Stop guessing. Start knowing your data quality.

Sohovi profiles your datasets in minutes — surfacing completeness gaps, type mismatches, and duplicate patterns before they reach production.

No credit card required · Free forever plan