# How to Build Impact Indicators Without Exposing Beneficiaries | KEZ$

Canonical: https://kezs.com.br/en/learn/nonprofit-projects/how-to-build-impact-indicators-without-exposing-beneficiaries
Language: en-US

Learn to measure nonprofit results, agree on evidence with funders, and report how resources are used while protecting the people receiving support.

[Nonprofit projects](https://kezs.com.br/en/learn/nonprofit-projects)

# How to Build Impact Indicators Without Exposing Beneficiaries

Learn to measure nonprofit results, agree on evidence with funders, and report how resources are used while protecting the people receiving support.

KEZS team · Educational guide

[impact indicators](https://kezs.com.br/en/learn?tag=impact%20indicators)[nonprofit accountability](https://kezs.com.br/en/learn?tag=nonprofit%20accountability)[data protection](https://kezs.com.br/en/learn?tag=data%20protection)[nonprofit funding](https://kezs.com.br/en/learn?tag=nonprofit%20funding)[project evaluation](https://kezs.com.br/en/learn?tag=project%20evaluation)[social entrepreneurship](https://kezs.com.br/en/learn?tag=social%20entrepreneurship)[governance](https://kezs.com.br/en/learn?tag=governance)

[← Nonprofit projects](https://kezs.com.br/en/learn/nonprofit-projects)[1. Start with the decision the indicator must support](https://kezs.com.br/en/learn/nonprofit-projects/how-to-build-impact-indicators-without-exposing-beneficiaries#topic-1)[2. Distinguish activities, outcomes, and impact before choosing numbers](https://kezs.com.br/en/learn/nonprofit-projects/how-to-build-impact-indicators-without-exposing-beneficiaries#topic-2)[3. Define each indicator so another person can reproduce it](https://kezs.com.br/en/learn/nonprofit-projects/how-to-build-impact-indicators-without-exposing-beneficiaries#topic-3)[4. Design the minimum dataset around actual risks](https://kezs.com.br/en/learn/nonprofit-projects/how-to-build-impact-indicators-without-exposing-beneficiaries#topic-4)[5. Make evaluation participation understandable and free from pressure](https://kezs.com.br/en/learn/nonprofit-projects/how-to-build-impact-indicators-without-exposing-beneficiaries#topic-5)[6. A complete example: keep enrollment, completion, and response groups separate](https://kezs.com.br/en/learn/nonprofit-projects/how-to-build-impact-indicators-without-exposing-beneficiaries#topic-6)[7. Make missing responses visible](https://kezs.com.br/en/learn/nonprofit-projects/how-to-build-impact-indicators-without-exposing-beneficiaries#topic-7)[8. Describe contribution and uncertainty without promising causation](https://kezs.com.br/en/learn/nonprofit-projects/how-to-build-impact-indicators-without-exposing-beneficiaries#topic-8)[9. Connect costs and outcomes without putting a price on people’s dignity](https://kezs.com.br/en/learn/nonprofit-projects/how-to-build-impact-indicators-without-exposing-beneficiaries#topic-9)[10. Use cash flow to identify evidence requirements that threaten delivery](https://kezs.com.br/en/learn/nonprofit-projects/how-to-build-impact-indicators-without-exposing-beneficiaries#topic-10)[11. Produce verifiable evidence without handing over participant databases](https://kezs.com.br/en/learn/nonprofit-projects/how-to-build-impact-indicators-without-exposing-beneficiaries#topic-11)[12. Prevent aggregate tables from revealing individuals](https://kezs.com.br/en/learn/nonprofit-projects/how-to-build-impact-indicators-without-exposing-beneficiaries#topic-12)[13. Agree on evidence and conflicts before funding arrives](https://kezs.com.br/en/learn/nonprofit-projects/how-to-build-impact-indicators-without-exposing-beneficiaries#topic-13)[14. Assign responsibilities and prepare to correct mistakes](https://kezs.com.br/en/learn/nonprofit-projects/how-to-build-impact-indicators-without-exposing-beneficiaries#topic-14)[15. Use adverse scenarios to improve decisions without punishing respondents](https://kezs.com.br/en/learn/nonprofit-projects/how-to-build-impact-indicators-without-exposing-beneficiaries#topic-15)[16. Close the cycle by connecting purpose, spending, findings, and limits](https://kezs.com.br/en/learn/nonprofit-projects/how-to-build-impact-indicators-without-exposing-beneficiaries#topic-16)[Frequently asked questions](https://kezs.com.br/en/learn/nonprofit-projects/how-to-build-impact-indicators-without-exposing-beneficiaries#faq)[References](https://kezs.com.br/en/learn/nonprofit-projects/how-to-build-impact-indicators-without-exposing-beneficiaries#references)

What you will find 

When a business owner funds a nonprofit project, asking whether the money helped produce a meaningful change is legitimate. Answering does not require unrestricted access to beneficiaries’ lives. Reporting can demonstrate how money was spent, how well activities were delivered, and how results developed without publishing names, identity documents, addresses, or recognizable personal histories. Start by identifying the decision each piece of evidence will support.

01 

## Start with the decision the indicator must support

An indicator might inform whether to continue a workshop, change its schedule, or release a donation installment. Each decision requires different evidence. Spending calls for financial records and appropriate supporting documents. Learning may be better assessed through a practical exercise than a photograph. Understanding why people stay or leave may require a voluntary conversation about barriers, rather than sending an attendance list to the funder.

Collecting excessive information also creates operating costs: more storage, access controls, review work, and consequences if information leaks. It can undermine trust and distort answers. Someone who believes that admitting difficulty might jeopardize support may say what they think the organization wants to hear. Protecting participants therefore improves measurement itself; it is not an optional safeguard added after the report is written.

Express the management question in one sentence: we need to know whether this support enables a particular change so we can decide how to use the next installment. If nobody can explain how a requested field will inform a decision, reconsider collecting it. This discipline connects purpose, results, and accountability while preventing beneficiaries from becoming promotional material for decisions already made.

02 

## Distinguish activities, outcomes, and impact before choosing numbers

Imagine a nonprofit providing business management training to people running small businesses. Holding sessions is an activity; counting participants and guidance hours measures outputs. Checking whether participants begin recording costs examines an outcome. Claiming that training increased income requires additional analysis: prices, demand, health, seasonality, and other support also affect businesses. Each level answers a different question and supports a different strength of conclusion.

An explicit change pathway might be: accessible guidance helps people understand costs; understanding encourages regular records; records support pricing and purchasing decisions; better decisions may contribute to greater economic stability. These links remain project hypotheses, however plausible they seem. Evidence can strengthen them, but drawing arrows in a presentation does not establish that every link actually occurs.

Choose a main indicator at a point in that pathway that matters and can be observed within the available period. Measuring economic stability after only a few weeks may produce weak conclusions. Observing a management practice after several months may be more defensible. A supporting entrepreneur needs to accept this relationship between ambition and demonstrability, rather than demanding that an immediate output prove lasting transformation.

Record possible unintended effects as well. Someone may learn to calculate costs and discover that a business activity is unsustainable. Closing it could be a prudent decision, even though it reduces the number of active businesses. An evaluation that celebrates continuation alone would misclassify that case. The project’s purpose should guide interpretation, including when a sensible participant decision contradicts the story initially expected.

03 

## Define each indicator so another person can reproduce it

An indicator specification should include its name, management question, definition, numerator, denominator, reference period, source, and responsible person. Add collection frequency, limitations, and a rule for missing information. Expressions such as lives transformed or autonomy achieved leave too much room for incompatible interpretations. Without precise definitions, two analysts can obtain different numbers from the same records while both believing they followed the agreement.

The main outcome in this example is the proportion of participants reporting that they recorded business receipts and costs during the four weeks before interview. The question does not request revenue amounts, bank balances, or customer names. The specification explains what qualifies as a record, accepts paper and digital methods, and includes a brief comprehension check without requiring copies of personal business ledgers.

This measures a reported practice, with limitations involving memory and the desire to please. A check using a fictional situation can show whether someone understands recordkeeping, but cannot prove continuous use. Keep that distinction visible. If volunteers demonstrate their methods, report that sample separately with its own denominator. Observing some participants does not establish that everyone else used the same practice.

Specify when an indicator may be revised. If wording excludes a valid recordkeeping method, correct it and document the change. Do not silently alter criteria after seeing performance. Preserve previous definitions and explain which comparisons remain valid. This allows the funder to understand whether a movement reflects participant behavior, a revised question, or both, and prevents improvements in measurement from being mistaken for improvements in outcomes.

04 

## Design the minimum dataset around actual risks

Tracking individual change may require connecting an initial response with a later one. The analytical form does not therefore need names, identity numbers, and telephone numbers. Random codes can link responses while contacts remain in a separate operational database. Restrict the link between codes and people to staff who need it for follow-up, and give the evaluation dataset only the fields necessary for analysis.

Separation reduces exposure but does not automatically anonymize records. If a linking key exists, reidentification remains possible. Even without it, a small neighborhood, unusual occupation, and household composition may identify someone together. Financial hardship information can also cause embarrassment or discrimination, even when its legal classification differs from sensitive personal data. Assess risk through context and possible use, rather than labels alone.

Before collecting each field, ask which decision depends on it, who needs access, and what disclosure could cause. A travel-time range may be enough to investigate attendance, without a full address. Availability by morning, afternoon, or evening may support scheduling without recording a family’s detailed routine. Greater precision is justified only when its added usefulness is proportionate to the exposure it creates.

Treat free-text fields carefully. A question about difficulties may elicit accounts of violence, illness, or family conflict. Ask respondents not to name third parties and offer a separate channel for support requests. Prepare staff to handle serious disclosures: discovering a worrying situation does not authorize publication of the story and may require a protected referral outside the evaluation process.

05 

## Make evaluation participation understandable and free from pressure

Participants need to understand why questions are being asked, how information will be used, and what happens if they decline. A short explanation in everyday language, read aloud where helpful and available for later consultation, may work better than a long statement accepted without understanding. The organization must establish and document applicable data-processing grounds and conditions with suitable guidance. A signed form alone does not settle those questions.

Power imbalances matter. Someone depending on assistance may interpret an invitation as compulsory, especially when it comes from the person deciding whether support continues. Where possible, use an evaluator who does not control the benefit. Clearly distinguish information needed to deliver the service from optional research questions, rather than combining both under one broad authorization.

Agreeing to an interview does not authorize photographs, publication of a personal story, or commercial image use. Address those purposes separately. Refusing publicity should not affect access to assistance. Explain practical limits on withdrawal too: once an aggregate report has been published, retrospectively removing one person’s contribution may be impossible without identifying their response. Describe these limits before participation.

In Brazil, guidance from the ANPD, the national data protection authority, is a useful starting point for further questions, while the particular situation still requires assessment. The supporting entrepreneur should fund an understandable, respectful process rather than demand universal research participation. A lower response rate obtained without pressure may provide more trustworthy evidence than complete forms filled in because people fear losing support.

06 

## A complete example: keep enrollment, completion, and response groups separate

All figures throughout this example are hypothetical arithmetic illustrations. A nonprofit runs a six-month project for 120 adults with small businesses. All complete an initial interview; 100 begin activities, and 90 meet a completion criterion established beforehand. At the end, 84 of the original 120 answer a follow-up interview. Invitations include people who did not complete the project, so evaluation does not listen only to the most engaged participants.

Initially, 42 of 120 report keeping records of business receipts and costs over the preceding four weeks: 35%. To examine change among the same people, however, use the 84 who answered at both stages. Within that group, 30 reported the practice initially and 54 at follow-up. These are comparable counts. Placing 42 beside 54 without explaining the different denominators would obscure what changed.

The start rate is 100 divided by 120, or 83.3%. Completion is 90% of those who started and 75% of those enrolled. Follow-up coverage is 84 divided by 120, or 70%. Report each denominator: these measures describe actual access, retention, and evaluation coverage. None replaces the others, and a strong completion rate among starters does not erase losses before activities began.

Among the 84 people with paired responses, reported practice rises from 35.7% to 64.3%, approximately 28.6 percentage points. The net difference is 24 additional positive reports. This does not necessarily mean exactly 24 people started keeping records: some may have adopted the practice while others stopped. Without examining individual transitions, the evidence establishes a net change between two counts, not every pathway behind it.

07 

## Make missing responses visible

There are 36 people without a final interview. Do not assume they all improved or automatically classify them as failures. Missing responses may reflect changed telephone numbers, lack of time, dissatisfaction, economic hardship, or a decision not to participate. Report contact attempts and the authorized channels used, together with how staff avoided excessive persistence or exposing participants to other people during outreach.

Arithmetic bounds can show uncertainty without inventing answers. There are 54 positive final reports among 120 enrolled people. If none of the 36 missing respondents maintained the practice, the final proportion would be 45%. If all did, it would be 90 out of 120, or 75%. This 45%–75% range is not a statistical confidence interval; it shows possible extremes given the unobserved responses.

Compare respondents’ and nonrespondents’ initial characteristics internally, disclosing findings only with suitable protection. If missing responses are concentrated among people with greater reading difficulties, the final picture may be overly favorable. More telephone calls alone may not help. Oral interviews, different scheduling, or a shorter questionnaire may reach experiences the original instrument excluded, provided participation remains voluntary.

Keep nonresponse, incomplete responses, and nonapplicable questions distinct. Someone who closed a business may no longer fit a question about current recordkeeping but can still explain decisions and effects of support. Apply classification rules set before analysis and explain how they affect denominators. Relabeling missing people as nonapplicable merely to improve the final percentage hides obstacles and undermines the funder’s ability to assess the project.

08 

## Describe contribution and uncertainty without promising causation

The example shows an increase in reported practice within the followed group. Claiming that the project caused the entire increase goes further. Participants might have received guidance elsewhere or begun recording costs because suppliers required it. The initial interview itself may have drawn attention to recordkeeping. Without a suitable comparison, there is no basis for assigning the whole difference to training.

Contribution analysis combines the expected change pathway with delivery evidence, accounts of practical use, and alternative explanations. Check whether content was delivered, whether participants could practice, and where they say they used what they learned. These pieces can strengthen interpretation without eliminating other causes. Appropriate wording is that the findings are consistent with a contribution from the project, with remaining uncertainty stated clearly.

A comparative evaluation may be useful before substantial expansion. But comparing participants with people who never enrolled can introduce differences in motivation, resources, and availability. Withholding necessary assistance merely to improve research design would also be unacceptable. Choose methods according to the question, scale, ethical conditions, and implementation capacity, preferably with independent technical support when the decision warrants it.

Evidence requirements should match the decision. Changing workshop hours may need operational records and participant feedback. Committing substantial resources over several cycles requires stronger evidence that outcomes persist. Neither decision permits promises of higher income or financial returns. A useful report distinguishes what was observed, what remains uncertain, and which additional evidence would reduce that uncertainty enough to inform the next commitment.

09 

## Connect costs and outcomes without putting a price on people’s dignity

The hypothetical project budget is R$ 120,000. Executed spending comprises R$ 72,000 for activities and field staff, R$ 18,000 for materials and access support, R$ 12,000 for monitoring and information security, R$ 10,000 for coordination, and R$ 8,000 for logistical adaptations actually made. Together these equal R$ 120,000. These are incurred expenses, not a contingency reserve still available to spend.

Dividing total spending by 120 enrollments gives R$ 1,000 per enrolled person. Dividing by 90 completions gives approximately R$ 1,333.33 per completion. Dividing by 54 positive final reports gives approximately R$ 2,222.22 per observed positive report. The last figure is not a cost per causally attributable impact: it includes people already keeping records before the project and depends on interview coverage.

These calculations can help compare the organization’s own cycles when scope, quality, and criteria are comparable. A lower cost per completion might reflect efficiency, or exclusion of people needing more support. Serving geographically dispersed communities may increase travel costs without implying poor execution. Ask which conditions produced the cost rather than using the quotient to create a simplistic ranking of organizations.

Budget for monitoring and protection. Cutting access controls, training, or follow-up may make reporting cheaper while making evidence less reliable. Agree on an evidence standard the available resources can support. If a funder requests additional research, discuss who pays and which decision justifies it, while preserving enough money for the underlying service. A more elaborate evaluation is useful only if its information warrants its financial and participant burden.

10 

## Use cash flow to identify evidence requirements that threaten delivery

In the same hypothetical example, the R$ 120,000 donation arrives in two R$ 60,000 installments, scheduled for the first and fourth months. Monthly expenses are R$ 18,000, R$ 22,000, R$ 24,000, R$ 22,000, R$ 20,000, and R$ 14,000, totaling R$ 120,000. Spending reaches R$ 64,000 by the end of month three, leaving a R$ 4,000 shortfall before the next installment if the nonprofit relies on the first payment alone.

Suppose the nonprofit makes R$ 10,000 of its own unrestricted reserves available with internal authorization. After the first installment, cash is R$ 70,000. Closing balances for the first three months are R$ 52,000, R$ 30,000, and R$ 6,000. With the second installment arriving at the start of month four, subsequent closing balances are R$ 44,000, R$ 24,000, and R$ 10,000. The reserve returns to its initial amount at completion.

Now suppose the second installment is delayed by one month because the funder demands proof of an outcome measurable only at completion. Without that payment in month four, R$ 6,000 cannot cover R$ 22,000 in expenses: the shortfall is R$ 16,000. Demanding premature proof can interrupt activities or encourage manipulated reporting. The problem lies in disbursement design, not unwillingness to account for resources.

Instead, tie the intermediate installment to evidence available by month three: reconciled expenses, completed activities, working controls, and a response plan for identified difficulties. Measure final outcomes at the appropriate time. Any bridging loan requires separate analysis of obligations, costs, and repayment capacity. A conditional future donation is not automatically an eligible receivable; an existing eligible claim differs from hoped-for donations or future sales.

11 

## Produce verifiable evidence without handing over participant databases

Organize accountability through access layers. A public report can present methods, aggregate figures, financial execution, and limitations. The funder receives additional information required by the agreement and necessary for oversight. Operational staff retain identifiable records essential to service delivery. An authorized evaluator or auditor can examine underlying evidence in a controlled environment, with a defined scope and a record of the work performed.

To verify activities, combine schedules, delivery records, contracts, financial documents, and sample inspections. Photographs of faces are not the only evidence of attendance. An authorized visit can observe an activity without recording participants. Where documents contain personal information, mask unnecessary fields or use supervised review rather than sending complete copies through messaging services or email. Access should match the verification task.

Explain what each source proves. A payment record supports a particular transaction but does not establish workshop quality. Attendance supports presence, not learning. A personal account can illuminate how change happened without showing how common it was. Reliability comes from consistency across different kinds of records and a process for investigating discrepancies, rather than treating one document as proof of every claim.

Technology can support version control, access management, and file integrity; it cannot turn an inaccurate record into truth. Blockchain integrity likewise does not establish factual accuracy, legal title, reserve solvency, or payment guarantees. A false statement may remain perfectly intact. Do not place beneficiaries’ private documents on public blockchains: permanence and replication can magnify exposure that is difficult to correct or reverse.

12 

## Prevent aggregate tables from revealing individuals

Removing names is insufficient when groups are small. Reporting that the only person in a particular occupation within a community experienced difficulty can identify them. Combinations of age, neighborhood, and life history can do the same. Before publishing a breakdown, consider what someone familiar with the area could infer by combining it with outside knowledge. Identification risk extends beyond the visible columns.

An editorial policy can set a minimum group size for publication, but that is no absolute guarantee. In a hypothetical breakdown of eight participants, publishing a subgroup of five while hiding three may still reveal the hidden count through totals and other subtotals. Review the whole table and earlier publications together. Combining categories or suppressing an additional subtotal may be needed to prevent reconstruction.

Rounding, ranges, and broader categories can reduce risk when explained without making findings misleading. If figures are rounded, disclose that component values may not add to the displayed total. More detailed information may be necessary for an internal accessibility decision under restricted access. Service delivery, evaluation, funder oversight, and public communication do not all require the same level of detail.

Review qualitative accounts too. Changing a name leaves a rare occupation, distinctive sequence of events, or recognizable phrase intact. A synthesis of recurring patterns can preserve learning without reproducing one person’s story. Label any composite illustration assembled from different situations as an explanatory construction, never a real testimonial. If an essential detail still exposes someone, withholding the account may be the responsible editorial decision.

13 

## Agree on evidence and conflicts before funding arrives

A funder may request the full beneficiary list to verify authenticity while the nonprofit considers disclosure excessive. A vague appeal to trust does not resolve this. Identify the actual risk: duplicate enrollment, fictional activities, or diverted funds. Each calls for a specific check, potentially less intrusive than sharing the complete database. This makes the discussion about verifiable oversight rather than unrestricted access.

For duplication concerns, an independent verifier could test records and report inconsistencies without passing on names. Delivery concerns may be addressed through documents and visits. If a concrete obligation requires identifiable information, examine its scope, purpose, and conditions. A sponsor’s interest in personal stories does not by itself make those details necessary for oversight, and contractual requests still need scrutiny.

Address conflicts of interest explicitly. A supporting company might want contacts for marketing or to select participants as sales prospects. Those purposes differ from evaluation and must not be introduced silently. Equally, a team paid only for favorable results may feel pressure to exclude difficult cases. Make compensation, selection criteria, and the right to record unfavorable findings transparent, including relevant intermediary incentives or fees.

Agree on deliverables, review periods, access, evaluation expenses, and procedures for disagreements. Define who can stop a risky publication and how corrections will be communicated. The entrepreneur retains meaningful oversight while accepting that not every curiosity belongs in accountability reporting. The nonprofit commits to verifiable evidence and does not use privacy as a generic answer to every legitimate question about implementation or spending.

14 

## Assign responsibilities and prepare to correct mistakes

Even a small team should distinguish functions. Coordination defines the question and authorizes changes; service staff maintain contacts; evaluators analyze coded data; finance reconciles expenses; a designated person handles protection matters and participant requests. Individuals may hold several roles without having identical access. The final report should receive a review independent of its author’s analysis, even if another staff member performs it.

Check for errors that change decisions: duplicate codes, responses outside the reference period, incorrect denominators, and formulas omitting rows. In the example, dividing 54 positive reports by 90 completions produces 60% but mixes different groups. The paired indicator uses 84 as its denominator. A correction should replace the inaccurate figure, explain its origin, and identify whether a funding or operational decision was affected.

Maintain internally versioned records, change histories, and publication approvals without inserting unnecessary personal data into the audit trail. Set retention periods according to purpose, legitimate verification needs, and applicable obligations. Keeping everything indefinitely increases cost and risk; deleting too early may prevent service delivery or legitimate checks. Include exported files, copies, and documents on personal devices in the retention and deletion rules.

If information is sent improperly, prioritize containing access, preserving evidence needed to understand the event, and notifying those responsible for incident handling. Assess scope, affected people, appropriate measures, and any required communications. Deleting a sent message does not prove nobody copied it. Record confirmed facts and remaining uncertainties, neither minimizing exposure nor announcing consequences that have not been established.

15 

## Use adverse scenarios to improve decisions without punishing respondents

Suppose follow-up shows better recordkeeping alongside reports of overload among participants balancing paid work and family care. The positive indicator does not cancel the burden. Test shorter sessions or access support and monitor retention and quality. Funders should allow justified adjustments with documented expected effects. Preserving an original format solely for comparability can sacrifice usefulness and participation.

An economic downturn could reduce sales while participants improve their cost records. Both findings can be true. Do not automatically attribute falling income to the project, or use an external shock to explain away every weakness. Examine the sequence of events, compare sources, and state what the evidence supports. Selecting only the favorable indicator would conceal information needed to judge whether the intervention remains appropriate.

Convenient selection is another risk. Prioritizing people who already seem organized may raise final percentages while excluding those facing greater barriers. Document admission criteria, refusals, and offered adaptations, reporting these in aggregate. Performance targets must sit alongside an equity question: whom did the project intend to reach, and who could actually participate? A higher outcome rate can coexist with poorer access for the intended population.

Before implementation, agree on responses to warning signs: revise confusing questions, offer another channel when response is low, or suspend a publication that enables identification. These are decision rules, not promises of success. Unfavorable findings may justify redesign, smaller scale, or responsible closure. Accountability should help prevent purpose and reputation from sustaining an intervention that the evidence no longer supports.

16 

## Close the cycle by connecting purpose, spending, findings, and limits

Structure the final report around intended change, participants and admission criteria, executed spending, outputs, outcomes, gaps, and decisions. In the example, report 120 enrollments, 100 starts, 90 completions, and 84 follow-ups. Explain the rise from 30 to 54 positive reports among the same 84 people and the 36 missing interviews. Show total cost and the reserve needed to bridge the disbursement schedule.

Connect each conclusion to a proportionate action. Practical learning with retention difficulties may justify preserving content while improving access. Extensive missing follow-up may require better contact methods and questions before expansion. Give participants accessible feedback so they can recognize mistaken interpretations and understand how answers were used, without disclosing anyone else’s experience. Their feedback can identify errors that a numerical review misses.

KEZS connects entrepreneurial projects and potential funders through technology and intermediation. Project documentation can help organize purpose, evidence, and risks for analysis; KEZS does not assign a credit score or replace funders’ own diligence and decisions. Identify the instrument: nonprofit donations are not equity or promised financial returns. Loans, receivables, equity, and convertibles create different rights, obligations, costs, and loss risks. A MEI cannot admit equity partners while remaining a MEI.

Begin with a relevant question, a few reproducible indicators, minimum necessary data, an evaluation budget, and access rules. Record actual events and correct unsupported assumptions. Public securities distribution and regulated financial activities require appropriate structures and providers; technology or a CNAE business activity code is not permission. Guidance sources are reference entry points, not proof of KEZS authorization. Clear documentation guarantees neither approval, resources, repayment, profitability, liquidity, capital protection, nor social success.

FAQ 

## Frequently asked questions.

### Can a nonprofit report to funders without publishing beneficiaries’ names or photographs?

Yes. Financial reporting, activity records, aggregate indicators, and sample verification can support accountability. Match evidence to the claim: a payment, an output, and a participant outcome require different checks. If an identifiable document is genuinely necessary, authorized people can examine it under defined purposes and access conditions without reproducing it publicly. Photographs and personal stories need their own assessment and appropriate authorization; they do not substitute for outcome evidence. Restricted verification must still be sufficient to investigate inconsistencies rather than simply asking the funder to accept unsupported claims.

### Which main indicator should a nonprofit present to a supporting entrepreneur?

Choose according to the intended change, the decision being made, and the period in which change can reasonably be observed. For management training, an applied practice may be more informative than session counts and more defensible than a promised income increase. Define the numerator, denominator, reference period, source, missing-data rule, and limitations. Add access and delivery measures: a strong result among a small responding group does not describe overall reach. The aim is a useful decision, not the most impressive percentage or a comparison that ignores differences between projects.

### Why should the example’s 42 initial reports not be compared directly with its 54 final reports?

The counts come from different groups. The 42 initial reports relate to all 120 enrolled people, while the 54 final reports come from 84 follow-up respondents. The matched comparison is 30 versus 54 among those same 84 people: approximately 35.7% versus 64.3%, a rise of 28.6 percentage points. It shows net change in reported practice, without proving causation or identifying every individual transition. Disclosing the 36 missing responses prevents differences in coverage from being mistaken for behavioral change. The arithmetic bounds of 45%–75% for all enrollees are possible extremes, not a confidence interval.

### Can a funder make an installment conditional on an impact result?

Conditions need to fit the measurement schedule and allow implementation to continue. Demanding a final outcome midway through delivery can create an impossible dependency and a cash shortfall. Intermediate installments can instead depend on reconciled spending, verifiable outputs, functioning controls, and responses to identified problems. Reserve final outcome assessment for when observation is feasible. Address delays, disputed findings, and corrective measures in the agreement. A conditional donation should not automatically be treated as an eligible receivable, and any bridging credit needs its own repayment and cost analysis. An uncertain indicator is not a guarantee of social success.

### Does replacing names with codes make a dataset safe to share?

No. A key linking codes to people preserves the possibility of identification. Even without that key, combinations of occupation, location, household characteristics, or distinctive events may reveal someone to a reader familiar with the community. Assess purpose, necessity, fields, recipients, and opportunities to combine sources. Separate contacts from analysis, restrict access, and reduce detail, but still review the actual disclosure risk. An internal dataset may remain unsuitable for a sponsor or the public. Small aggregate tables can also reveal suppressed groups through subtraction, so review totals, subtotals, and earlier publications together.

### What should happen when the nonprofit and funder disagree about the findings?

First check definitions, denominators, versions, and sources, distinguishing calculation errors from interpretive differences. Then state alternative explanations and assess whether evidence can distinguish them. Independent technical review may help with consequential decisions. If uncertainty remains, present the main interpretation together with the limitation preventing a stronger conclusion. A funder should not demand a favorable narrative in exchange for resources, and a nonprofit should not invoke impact complexity to avoid verifiable questions. Agreed procedures should protect the ability to publish unfavorable findings and stop disclosures that could expose participants.

### Does organized documentation mean a project will obtain funding or generate financial returns?

No. It makes purpose, evidence, costs, and risks easier to examine, while funders conduct their own diligence and make their own decisions. Donations provide no equity or promised financial return; loans create repayment obligations; receivables depend on eligible existing claims; equity and convertibles involve distinct contractual and ownership risks. Future sales are not existing receivables, and a MEI cannot take equity partners while retaining MEI status. Regulated activities require appropriate structures and providers. Neither documentation, technology, a CNAE code, nor a favorable indicator guarantees authorization, approval, capital availability, repayment, liquidity, profitability, or protection against loss.

## References and context

Institutional reference points for checking applicable rules and guidance. Calculations and scenarios are original examples, not quotes, historical results or individualized recommendations.
- [Banco Central — correspondentes no País ↗](https://www.bcb.gov.br/estabilidadefinanceira/exibenormativo?tipo=Resolu%C3%A7%C3%A3o%20CMN&numero=4935)
- [CVM — Resolução 88, texto e alterações ↗](https://conteudo.cvm.gov.br/legislacao/resolucoes/resol088.html)
- [CVM — Resolução 175, fundos de investimento ↗](https://conteudo.cvm.gov.br/legislacao/resolucoes/resol175.html)
- [Portal do Empreendedor — perguntas sobre MEI ↗](https://www.gov.br/empresas-e-negocios/pt-br/empreendedor/perguntas-frequentes)
- [ANPD — orientações e perguntas frequentes ↗](https://www.gov.br/anpd/pt-br/acesso-a-informacao/perguntas-frequentes)

AI-assisted educational content. Decisions require verification of the information, instruments and providers in each actual case.

## Keep learning.[Explore the category ↗](https://kezs.com.br/en/learn/nonprofit-projects)

[

Nonprofit projects 

### Nonprofit budgeting: connect resources to social outcomes

Build a nonprofit project budget that connects full costs, social goals, cash flow, evidence, and clear responsibilities for funding and delivery.Explore the article ↗  ](https://kezs.com.br/en/learn/nonprofit-projects/nonprofit-budgeting-connect-resources-to-social-outcomes)

[

Nonprofit projects 

### Street Dog Rescue: Planning Shelter, Care and Responsible Adoption

Plan a dog rescue project with realistic shelter capacity, a complete care budget, responsible adoption and transparent reporting on funding and impact.Explore the article ↗  ](https://kezs.com.br/en/learn/nonprofit-projects/street-dog-rescue-planning-shelter-care-and-responsible-adoption)

[

Nonprofit projects 

### Donations, Sponsorships and Loans: Different Ways to Fund a Nonprofit Project

Understand how donations, sponsorships and loans affect rights, cash flow and accountability when a business supports a nonprofit project in Brazil.Explore the article ↗  ](https://kezs.com.br/en/learn/nonprofit-projects/donations-sponsorships-and-loans-different-ways-to-fund-a-nonprofit-project)

From knowledge to decisions 

## Plan your next step with more clarity.[Get started ↗](https://kezs.net/?lang=en)