What is social impact?
Social impact concerns the effects an organization, activity or investment has on people and communities. Those effects can be positive or negative, intended or unintended, and distributed unevenly. A credible account describes who experiences the change, what changes and how the evidence supports the claim.
The term is used broadly in business, public services, philanthropy and evaluation. Define it for the decision at hand. A company might examine working conditions in its supply chain; a service provider might examine access and wellbeing; an investor might examine changes associated with portfolio activity.
Why social impact matters
An organization can deliver what it promised and still miss what people need. Counting services or spending shows activity, but it does not explain whether access improved, whether a benefit lasted or whether some people were left worse off.
Understanding these effects supports decisions about design, resource allocation and accountability. It can reveal trade-offs: a service may become more efficient for most users while becoming harder to access for a smaller group. Reporting only the average benefit hides that problem.
Social impact is relevant beyond nonprofit work. Employment conditions, product accessibility, partner practices and community effects can matter to a commercial organization even when they are not captured in revenue measures.
Outputs, outcomes and impact are different claims
| Claim | Illustrative example | Evidence needed |
|---|---|---|
| Output: delivery | A provider delivers 40 training sessions. | Delivery records establishing what took place. |
| Outcome: change | Participants demonstrate a relevant skill after training. | A suitable assessment, timing and clear account of who was assessed. |
| Impact: wider effects | Improved employment conditions over time. | Evidence about the wider change and the role of the intervention among other influences. |
OECD’s evaluation glossary describes impact in terms of significant higher-level effects, including positive and negative, intended and unintended effects. The distinction helps prevent an activity total from being presented as evidence of a wider change. Read the glossary.
Examples of social impact questions
| Setting | Useful question | What a count alone misses |
|---|---|---|
| Workforce development | Do participants enter work that meets their needs and remain in it? | Training completions do not establish job quality or stability. |
| Customer access | Can people with different needs use the service successfully? | Total users can hide barriers for particular groups. |
| Employee experience | Do changes in management practice improve people’s working experience? | Attendance at manager training is not evidence of behavior change. |
| Supply-chain partnerships | Do agreed practices improve conditions for people affected? | A submitted audit is evidence to examine, not a benefit by itself. |
| Community facilities | Who gains access, and can the facilities continue to serve them? | Visits do not reveal exclusion, maintenance problems or changes in experience. |
Choose the question before the metric. If the question is about accessibility, collecting only overall satisfaction will leave an important gap.
How to measure social impact
Begin with the people affected and the decision the evidence should support. Do not start with a long list of available metrics and assume more collection will produce a clearer answer.
- Describe the intended change: connect activities to outcomes in a theory of change and state important assumptions.
- Choose a manageable set of measures: include the extent of change, its distribution and relevant negative effects.
- Gather complementary evidence: use appropriate surveys, interviews, observations, service records and documents.
- Examine explanations: consider other influences and the limits of your comparison.
- Use and revisit the findings: record a decision, an owner and what evidence will be reviewed next.
Keep the theory of change available alongside the data. It explains why the measures matter and helps identify missing evidence.
Who experienced the change?
A total can conceal differences between locations, groups or periods. Ask who participated, who did not, who benefited and who faced a burden. Describe the basis for those comparisons and protect privacy when groups are small.
Collecting a person’s name is not always necessary. Use identifiable or pseudonymous records only where the question and permissions justify them. Aggregate reporting can be appropriate when the decision concerns a group and individual tracking is unnecessary.
Follow-up matters when the claim concerns sustained change. Report the number reached and the people whose status is unknown. Avoid treating a missing response as a negative outcome or a successful one.
Contribution is not the same as attribution
A change may be influenced by several organizations, economic conditions, personal circumstances and earlier experiences. A before-and-after comparison can describe movement but rarely settles how much one intervention caused.
Contribution analysis examines whether evidence supports the proposed explanation while considering other influences. It can help a team make a reasoned account without pretending it has isolated a precise causal share. See BetterEvaluation’s explanation.
Be explicit about the claim. “Participants reported improved confidence” is different from “the program caused improved confidence.” Choose an evaluation design that matches the strength of claim and the decision involved.
Community impact and societal impact
Community impact usually concerns effects in a particular place or group. Societal impact concerns wider systems or conditions. The boundary depends on the intervention rather than a fixed size threshold.
For a community facility, access, safety and the experience of local users may be immediate concerns. Wider social cohesion is a more demanding claim. It requires a clear definition and evidence beyond facility usage. Ask communities what matters rather than assuming that increased participation means the same thing to everyone.
What to include in a credible report
- The question, scope, people affected and reporting period.
- The intervention’s intended pathway and important assumptions.
- Sources, collection methods, definitions and limitations.
- Findings separated from interpretation and recommendations.
- Positive, negative, expected and unexpected effects where evidence supports them.
- A clear account of what will change and what remains uncertain.
Use a quotation to illuminate a finding, not to imply that one person represents everyone. Keep the original context available and obtain appropriate permission for public use.
Make evidence easier to revisit
When information arrives through surveys, partner files, interviews and operational documents, give it consistent dates, definitions and ownership. Retain the original source when correcting or interpreting a submission.
An AI-assisted workflow can help organize new qualitative evidence and prepare questions for review. People remain responsible for checking the interpretation, deciding what is material and making claims proportional to the evidence. The practical goal is better visibility and decisions, not a larger collection of dashboards.
Continue learning in the Academy
Use these existing Academy guides for the practical next step. They are suggested companion readings; follow each guide’s course navigation for the full sequence.
Use the practical guides to connect the model to a collection plan, review new evidence and decide what to improve.
Frequently asked questions
Is social impact only relevant to nonprofits?
No. Businesses, public services, investors and other organizations affect people through their products, employment practices, operations and relationships.
Can social impact be negative?
Yes. Include adverse and unintended effects rather than defining impact as positive by default.
How is social impact different from social responsibility?
Responsibility concerns commitments and conduct; impact concerns effects on people and communities. Responsible intentions do not establish beneficial results.
Do we need an SROI ratio?
Not necessarily. SROI is one approach to valuing outcomes. Use it when monetary valuation serves the decision and can be supported transparently.
Do we need software to begin?
No. Begin with a clear question, definitions and feasible collection. More complex reporting may justify systems that help maintain identity, source evidence, access and consistency over time.
