An investment syndicate can hold stakes in ten businesses, work with ten founders and sign ten contracts while remaining exposed to the same cause of loss. This happens when their customers cut spending together, they rely on one distribution channel, or they all need another funding round under similar conditions. Counting holdings describes the portfolio, but says little about its resilience. For a group focused on a particular sector or region, the useful question is which events could damage several portfolio companies at once.
Diversify by the mechanism that causes a loss
Diversifying by risk source means identifying those events and spreading exposure across economic mechanisms that are sufficiently distinct. One company might lose revenue because its customer runs out of cash; another because an integration stops working; a third because a major contract is not renewed. These effects can interact, but their origins differ. The task is to explain those differences before investing and check whether they remain meaningful as the businesses develop. A category label alone does not establish independence.
A focused investment thesis does not have to be abandoned to conduct this analysis. Sector knowledge can improve judgments about products, customers and execution. The problem arises when that knowledge advantage is mistaken for financial protection. Specialization can improve individual decisions while increasing collective exposure to one shock. Recognizing both effects makes deliberate concentration possible, without describing a specialist portfolio as protected against every business cycle. The group should understand the risk it accepts in exchange for its expertise.
Turn the thesis into limits that guide decisions
A thesis such as investing in enterprise software within a chosen Brazilian region is still too broad to resolve difficult choices. The group must specify which problems it understands, which customers it can investigate and which business stages it can monitor. It should also state exclusions: dependence on a license nobody can assess, capital needs beyond the pool's resources, or no access to essential information. These boundaries preserve the value of specialization when an attractive opportunity falls outside the group's competence.
Translate preferences into observable criteria. Instead of seeking diversified revenue, identify the main payers and analyze the consequences of losing the largest contract. Instead of accepting any nearby business, examine where its customers, suppliers and infrastructure are located. The registered office may make meetings and monitoring easier, but does not reveal the origin of demand or how a recession would reach the company. A locally based business can depend on distant customers whose budgets move together.
Set limits before enthusiasm about a particular founder dominates the discussion. They might cover exposure per company, shared dependencies, refinancing needs and decision-making concentrated in the leader. No percentage is universally appropriate. Each limit reflects a trade-off among variety, diligence depth, cost and monitoring capacity. Its purpose is to make the choice reviewable: the opportunity fits the mandate, requires a formal exception, or should be declined. A limit that changes informally for every attractive proposal provides little discipline.
Build a map of overlapping dependencies
Start with the paths connecting each business to cash. The product user may differ from the party that pays, approves the budget or bears the risk of nonpayment. Software used by clinics, for example, may depend economically on payments from health plan operators. Software sold to manufacturers may depend on export orders. Calling both businesses technology companies hides the source of their money. Record the economic payer and the event that could interrupt the flow, including material intermediaries between the customer and the cash received.
Next map operational and financial dependencies: cloud infrastructure, distribution, banking integrations, logistics, suppliers that are difficult to replace and access to later funding rounds. Some dependencies are directly shared; others are transmitted through an intermediary. Two companies may use different providers but depend on the same integrator, or serve different sectors while facing the same purchasing committee. Investigate that link when it is material. The aim is to identify consequential connections, rather than document every minor supplier.
Exposure dimensions overlap. A R$70,000 investment may belong both to an industrial-demand group and to a cloud-dependency group. Appearing in two categories does not turn it into R$140,000 invested. Each dimension answers a separate stress question: what happens if industrial demand weakens, and what happens if the cloud service fails? Adding all columns would create misleading totals. Compare each column with its own limit to identify events that could affect a material share of the portfolio.
A numerical example with separate commitments
Consider a hypothetical R$1,000,000 pool, used solely to illustrate the arithmetic. Its plan allocates R$700,000 to initial investments, R$250,000 to follow-on investments and R$50,000 to vehicle and operating expenses. Ten initial investments, labeled A through J, receive R$70,000 each. The example assumes every planned capital commitment is paid and total costs stay within budget. These are assumptions, not evidence that future capital calls will be honored or that expenses will remain stable. The spending allocation is specific to this exercise.
All ten companies fit the thesis of enterprise software in the chosen region, but their customers earn money from different sources. A, B, C and D depend on discretionary consumption, totaling R$280,000. E, F and G depend on healthcare payment flows, totaling R$210,000. H, I and J serve industrial export chains, also totaling R$210,000. For this exercise, the group sets a hypothetical R$350,000 ceiling for each demand source and a R$140,000 ceiling on invested capital per company.
The cross-portfolio map adds three dependencies. A, E and H share a material cloud dependency; B, F and I depend on a banking integration; C, G and J depend on a logistics partner. Each group initially contains R$210,000, with a hypothetical operational-dependency ceiling of R$350,000. D is outside these three groups but has other risks. Limits use capital invested, including follow-ons, rather than later valuations. Estimated economic value and the severity of each dependency must therefore be monitored separately.
Assess the risk each opportunity adds
Assess both the individual merits of a prospective investment and its contribution to the portfolio. A company may have a capable team, satisfied customers and negotiable terms while increasing an already substantial dependency. Another may introduce a new source of demand but fail the group's minimum quality standard. Diversification cannot turn a weak business into a sound investment. Establish the minimum standard first, then examine how an opportunity that meets it changes shared exposures, cash requirements and monitoring demands.
In the example, another company linked to discretionary consumption would enter a category already containing R$280,000. A R$70,000 investment would reach the R$350,000 ceiling and remove the headroom for additional investment in A, B, C or D. That lost option is part of the decision's cost. Including the candidate in the initial plan would require replacing one of the ten companies or reducing other investments to preserve the R$700,000 budget. Quietly taking money from the reserve would change the approved financial mandate.
This marginal analysis also prevents labels from becoming shortcuts. A healthcare company whose actual buyer is a large employer's benefits department may suffer alongside consumer-facing businesses. Ask which budget funds the purchase and who can cancel it. The decision record can identify a provisional classification and require validation. If available information cannot establish distinct risk sources, treat independence as a hypothesis. Do not claim diversification that has not been demonstrated, especially when that claim would justify using scarce concentration headroom.
Distinguish evidence, estimates and narrative
A dependency map is only as reliable as its supporting information. Sales presentations explain a proposition but do not establish revenue concentration, renewals or collection capacity. Due diligence can combine contracts, reconciled billing records, receipts, financial statements, authorized interviews and operating information. Each answers a specific question. A signed contract establishes agreed obligations; it does not establish that the customer will keep using the service or pay on time. Reconcile the evidence with the economic claim being tested.
A useful investment record separates the claim, evidence, interpretation and remaining gap. The claim might be that no customer is critical. Evidence might show that three payers account for a large share of receipts. The interpretation is economic concentration despite many users. The gap might concern whether apparently separate payers belong to the same corporate group. This structure allows participants to challenge a conclusion through facts, rather than reducing disagreement to confidence in the founder or the reputation of another investor.
Estimates also need a basis. If a supplier is described as easy to replace, investigate alternatives, engineering effort, cost and a plausible migration timetable. An exploratory conversation supports less confidence than a migration completed in a test environment. Missing information may justify additional diligence, a contractual condition or rejection. Record what evidence could change the conclusion. Monitoring can then seek information that matters to the decision, including information that contradicts the original case, rather than merely collecting reassuring updates.
Give the leader a mandate that can be checked
The syndicate leader often concentrates deal sourcing, negotiations and founder relationships. This reduces friction but creates dependence on one person and the risk of insufficient challenge. The mandate should state what the leader can do alone, what requires approval and what information must be reported. Speaking to a company, commissioning diligence within an agreed budget and committing investment capital are different acts. Authority for one should not be interpreted as authority for all, even when the acts concern the same opportunity.
One possible arrangement lets the leader conduct preliminary analysis and negotiate conditional proposals, while a committee approves investments, exceptions and material changes. Documents should also address representation at company meetings, signing authority, data access, exercise of rights and incident reporting. Whether each act can legally be delegated depends on the structure used. An expedited procedure can address urgency, but needs objective limits, accountability and exclusion of conflicted decision-makers. A short deadline does not itself create legal authority.
Compensation and incentives require equal clarity. A completion fee may encourage volume; a share of investment gains may encourage greater risk; fixed compensation may fail to reflect workload. No model automatically removes conflicts. Specify calculations, reimbursable expenses, personal investments, commercial relationships and preferential access to opportunities. Also provide for replacing the leader and transferring records in an orderly way. The portfolio must remain manageable if its central person becomes unavailable, including access to decisions, contracts and pending obligations.
Design decision rules that can withstand disagreement
The decision process should identify who decides, what documents they receive, how quorum is calculated and how conflicts and abstentions are handled. Voting per person, per unit of capital or per committee seat creates different incentives. Larger financial contributors may seek proportional influence, while those responsible for analysis may favor a technical committee. The arrangement must match the applicable structure and contractual rights. Investors should understand their actual powers, including whether a vote concerns the vehicle's decision or direct rights in a company.
Separate ordinary approvals from exceptions to the mandate. An investment within the limits can follow the standard procedure; exceeding an exposure ceiling or releasing reserved money requires a specific decision. Minutes should record the amount, instrument, conditions, conflicts and rationale. Silence should not become consent for convenience unless that consequence has been validly agreed. A company's short fundraising window does not fix defective representation. If the required approval is unavailable, the group may need more time or may have to pass.
Suppose the leader proposes supporting a company that recently lost its largest customer. One member sees a temporary setback; another sees structural deterioration. Useful disagreement produces verifiable questions: is replacement revenue contracted, how much cash remains and which expenses can be reduced? The vote might approve a tranche conditional on evidence or reject the investment. Recording the minority view supports later review. The aim is a defensible decision even when participants retain different judgments, rather than an appearance of unanimity.
Reserve for follow-ons without promising a rescue
The example's R$250,000 reserve does not belong in advance to the ten portfolio companies. It preserves the ability to decide as new information emerges. It might fund demonstrated expansion, maintain ownership where economically justified or support a transition with a credible rationale. It can also remain unused if the available choices are poor. Treating it as an obligation to rescue every company converts a decision resource into a commitment the budget may be unable to support, especially if several companies need help together.
Distinguish a contractual right to participate, an intention to participate and cash actually available. A right may have deadlines and conditions; an intention may change after diligence; an investor's committed money may not yet have reached the vehicle. Communications to founders must reflect these differences. Suggesting that future backing is certain can influence hiring and spending without corresponding authority or resources. A reserve plan should therefore describe the decision process and constraints without becoming an informal funding promise.
Follow-on criteria should examine the use of new capital, verifiable milestones, financing alternatives and concentration effects. Compare investing with the consequences of declining. Preserving an ownership percentage is insufficient by itself: accepting dilution may conserve cash when pricing or risk has worsened. Money already invested does not independently justify another investment. Assess the additional capital, the rights it buys and the scenarios it can finance from that point forward. An earlier loss cannot be undone simply by increasing exposure.
Connect capital calls, cash and the sequence of decisions
In the hypothetical scenario, participants initially pay in R$750,000: R$700,000 for the ten investments and R$50,000 for operations. The remaining R$250,000 stays committed for later capital calls under the applicable documents. A commitment is not a bank balance. Before assuming a follow-on obligation, the finance lead must confirm payment deadlines, contractual conditions and receipt of funds. A promise to a founder should not depend on uncertain informal collection. Approval and available cash are separate checks, and both matter before disbursement.
After the first monitoring round, E, H and J each present an approved R$70,000 follow-on opportunity. To simplify the demonstration, assume the entire reserve is called and received before those payments. The group invests R$210,000 and retains R$40,000. Total capital invested reaches R$910,000. Assuming the full R$50,000 operating budget is eventually spent, the final reconciliation is R$910,000 invested, R$50,000 spent and R$40,000 in cash. Together these account for the full R$1,000,000 paid in, without treating expenses as portfolio assets.
Demand exposures change too: consumption remains R$280,000, healthcare rises to R$280,000 and export industry reaches R$350,000. E, H and J each reach R$140,000 invested. Cloud exposure through A, E and H reaches R$350,000; logistics through C, G and J reaches R$280,000; banking integration remains R$210,000. Every example limit is respected, but export industry and cloud have no headroom. Another investment in either group requires an explicit review; the remaining R$40,000 does not itself authorize it.
Test shocks without manufacturing precision
A stress test asks what happens if a dependency fails and what responses are possible. It need not begin with a numerical probability. For consumption exposure, assume weaker demand, cancellations and slower collection. For healthcare, consider delays in payments reaching customers. For export industry, consider reduced orders. Estimate effects on revenue, margins and cash, distinguishing contractual facts from judgment. A scenario is useful when its assumptions and transmission channels are understandable, even if the group cannot credibly assign it a precise likelihood.
After the follow-ons, a complete loss of A, B, C and D would eliminate R$280,000 of invested capital, leaving R$630,000 invested in the other companies. That balance is not guaranteed recoverable value; it only identifies capital allocated to survivors. Separately, a cloud interruption would affect A, E and H, representing R$350,000 of exposure. It would not automatically cause a R$350,000 loss. The result would depend on duration, redundancy, contractual responsibility and migration capacity. Exposure identifies what is at risk, not a predetermined outcome.
The most useful tests include simultaneous effects. Falling revenue may coincide with difficult fundraising, prompting several companies to seek support when the pool has only R$40,000 free. Prioritization, expense reductions and negotiations then become critical. A portfolio can have varied commercial customers while remaining concentrated in dependence on external financing. Stress testing should produce preparatory decisions and warning indicators, such as information to request or options to investigate, without presenting hypothetical scenarios as statistical forecasts or assured loss estimates.
Read aggregate results alongside illiquidity
To complete the arithmetic, imagine a simplified adverse outcome after the follow-ons. A, B, C and D return no capital. E produces total receipts of R$280,000 on R$140,000 invested; H produces R$210,000 on R$140,000; J produces R$70,000 on R$140,000. F, G and I each return R$70,000. Receipts from the investments therefore total R$770,000. Adding the R$40,000 left in cash produces R$810,000 available. These are hypothetical total receipts, rather than claims about actual companies or achieved performance.
Against R$1,000,000 paid in, the result is a R$190,000 loss, or 19%, including the planned R$50,000 in costs. The calculation assumes no additional expenses, taxes, obligations or compensation, and ignores the time value of money. It reconciles inflows and outflows rather than estimating a likely return. E's positive result does not offset four complete losses, J's partial recovery and expenses. Reporting only successful holdings would misrepresent the pool's performance; a gain on one investment is different from a gain for participants overall.
Actual receipts may arrive at very different times while obligations continue to fall due. A valuation above the entry price puts no cash in the bank. Liquidation preferences, dilution and exit costs can also separate an announced value from proceeds received. Reports should distinguish invested capital, estimated holding values, realized cash and distributions to participants. Even realized cash may be needed for remaining obligations before distribution. Diversification can reduce reliance on one outcome, but cannot create a buyer or guarantee liquidity.
Negotiate rights that make monitoring possible
Monitoring capacity starts during negotiation. Investors need to know what information they will receive, how often and at what level of detail. Useful requests include cash balances, developments involving major customers, material contract changes and planned fundraising. Requirements should be proportionate to company size. An overly complex package can consume management time without improving decisions; a report containing only founder-selected indicators may conceal the dependencies behind the investment case. Agree information needs around the risks the group actually intends to monitor.
Information rights, preference rights, participation rights in future rounds and consent requirements serve different purposes. Information helps reveal changes; a preference right provides a contractual opportunity; consent can restrict specified acts. None removes execution risk or ensures payment. Assess whether vetoes could impair operations or obstruct later financing. A protection that is expensive to monitor and difficult to enforce may be worth less than its wording suggests. The practical ability to exercise a right matters alongside its legal formulation and the cost of enforcement.
If confidentiality prevents disclosure of customer identities, consider aggregated information, restricted access or independent confirmation where feasible and legally appropriate. Seek sufficient evidence without unnecessary exposure. Legal costs weigh more heavily as a percentage of a small investment. Templates can organize negotiations, but do not replace review of the actual rights and structure. If the minimum cost of contracting and monitoring makes the opportunity uneconomic, include that fact in the investment decision rather than assuming documentation can simply be reduced without consequence.
Establish the structure before collecting money
Gathering investors in a messaging group does not automatically create an authorized investment fund. A pool may need a specific legal structure depending on fundraising, representation, investment arrangements and distribution of proceeds. Before receiving money, establish who will own the assets, who can move funds, how each participant's rights will be recognized and which service providers are required. Technology and a Brazilian CNAE business activity classification do not replace applicable permissions. An operating description is not authorization to conduct a regulated activity.
The instrument changes the rights and risks being monitored. Equity creates an ownership interest with exposure to residual value. Convertibles depend on conversion conditions and other agreed rights. Loans create payment obligations but retain default risk. Existing eligible receivables require checks on origin, legal ownership, disputes and collection; hoped-for sales are not the same asset. A donation to a nonprofit is neither equity nor a promise of financial return. A Brazilian MEI cannot admit equity partners while retaining MEI status. These distinctions must survive presentation in a common interface.
Public securities distribution and regulated financial activities require appropriate structures and providers. Materials from Brazil's CVM on offerings and funds, Banco Central on banking correspondents, and Portal do Empreendedor on MEI provide entry points for specialized research. They do not prove any platform's authorization. A syndicate should compare cash flows, priorities, obligations and potential losses with professional support suited to its chosen structure. Presenting several instruments together does not make their protections, repayment terms or legal treatment equivalent.
Assign work and correct information errors
A small operation may combine several functions in one person, but responsibilities still need to be distinguished. The leader coordinates the thesis and relationships; the analyst maintains hypotheses and evidence; finance reconciles commitments, cash and expenses; legal support checks documents and authority; the committee decides within the mandate. Assign a monitoring owner and a substitute for each company. This prevents everyone from assuming someone else examined an alert or confirmed a payment. Where roles overlap, identify which review or approval provides a meaningful check.
Update the dependency map when a company changes. A business previously serving exporters may acquire a major domestic buyer; a replaceable integration may become critical. If F and I are found to serve companies in the same corporate group, reclassify the exposure, identify affected decisions and assess the consequences. Preserving the old category to maintain apparent compliance turns an initial error into continuing distortion. Keep a history showing what changed and why, so subsequent decisions use the corrected picture and earlier reasoning remains understandable.
Private documents require access, purpose and retention controls suited to their contents. Customer and personal data should not circulate freely among all participants; ANPD guidance can inform the analysis. Where justified, recording a cryptographic digest on a blockchain can support integrity checks, but does not prove truth, legal title, reserve solvency or payment. Private documents should not be placed on public blockchains. Reliable records still require verification, context, controlled access and responsibility for correcting errors. A durable entry can preserve inaccurate information just as durably.
Use documentation to decide and learn
Organize the method around each project: its economic problem, proposed instrument, evidence, dependencies, decision and monitoring. This can make smaller opportunities comparable without erasing their differences. KEZS connects entrepreneurial projects and potential funders through technology and intermediation. Investors conduct their own diligence and make their own decisions; KEZS does not assign a credit score. Clear documentation may support analysis, but cannot guarantee approval, repayment, profitability, liquidity or preservation of capital. Its usefulness depends on the quality and relevance of the underlying information.
Before investing, the group should explain why the company fits the thesis, which risk source it adds and which shared exposure it increases. It should also show who authorized the decision, which rights will be acquired and how disbursement fits available cash. Monitoring then checks the facts supporting that conclusion. A changed main customer, depleted reserve or altered cost structure must affect the next decision, even if the investment still carries an attractive estimated valuation. An earlier approval does not settle every later allocation choice.
Finish with a joint review of mandate, map and budget. Reconcile amounts, identify gaps, confirm responsibilities and record events requiring another decision. In the example, export industry and cloud have reached their ceilings, R$40,000 remains available and enthusiasm alone cannot authorize another obligation. That picture helps guide negotiations, conditional proposals and refusals. A coherent portfolio combines specialist knowledge, monitoring capacity and explicit financial limits, keeping potential usefulness, costs, uncertainty and possible loss visible throughout the investment process.
Frequently asked questions.
How many companies does a syndicate need to be diversified?
No holding count resolves the question by itself. Ten companies may depend on the same customer budget, while a smaller group may have more distinct economic loss mechanisms. Assess the number alongside concentration, investment sizes, costs and monitoring capacity. Adding holdings without sufficient diligence can spread administrative attention thinly while preserving the common risk. The group should explain which shocks could affect each company and how much capital each shock reaches. Those exposures can overlap, so they should be assessed by dimension rather than added into an inflated total.
Does a regional thesis prevent diversification by risk source?
No, but it creates concentrations that must be acknowledged. Nearby companies can sell into different markets and receive money from different economic payers. They may also share infrastructure, specialist labor or exposure to a local disruption. Separate the location of the registered office, the location of operations and the origin of revenue. If the group knowingly accepts a regional dependency because it understands that market better, record the choice and test its effects. Proximity may improve access and monitoring, but does not itself provide financial protection.
Can the leader decide alone when an opportunity needs a quick response?
Only within powers validly granted through the group's structure and documents. A mandate can provide limited authority or an urgent procedure with amount limits, conflict restrictions and accountability. A founder's requested deadline does not expand those powers. If the necessary approval cannot be obtained, the leader may need to negotiate more time or let the opportunity pass. Granting discretion after a signature does not automatically cure an obligation entered into without proper authority. Negotiating a conditional proposal and committing capital should be treated as distinct actions.
Should the follow-on reserve be divided equally among portfolio companies?
An equal division can be a planning reference without becoming an obligation. Future needs, evidence and terms will differ. One company may justify funding to deliver signed contracts; another may seek money merely to postpone a problem with no demonstrated solution. Assess the additional capital on its merits and consider the resulting concentration. Existing contractual funding commitments must be separated from a discretionary reserve and included in cash obligations. Also distinguish money committed by participants from money received by the vehicle: an allocation on paper cannot fund an immediate disbursement.
What should the group do with an excellent opportunity that exceeds a concentration limit?
First verify the classification and arithmetic. Then consider a smaller investment, replacing another planned allocation or declining the opportunity. If the documents permit a limit change, use the prescribed exception procedure and record the resulting exposure and reasons. Renaming a category to make the investment fit does not reduce the underlying risk. An exception changes the risk accepted by the group and must be understood by those responsible for deciding. Available cash is a separate issue: having enough money does not mean the concentration mandate permits its use.
How can performance be assessed before any holdings have been sold?
Separate realized facts from estimates. Paid-in capital, investment amounts, expenses and receipts can be reconciled against records. Holding valuations depend on methods and information that should be disclosed, especially where trading is limited. A funding round at a higher price can inform the analysis, but differing rights and transaction conditions may limit comparability. Reports should show available cash, commitments and uncertainty without presenting estimated appreciation as distributable money. They should also distinguish proceeds received by the vehicle from amounts actually distributed to participants.
What is the minimum documentation for a small opportunity?
A smaller amount does not remove the need to identify the parties, their authority, the instrument, use of funds, rights, risks and disbursement conditions. Obtain sufficient evidence about decisive dependencies and agree information for monitoring. Proportionate formats and standard documents can help if they preserve the relevant substance. Where legal, operational and diligence costs make the investment uneconomic, adjust the design or decline it rather than discard essential controls. The minimum depends on the actual rights and risks, not simply on the number of pages or the investment amount.
Does a platform connecting projects take over the syndicate's due diligence?
Providing information does not automatically transfer responsibility for analysis or investment decisions. Understand the contracted scope of each service and verify evidence in proportion to the risk. KEZS connects entrepreneurial projects and potential funders through technology and intermediation; this is neither a credit score nor a guarantee. Investors remain responsible for their diligence and decisions. The group should maintain its own criteria, document information gaps and use appropriate professionals or providers when the structure or activity requires specific expertise or authorization. Reference materials do not establish platform authorization.
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.
