An impact investment agreement must answer two separate questions: what can be required of the recipient, and what change justifies financing the project? A business may be able to repay while generating little additional impact. An organization may deliver substantial social benefits without generating cash to return funding. Purpose, financial capacity and legal obligations therefore require separate analysis. Evidence supporting one does not establish the others.
The first right is to understand the relationship being agreed
A donation transfers resources without, by itself, creating a right to financial returns. Sponsorship may include brand exposure and communication commitments. A grant depends on its source, terms and applicable rules; it does not automatically become debt or equity. Financing that seeks a financial return must identify both the payment source and the supporting right, such as loan repayment, collection of a receivable, profit distributions or sale of an ownership interest. Calling every arrangement an investment does not give them equivalent consequences.
Before negotiating veto powers, write a statement that can be checked: who provides the money, who receives it, for what purpose, through which instrument and subject to which obligations. Identify the service provider and the party responsible for operations. They may differ where a company and a nonprofit collaborate. Documents must show that separation and the flows between them. Otherwise, rights may be negotiated against an entity that controls neither the funds nor the operation or expected revenue.
Legal form determines which rights can be offered
A Brazilian nonprofit association or foundation cannot offer ownership interests in the same way as a limited liability company, or LTDA. Supporting an NGO does not make the funder an owner of part of its assets or authorize distributions of surplus to donors. Requiring a percentage of financial results while calling the transfer a donation creates a contradiction to resolve before funding. The economic intention must match a legally appropriate instrument compatible with the recipient.
Equity, loans and convertible instruments also create different positions in a business. An equity holder takes business risk and exercises rights under the corporate structure and governing documents. A creditor holds a payment obligation, subject to the agreed terms and practical ability to collect. A convertible instrument guarantees neither conversion nor a buyer or liquidity. A Brazilian individual microentrepreneur, or MEI, cannot admit equity partners while remaining a MEI; an equity transaction requires addressing that incompatibility first.
Diligence should examine formation documents, authorized representatives, approval powers and restrictions on the use of funds. A signature cannot cure a signer's lack of authority. Confirm whether projected revenue belongs to the funded entity or another organization. If an NGO supports participants while a company sells services, the company's cash does not automatically belong to the NGO. Economic arrangements between them need a proper basis, defined scope, pricing where relevant and procedures for conflicts of interest.
Translate the impact thesis into verifiable commitments
An impact thesis suitable for an agreement identifies the problem, target population, intervention and expected change. Financing economic inclusion is a broad intention. Financing training and job placement support for people facing identified barriers allows clearer responsibilities. Enrollment, course completion and continued employment remain different measures. The agreement should distinguish activities the operator controls from outcomes influenced by employers, economic conditions and participants' decisions.
Additionality asks what the funding enables beyond what would probably happen without it. Financial additionality may arise when support enables an operation unable to obtain suitable financing elsewhere. Impact additionality may arise when an intervention expands or improves outcomes. Neither follows automatically from a social purpose. A convincing proposal identifies the constraint, alternatives considered and specific difference the funding should make, while acknowledging evidence that could weaken that interpretation.
Record targets with their assumptions, measurement method, observation period and responsible parties. Separate delivery obligations from indicators used for evaluation and learning. Missing a social target is not automatically fraud or breach: an underlying hypothesis may have failed. Withholding information, diverting funds and manipulating indicators require distinct responses. This distinction supports accountability for integrity without encouraging teams to conceal difficulties to preserve funding.
A worked example: two entities with separate funding
Consider an entirely hypothetical vocational training project for 120 people, with a total budget of R$ 300,000. An NGO handles outreach, participant support and evaluation; a company delivers training and runs the commercial operation. A supporter donates R$ 120,000 to the NGO and an investor lends R$ 180,000 to the company. Even if one person fills both roles, the instruments remain separate. The donation buys no ownership and is excluded from loan repayment calculations.
The NGO allocates R$ 36,000 to outreach and selection, R$ 48,000 to participant support, R$ 24,000 to monitoring and evaluation, and R$ 12,000 to administration. The company allocates R$ 72,000 to equipment, R$ 60,000 to training delivery, R$ 30,000 to working capital, and R$ 18,000 to a cash reserve. Each budget adds up to its funding amount. These categories authorize uses; they do not prove that spending occurred, suppliers delivered or prices are reasonable.
A cooperation agreement defines deliverables, necessary information sharing and responsibilities toward participants. Each entity keeps records tracing its own allocation. For simplicity, the donation is not transferred to the company, neither entity guarantees the other's obligations, and the NGO has no duty to repay the company's debt. Commercial revenue is the company's projected repayment source. The impact hypothesis is 120 participants, 90 completers and 54 verified job placements under a defined protocol. Here, all 54 placements involve people among the 90 completers. Individual records must establish that overlap rather than assuming every placement involved a completer.
Staged disbursements must match the cash schedule
The hypothetical R$ 180,000 loan is released in three installments of R$ 60,000: at the start, in month 2 and in month 4. The first requires executed documents and minimum delivery capacity. The second requires evidence of initial spending and arrangements for the resources needed for training. The third depends on delivery evidence and an updated cash forecast. The budget assigns purposes; a separate payment schedule assigns timing, preventing commitments to spending before the corresponding money arrives.
For simple arithmetic, assume total repayment of R$ 198,000 through 12 equal installments of R$ 16,500, from month 7 through month 18. The nominal R$ 18,000 difference is the gross financing cost agreed for this example. It is neither a market rate nor net investor profitability, and it does not adequately measure investment performance. Disbursement timing, costs, potentially applicable taxes and loss risk matter. A real instrument must explain how charges are calculated and address early repayment, late payment and renegotiation.
The company projects R$ 22,000 a month after operating expenses and before debt service throughout the 12 repayment months. Each R$ 16,500 installment consumes 75% of that cash, leaving R$ 5,500. This headroom is a forecast, not an existing reserve. The budgeted R$ 18,000 reserve is separate and financed by the loan itself. Counting it again as operating revenue would exaggerate repayment capacity and obscure how much protection comes from the original funding.
Information rights should support decisions without obstructing operations
Information rights should specify content, frequency, format and responsibility. Useful reporting combines cash balances, budget versus actual spending, overdue obligations, expected receipts and service indicators. NGO reporting addresses use of the donation under its terms; company reporting addresses debt repayment capacity. Reports may share definitions, but consolidated figures must not imply that restricted money held by one entity is available to settle the other's obligations.
Define events requiring prompt notification outside routine reporting: loss of a major customer, interrupted services, suspected fraud, a data incident or inability to pay essential suppliers. A polished monthly report arrives too late if a serious event remains undisclosed until the reporting period closes. Conversely, approval for every minor purchase turns oversight into daily management. Decision limits and materiality criteria should reflect project size, operating speed and administrative capacity.
Investors need to reconcile figures, request explanations and, in agreed circumstances, commission independent checks. This does not grant unrestricted access to case files, personal documents or individual interviews. Financial information and aggregated data can answer many questions. Specify how discrepancies are challenged, response deadlines and responsibility for extraordinary audit costs. Controls have a cost: disproportionate requirements can consume service funding and make oversight itself impractical.
Veto powers, representation and limits on intervention
A veto can protect against decisions that materially change the agreed risk, such as substantial additional debt, disposal of essential assets, a change in the funded purpose or related party transactions outside approved terms. Each reserved matter needs a clear definition and procedure. Broad wording such as any strategic decision invites disputes and may block routine contracting. Address investor silence and urgent exceptions needed for safety or continuity of services.
In an equity investment, representation on governance bodies depends on the corporate structure and applicable documents. In an NGO donation, oversight, committee participation or a nomination right does not create ownership or replace statutory powers. A project committee may review delivery and recommend corrections without possessing authority to impose them. A responsibility matrix should identify who recommends, approves, signs and implements each decision.
Greater operational intervention requires closer examination of responsibility, conflicts and practical capacity. Demanding immediate replacement of the entire team could interrupt services and destroy local knowledge. Escalation may instead move from clarification to a corrective plan, closer monitoring and proportionate further measures. Intervention rights need a legal basis and a workable execution plan. A clause promising takeover without staff, lawful access or suitable legal arrangements offers little protection during a crisis.
Beneficiary protection belongs in operational obligations
Beneficiaries are not collateral, a supply of promotional stories or a compulsory source of data. Agreements should establish service standards, selection criteria, accessibility, complaint handling and responses to vulnerability. Selecting the example's 120 participants must reflect the barriers the project intends to address. Choosing only people already likely to find work could improve the headline placement rate while reducing the contribution to the original social problem.
Monitoring arrangements should define necessary data, permitted users, retention periods and purposes. The appropriate legal basis for processing requires contextual analysis; blanket permission does not solve every situation. Participants should not be pressured to permit promotional use of their image as an improper condition of support. Aggregated or pseudonymized reporting may provide oversight with less exposure. Sensitive information requires additional care and access proportionate to demonstrated need.
A suspension or closure plan should cover commitments to current cohorts, communication with participants, lawful handling of records and feasible referrals to other services. Protection does not require indefinite support for an unviable project. It does require avoiding an improvised withdrawal that transfers every consequence to participants. Budget and governance arrangements should provide for transition. A social obligation without an owner, resources or a procedure is unlikely to withstand cash pressure.
Useful metrics and the limits of attributing outcomes
In the example, 90 completers out of 120 participants produce a 75% completion rate. If 54 people obtain verified placements, they represent 45% of the original group and 60% of completers. Those percentages answer different questions; showing only the second conceals the experience of noncompleters. Define denominators, completion and qualifying placement in advance, including treatment of existing employment, very short engagements and information that cannot be confirmed.
Verification six months after completion may reveal job retention and placement quality, but requires funding and a contact strategy. People who do not respond must not silently disappear from denominators. Report how many were reached, the evidence supporting results and the implications of missing follow-up. Enrollment records establish registration; attendance records support participation claims; employment evidence supports a different claim. None alone establishes that the project caused the observed change.
Suppose, purely for analysis, that 36 of the 120 people would find work without the intervention. Compared with 54 observed placements, that implies 18 additional placements. This depends on an unproven counterfactual assumption. Dividing R$ 300,000 by 54 gives approximately R$ 5,555.56 per observed placement; dividing it by 18 gives approximately R$ 16,666.67 per hypothetical additional placement. Both calculations use the whole budget, not just the donation. Neither alone measures benefit quality, duration or causality.
Financial difficulties can arrive before impact evaluation
Consider an adverse scenario in which monthly cash before debt service falls from the projected R$ 22,000 to R$ 12,000 for three repayment months. With installments unchanged at R$ 16,500, the monthly shortfall is R$ 4,500, totaling R$ 13,500. If the R$ 18,000 reserve is fully available and its use permitted, it covers that gap and ends at R$ 4,500. This demonstrates how a particular buffer absorbs a specified shock, not that the business has recovered.
Require an updated forecast before the reserve is exhausted. A persistent decline may call for spending reductions compatible with service commitments, renegotiation, new funding or operational restructuring. The NGO's R$ 120,000 donation is not unrestricted company cash and does not become security for the debt because both entities participate in the project. Changing that separation requires its own analysis and compliance with the relevant purposes and instruments; an informal accounting adjustment cannot achieve it.
Repayment and evidence may also follow different schedules. Loan payments begin in month 7, while some employment outcomes can only be assessed after activities finish and the agreed follow-up period passes. An installment cannot implicitly depend on an unavailable metric. Any outcome-linked remuneration needs a measurement window, verifier, challenge procedure and identified payment source. A future impact promise cannot replace the actual schedule of cash receipts and payments.
Default, security and realistic recovery expectations
Default provisions should distinguish missed payments, misuse of funds, false information and remediable breaches of ancillary obligations. A late report may warrant a request and a cure period; established diversion of funds may justify immediate action. Responses depend on the instrument and seriousness. Treating every failure as grounds for immediate repayment can escalate disputes and trigger an avoidable crisis. Define how facts will be established when parties disagree about whether a default occurred.
Security requires examination of existence, ownership, availability, formalization, priority and enforcement costs. Essential equipment may have little resale value while remaining necessary to complete a cohort's training. Existing eligible receivables differ from hoped-for sales: a forecast does not become an established claim because it appears in a spreadsheet. Check assignments, liens and other commitments before counting revenue as available. Even properly constituted security may fail to deliver full recovery within the investor's required timeframe.
Recovery planning should compare concrete alternatives: negotiation, orderly asset sales, replacement of a provider where feasible, or appropriate collection proceedings. Legal expense, elapsed time and operational consequences belong in that comparison. Preserving services and recovering money may align, but they can also conflict. The agreement needs a procedure for choosing without promising protection for everyone. Impact investment remains exposed to loss, including where a project has been run honestly.
Incentives and conflicts must be visible
Company–NGO arrangements require attention where founders, managers, suppliers or relatives overlap. A company may receive commercial revenue while the NGO bears costs that make it possible. This is not automatically misconduct, but the economics must be explained and each contract justified by the paying entity's purpose. Investors should understand related parties and approval procedures, including abstention from voting or independent review where appropriate.
Indicator-linked compensation can distort behavior. Paying only for completion may encourage weaker standards; paying only for placements may exclude people facing greater barriers. One approach combines access criteria, delivery quality and outcomes, with review of withdrawals and complaints. No formula removes every unwanted incentive. Monitoring should look for opportunistic selection, duplicate records or pressure on participants to confirm results inconsistent with their experience.
Investors may also have conflicts, including an interest in buying the funded company's services, publicizing results quickly or favoring another portfolio business. Declare these interests and address them in decision procedures. An evaluation controlled entirely by the funder demanding it has reduced credibility. Rules for appointing evaluators, disclosing limitations and allowing responses help distinguish verification from promotion. Governance and evaluation costs must appear in the budget; independence needs practical resources.
Resolve disagreements before turning them into clauses
Suppose an investor wants every disbursement suspended if placements fall below target, while the NGO argues that participants facing greater barriers need longer support. Capital preservation, continuity and uncertainty create a real disagreement. Negotiation improves when each party names the risk it wants to control. Automatic suspension may withdraw support from those needing it most; unconditional funding may sustain an ineffective intervention. An intermediate decision process needs evidence and clear authority.
In the example, month 4 funding can depend on documented delivery, reliable information and updated viability without requiring all 54 placements already to exist. A material participation shortfall could trigger investigation, a revised plan and an updated budget. Inadequate transport calls for a different response from unsuitable training content or poor selection. A right to withhold funds should specify the review period, opportunities to correct problems and treatment of legitimately incurred commitments.
Indicator disputes need a process too. One party may count an occupation as a qualifying placement while another considers it outside the agreed objective. Prior definitions, decision records and technical review reduce opportunistic reinterpretation. Persistent disagreements can escalate from operational leads to representatives authorized to settle, followed by a suitable dispute resolution mechanism. Contract length cannot replace this design. Difficult decisions must be executable at proportionate cost while protecting obligations that cannot wait.
Documentation, technology and the scope of intermediation
Organized documentation connects the project, proponents, beneficiaries, budget, instrument and evidence. A project-based method should identify each item's source, responsible party and status: proposal, assumption, submitted document or information checked through a described procedure. KEZS connects entrepreneurial projects and potential funders through technology and intermediation. That role can help organize analysis, but investors conduct their own diligence and make their own decisions. KEZS does not assign a credit score.
The method's usefulness depends on document quality and how questions are resolved. A complete profile does not establish solvency, regulatory authorization or impact. Public distributions of securities and regulated financial activities require structures and providers appropriate to their classification. Technology or a Brazilian CNAE business activity code is not permission. Banco Central materials on correspondents and CVM materials on crowdfunding and investment funds are entry points for specialist analysis, not proof of authorization for KEZS or any project.
Digital integrity must also be distinguished from truth. Depending on the mechanism, a signature or record may support authorship, version history and the sequence of changes; it does not independently establish that an expense occurred or a signer had sufficient powers. Blockchain establishes neither truth nor legal title, reserve solvency or payment guarantees. Private documents and beneficiary data should not be placed on public blockchains. Document governance needs access, retention, correction and traceability controls without treating permanent exposure as transparency.
Correct design errors without erasing the record
A recurring error is calling every transfer funding and leaving financial returns for a later conversation. Start correction by reconstructing what was actually agreed, rather than relabeling the transaction to improve its appearance. Another error is using one report for a donation and loan, obscuring distinct obligations. Separate records by entity, instrument and purpose, reconciling any flows between them. Material changes need appropriate formalization; editing a spreadsheet does not amend a contract.
Measurement errors include duplicate participants, changed denominators and confusion between getting a job and staying employed. Preserve the earlier version, explain the cause, recalculate affected indicators and notify those who relied on the old figures. If two of the 54 recorded placements were duplicates, the corrected count would be 52, approximately 43.3% of the 120 participants. Cost calculations and comparisons with the counterfactual assumption also need revision. Correcting the headline alone leaves related decisions based on inconsistent evidence.
Some rights are impossible to exercise: data access without a secure channel, an audit without a budget or a veto without a response deadline. Assign an owner, resources and procedure, and formally amend unworkable obligations. Change history helps distinguish legitimate learning from opportunistic rule changes. Recording a limitation early allows expectations to be renegotiated. Concealing it until repayment falls due narrows the available options and increases conflict.
Complete the decision with monitoring and exit arrangements
Before signing, investors should be able to explain the instrument, any repayment source, negotiated powers and possible loss. Separate evidenced claims from assumptions awaiting execution. In the example, the R$ 300,000 budget, three R$ 60,000 disbursements and 12 R$ 16,500 repayments are proposed terms. The 54 placements and R$ 22,000 monthly cash are hypotheses. Later receipts, contracts and records may establish particular facts without automatically validating the whole thesis.
Include the cost and capacity to monitor. Investors unable to review reports, discuss deviations and exercise rights need a simpler arrangement or suitable support. Define closure as well: repayment and release of security for a loan, accountability for a donation, continuing confidentiality and treatment of data. An equity exit depends on actual conditions and interested buyers. A preferred holding period creates neither liquidity nor an obligation for a third party to buy.
A practical closing package combines a transaction summary, consistent instruments, a responsibility matrix, a time-based budget, an indicator protocol and a response plan for deviations. Legal, accounting and technical review should focus on material issues, including regulatory classification where relevant. Signing then becomes a decision about understood risks and usable rights. An impact commitment becomes more credible when documentation supports recognition of results, correction of errors and handling of losses while making clear who bears each consequence.
Frequently asked questions.
Can I demand my money back after donating to an NGO?
A donation does not, by itself, create a right to financial returns. Its terms may specify purposes, conditions, reporting duties and consequences for particular violations, subject to case-specific legal analysis. That differs from making repayment with remuneration a normal feature of the arrangement. If payments to the funder are the economic intention, examine and document that intention from the outset, taking the recipient's legal nature into account. A contribution should not be advertised as a donation while being treated internally as a loan.
Can an NGO give a funder the same powers as an owner of an LTDA?
There is no automatic equivalence. A nonprofit has its own structure, powers and purposes. It may agree to compatible oversight and obligations, but cannot offer ownership of its assets as a Brazilian limited liability company offers equity interests. Committee participation or reporting rights do not make a funder an owner. Check the bylaws, representatives' authority and scope of each right. A project agreement should not promise to displace powers belonging to the nonprofit's governing bodies.
Should every payment or disbursement depend on the final social outcome?
Such conditions are workable only when outcomes can be measured within the relevant period and a payment source is identified. Social outcomes may emerge after spending and depend partly on external factors. Making all funding conditional on job placements could prevent training from taking place. One possible structure combines delivery evidence, intermediate indicators and later evaluation, with different responses to disappointing outcomes, operational failure and fraud. Each condition needs evidence requirements and a way to challenge findings.
How can I assess whether a project generated additional impact?
Investigate what would probably have happened without the intervention. Comparing everyone served with zero can exaggerate contribution because some participants could obtain support or outcomes elsewhere. A suitable comparison group or another evaluation design may strengthen the analysis, but feasibility, ethics and cost matter. Where only estimates are available, disclose assumptions and scenarios. Observed outcomes provide useful evidence; establishing that the intervention caused them requires a different demonstration.
Do oversight rights allow access to all beneficiary data?
No. Oversight must respect purpose, necessity and applicable data processing rules. Aggregated reports, sample checks and controlled evaluator access may provide sufficient evidence with less exposure. Necessary personal information requires a defined legal basis, responsible parties, security and retention arrangements. Funding does not authorize promotional use of contacts, sharing identifiable stories or demanding unrelated documents. Guidance from Brazil's National Data Protection Authority, ANPD, is a starting point for deeper analysis, not permission for unrestricted access.
Does collateral or a blockchain record make an investment safe?
Neither eliminates loss risk. Security depends on ownership, formalization, priority, recoverable value and enforcement conditions. A blockchain record may help check a record's integrity without proving its truth or the existence of sufficient assets to pay. It does not independently create legal ownership of an asset. Review the supporting documents, financial flows and recovery mechanisms. Private information and personal documents should not be published on a public blockchain as a substitute for proper controls.
Which document deserves the most attention before funding?
The instrument creating the rights is central, but cannot be assessed alone. It must align with the entity's governing documents, budget, cash schedule and impact protocol. Look for concrete inconsistencies: does the signer have authority; does the debtor receive the revenue; can a target be measured before the decision depending on it; can required data be accessed lawfully? Resolve material mismatches before disbursement, recording responsibilities and conditions. Strong wording in one document cannot repair contradictions elsewhere.
How does technology-based intermediation affect investor responsibility?
It may help present projects, compare documents and organize questions, depending on the resources actually available. It does not transfer the decision to technology or establish a transaction's economic or social quality. KEZS connects projects and potential funders without replacing their own diligence or assigning a credit score. Investors must understand the instrument, check relevant evidence and assess necessary providers, including regulated structures where required. Intermediation provides no guarantee of approval, repayment, profitability, liquidity or preservation of capital.
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 ↗
- CVM — Resolução 88, texto e alterações ↗
- CVM — Resolução 175, fundos de investimento ↗
- Portal do Empreendedor — perguntas sobre MEI ↗
- ANPD — orientações e perguntas frequentes ↗
AI-assisted educational content. Decisions require verification of the information, instruments and providers in each actual case.
