Deal Sourcing, Screening and Investment Decisions

Introduction

Venture capital (VC) creates value not only by supplying capital but also by identifying uncertain opportunities before they become obvious to the wider market. Deal sourcing is therefore the beginning of an investment process, not an administrative activity. The quality of a fund’s opportunity pipeline influences portfolio construction, negotiation power, sector expertise, and ultimately returns.

Sourcing and screening are especially difficult because early-stage ventures have limited operating histories, incomplete financial information, uncertain markets, and evolving products. An investor must make a disciplined decision without pretending that uncertainty can be eliminated. The central task is to distinguish promising uncertainty—where learning and scale may produce exceptional value—from unmanaged risk, weak evidence, or attractive storytelling without economic substance.

This chapter presents an integrated process for building sourcing networks, screening opportunities, assessing founder–market fit, evaluating product and traction, estimating market size using TAM–SAM–SOM and bottom-up analysis, examining defensibility and scalability, preparing investment memos, using scoring models, and making investment-committee recommendations. It also addresses cognitive bias, governance, ethics, and the responsible use of technology and artificial intelligence.

Learning Objectives

By the end of this chapter, the learner should be able to:

1. Design a structured and measurable deal-sourcing network.
2. Apply staged screening criteria appropriate to venture stage and sector.
3. Assess founder–market fit, product quality, traction, defensibility, scalability, and market size.
4. Distinguish top-down market estimates from credible bottom-up calculations.
5. Construct an investment memo and a transparent scoring model.
6. Formulate an investment-committee recommendation under uncertainty.
7. Identify and control cognitive, process, ethical, and governance risks.

1. The Venture Investment Funnel

A VC fund typically moves through a funnel:

Sourcing → Triage → Initial diligence → Commercial and technical diligence → Term-sheet decision → Investment → Portfolio monitoring.

The funnel is not merely a sequence of meetings. Each stage should answer a different question:

A fund should define an “investment mandate” before reviewing individual deals. It may specify preferred sectors, stage, geography, initial cheque size, ownership target, reserves, impact constraints, and prohibited activities. Without a mandate, decisions become dominated by personality, novelty, or the most recent opportunity.

2. Designing Deal-Sourcing Networks

2.1 Sourcing channels

A diversified sourcing system commonly includes:

• Founder referrals: trusted founders often identify capable entrepreneurs early.
• Co-investor and angel networks: syndicates provide access and informal reference information.
• Accelerators, incubators, and universities: useful for pre-seed and technology ventures.
• Customers and industry experts: valuable for identifying painful problems and emerging suppliers.
• Sector communities and professional associations: support specialist sourcing.
• Targeted outbound research: investors identify companies that fit a thesis and initiate contact.
• Public databases and company registries: useful for market mapping, but not substitutes for verification.
• Events and demonstrations: helpful for relationship building, although presentation skill can distort judgment.

A strong network is relational rather than purely transactional. The investor should contribute insight, introductions, hiring assistance, or market knowledge before requesting a referral. Trust improves the quality of information and increases the probability that founders disclose problems early.

2.2 Sourcing architecture and metrics

The fund can manage sourcing as a funnel with stage-conversion metrics. For a period, the number of opportunities entering a stage can be recorded as follows:

The formula is descriptive, not a universal quality standard. It is meaningful only when stages are defined consistently. A high conversion rate may indicate excellent sourcing—or excessively weak screening. Relevant measures include qualified-deal rate, referral-to-meeting rate, time from first contact to decision, sector diversity, founder diversity, and the proportion of opportunities originating outside established networks.

A sourcing dashboard should distinguish activity metrics from outcome metrics. Number of meetings is an activity measure; investments, ownership acquired, portfolio performance, and later-stage financing are outcome measures. Activity without learning or quality is not productivity.

3. Screening: From Eligibility to Investment Hypothesis

Screening should be staged. A short initial screen prevents the team from spending weeks on ventures that clearly fall outside the mandate. However, a screen should not eliminate unconventional opportunities merely because they do not resemble existing portfolio companies.

3.1 Initial eligibility screen

The analyst should establish:

• sector and business-model fit;
• development stage and financing requirement;
• geography and legal entity;
• ownership and cap-table condition;
• regulatory eligibility;
• potential conflict of interest;
• whether the fund can invest the required amount and support the company.

3.2 Substantive screen

The substantive screen tests six linked propositions:

1. Founder–market fit: Can this team understand and execute in the target context?
2. Problem and product: Is the problem important, frequent, and sufficiently painful?
3. Traction: Is there credible evidence that customers adopt, use, pay for, or renew the solution?
4. Market: Can the venture become large enough for the fund’s return objectives?
5. Defensibility: Why will the company retain value as competitors respond?
6. Scalability and economics: Can revenue grow faster than the resources required to produce it?

The output should be an investment hypothesis, not a conclusion. For example: “If mid-sized hospitals continue shifting procurement toward integrated digital workflows, and this product reduces administrative time by at least 20%, the company may reach a large regional market with attractive expansion economics.” Each claim should have an evidence plan.

4. Founder–Market Fit

Founder–market fit is broader than résumé prestige. It concerns the relationship between the team’s capabilities and the market’s demands. Relevant dimensions include:

• direct experience with the customer problem;
• insight that is difficult for outsiders to obtain;
• technical and commercial competence;
• speed of learning and adaptation;
• ability to recruit complementary talent;
• integrity, candour, and treatment of stakeholders;
• resilience without destructive stubbornness;
• capacity to make decisions under uncertainty.

Investors should avoid confusing confidence with competence. Evidence may include prior execution, customer references, prototype iterations, domain knowledge, hiring quality, and how founders respond to adverse data. Reference checks should be conducted consistently and, where possible, with permission and appropriate confidentiality.

Team assessment also requires identifying gaps. A technically exceptional founder may need a commercial leader; a sales-oriented founder may require product or engineering leadership. A gap is not automatically a rejection if it is understood, recruitable, and reflected in the financing plan.

5. Product, Customer Value, and Traction

5.1 Product assessment

Product analysis should address:

• the specific customer and user;
• the problem and current alternative;
• frequency and severity of the problem;
• product workflow and adoption friction;
• measurable customer outcome;
• implementation, integration, and support requirements;
• product roadmap and technical feasibility.

A minimum viable product is not simply a product with few features. It is the smallest testable solution capable of generating meaningful learning about customer value.

5.2 Traction as evidence

Traction must be interpreted according to business model and stage. Useful evidence includes active usage, retention, paid pilots, recurring revenue, gross margin, bookings, renewal, waiting lists, and referenceable customers. Vanity metrics—such as downloads or social-media impressions—may matter only if they connect to customer value and future economics.

For a subscription business, one common revenue measure is monthly recurring revenue (MRR), the contracted recurring revenue expected in a month. A simplified net revenue retention measure is:

All revenue terms must refer to the same customer cohort and measurement period. “Churn” should be defined consistently, and one-time fees should not be mixed with recurring revenue. NRR above 100% indicates that the retained cohort expanded overall; it does not prove that new-customer acquisition is efficient.

Investors should examine cohort data rather than relying only on aggregate growth. A company may report rising revenue while older cohorts are shrinking and new sales merely replace lost customers.

6. Market Size: TAM, SAM, SOM and Bottom-Up Analysis

6.1 Definitions

• Total Addressable Market (TAM): the total annual revenue opportunity if the relevant product or service achieved maximum theoretical coverage under stated assumptions.
• Serviceable Available Market (SAM): the portion of TAM that the company can serve with its product, geography, regulatory position, and business model.
• Serviceable Obtainable Market (SOM): the realistically attainable share over a defined period, considering competition, capacity, distribution, and adoption.

TAM is not the same as global industry revenue unless the industry definition matches the venture’s actual economic opportunity. A credible analysis states customer boundaries, time period, price basis, geography, and inclusion or exclusion rules.

6.2 Top-down and bottom-up methods

A top-down estimate begins with an industry or population figure and applies narrowing assumptions. It is useful for orientation but can conceal incompatible categories.

A bottom-up estimate builds from observable commercial units:

For a multi-segment market:

where Nᵢ is the number of potential customers in segment i and Pᵢ is annual revenue per customer in that segment. The units must match: customers multiplied by currency per customer per year produces currency per year. If pricing varies by usage, the estimate should use expected annual usage multiplied by price per unit, while clearly stating assumptions.

SAM applies practical filters such as geography and product eligibility. SOM should be tied to a time horizon and operational capacity. For example, a five-year SOM may be based on the number of sales representatives, sales productivity, implementation capacity, churn, and competitive response—not merely “capturing 1% of a large market.”

Worked example: bottom-up market sizing

An enterprise software venture targets 4,000 eligible clinics in three regions. It expects an average annual contract value of $12,000.

Suppose only 1,500 clinics fall within the company’s current regulatory and integration capability:

The company believes it can acquire 180 clinics within five years. If the average contract value remains $12,000:

This is a market-revenue estimate, not a forecast of profit or valuation. It should be stress-tested for price discounts, implementation delays, churn, and changes in the eligible customer population.

7. Defensibility and Scalability

7.1 Defensibility

Defensibility is the ability to preserve customer value and economic returns against imitation and competition. Potential sources include:

• proprietary technology or data, where legally and economically meaningful;
• network effects, in which value increases with relevant participation;
• switching costs and workflow integration;
• trusted brand and distribution access;
• regulatory approvals or specialised compliance capability;
• learning effects and accumulated operational data;
• cost advantages at scale;
• exclusive partnerships, provided they are durable and enforceable.

A patent alone is not a moat. Investors should ask whether the intellectual property is valid, enforceable, difficult to design around, and connected to customer willingness to pay. Similarly, data becomes defensible only when it is lawfully obtained, difficult to replicate, and improves the product or economics.

7.2 Scalability and unit economics

A scalable venture can increase revenue without proportional increases in cost, complexity, or working capital. Software may scale efficiently, but customer acquisition, support, cloud usage, compliance, and implementation can still create variable costs.

A basic contribution-margin framework is:

where variable cost changes with units delivered. Fixed costs such as core salaries and rent are not subtracted in contribution margin. The framework is useful for assessing whether incremental sales contribute toward fixed costs and growth investment.

For a subscription company, customer acquisition cost (CAC) should be defined as attributable sales and marketing expenditure divided by the number of newly acquired customers in the same period, with care taken over timing and attribution. Lifetime value (LTV) is highly assumption-sensitive. A simple constant-churn approximation is:

This approximation assumes stable revenue, stable gross margin, constant churn, and a sufficiently long horizon. It excludes discounting and expansion unless explicitly incorporated. It should not be presented as a precise valuation.

8. Investment Scoring and Decision Quality

A scoring model creates consistency, but it does not replace judgment. An example weighted model is shown below.

If each criterion is scored from 1 to 5, the weighted score is:

where wᵢ is the criterion weight expressed as a decimal, sᵢ is the score, and the weights sum to 1. The result lies between 1 and 5. A score is decision support, not a probability of success. Scores should be accompanied by evidence, confidence level, and unresolved questions.

A robust process uses gates as well as averages. For example, a fund may require no unresolved legal prohibition, a minimum market threshold, and a credible path to ownership. A high average score should not compensate for a fatal regulatory issue or founder integrity concern.

9. The Investment Memo

An investment memo should allow an absent committee member to understand the opportunity, evidence, risks, and proposed action. A practical structure is:

1. Recommendation: invest, decline, or proceed with conditions.
2. Company and financing: entity, stage, round size, instrument, valuation or cap, and proposed cheque.
3. Investment thesis: two or three testable reasons the investment may generate superior returns.
4. Problem, product, and customer: use case, alternatives, adoption evidence.
5. Market analysis: TAM, SAM, SOM, assumptions, and sensitivity.
6. Traction and economics: growth, retention, margin, CAC, cash, and runway.
7. Competition and defensibility: substitutes, incumbents, barriers, and response risk.
8. Team and governance: capability, gaps, references, ownership, and board matters.
9. Risks and mitigations: ranked by severity and evidence required.
10. Valuation and returns: assumptions, dilution, exit scenarios, and sensitivity.
11. Diligence findings: completed work, exceptions, and open items.
12. Terms and recommendation: conditions precedent, ownership, reserves, and decision requested.

A memo should distinguish facts, management claims, analyst estimates, and assumptions. It should state what would change the recommendation.

10. Investment-Committee Recommendations and Bias Control

Investment committees are vulnerable to confirmation bias, anchoring on valuation or a famous competitor, availability bias, halo effects, overconfidence, survivorship bias, and groupthink. These biases are amplified when a charismatic founder presents a novel story under time pressure.

Useful controls include:

• independent preliminary scoring before group discussion;
• a designated devil’s advocate or red-team reviewer;
• explicit separation of evidence from interpretation;
• a pre-mortem asking why the investment failed;
• base rates from comparable ventures, used cautiously;
• structured reference checks;
• recording dissent and unresolved uncertainty;
• decision journals for later calibration;
• periodic review of outcomes by source, sector, stage, and investor;
• conflict-of-interest declarations and recusal rules.

The committee should ask: “What evidence would falsify our thesis?” and “Are we approving this company because it is strong, or because the fund has already invested time and reputation?” The sunk-cost fallacy is particularly dangerous after extensive diligence.

11. Applied Case: MedFlow Systems (Illustrative)

Background

MedFlow Systems is an illustrative early-growth company offering workflow software for outpatient clinics. It integrates appointment management, insurance documentation, and automated patient reminders. The company seeks $3 million to expand from one country into two neighbouring markets.

Problem

The fund must decide whether the company’s reported growth reflects durable customer value or temporary adoption driven by discounted pilots. The team also needs to determine whether the regional market is large enough for the fund’s return requirements.

Analysis

The founders include a former clinic administrator and a software engineer who previously built healthcare integration tools. Customer interviews confirm that documentation delays and missed appointments impose measurable costs. However, implementation requires integration with local systems, creating a scaling constraint.

The company reports 220 paying clinics, an annual contract value of $10,000, and 92% gross revenue retention over the most recent annual cohort. Several large customers expanded usage, but the company has not yet demonstrated consistent retention in the second country.

The bottom-up estimate identifies 4,000 eligible clinics across the initial three-country region. At $10,000 per year, the theoretical market is $40 million annually. Regulatory and integration constraints reduce the immediately serviceable population to 2,400 clinics, producing a SAM of $24 million. Management’s five-year target of 600 clinics would represent $6 million in annual recurring revenue before discounts and churn. This target is plausible but depends on hiring implementation teams and securing local integrations.

The principal risks are regulatory variation, long sales cycles, dependence on a small number of integration partners, and the possibility that retention weakens after promotional pricing ends. Defensibility is currently moderate: workflow data and integrations create switching costs, but no clear network effect exists.

Approach and recommendation

The analyst recommends proceeding conditionally rather than investing immediately. Conditions include verified cohort retention, legal review of data processing, signed integration agreements in the expansion markets, and evidence that implementation capacity can support the planned sales volume. The committee may approve a smaller initial tranche linked to milestones, with pro-rata rights subject to satisfactory progress.

Outcome and learning points

The case illustrates that a credible founder, useful product, and growing revenue do not eliminate execution risk. Market size must be connected to actual eligibility and capacity. The correct decision may be “advance with conditions,” not an artificial binary judgment. The memo should make uncertainty visible and translate it into diligence tasks, financing structure, and governance rights.

12. Ethical, Regulatory, and Technology Considerations

Investors handle sensitive financial, personal, technical, and customer information. Confidentiality agreements, secure data rooms, access controls, retention policies, and documented permissions are essential. Personal data should be assessed for lawful collection, purpose limitation, security, and appropriate use under applicable data-protection law.

Potential conflicts include investing in competing companies, using one founder’s confidential information to assist another, accepting undisclosed benefits, or pressuring founders through information asymmetry. Clear conflict policies and fair communication protect both the fund and the entrepreneurial ecosystem.

AI tools can assist with market mapping, document classification, anomaly detection, and synthesis of publicly available information. They should not replace primary diligence. Generated summaries may omit caveats, reproduce bias, or misclassify confidential information. Analysts should verify sources, avoid uploading restricted data to unapproved systems, maintain an audit trail, and disclose material use of automated analysis where governance requires it.

13. Best-Practice Checklist

Before an investment-committee meeting, the team should be able to answer:

• Does the company fit the fund mandate and portfolio construction needs?
• What is the investment thesis, and what evidence would disprove it?
• Are founder claims independently verified?
• Are product usage, retention, revenue, and margins defined consistently?
• Is the market estimate bottom-up, transparent, and time-bound?
• What limits scalability?
• What prevents competitors from replicating the offering?
• Which risks are fatal, material, or manageable?
• What are the valuation, dilution, ownership, and reserve implications?
• What governance rights and milestone conditions are required?
• Have conflicts and cognitive biases been actively managed?
• What information remains outstanding, and who owns each diligence task?

Chapter Summary

Deal sourcing and investment decisions require a designed process rather than opportunistic networking. Effective funds diversify sourcing channels, measure pipeline quality, and apply staged screening consistent with their mandate. Evaluation should integrate founder–market fit, customer problem, product evidence, traction, market size, defensibility, scalability, and economics.

TAM, SAM, and SOM become useful only when their boundaries and assumptions are explicit. Bottom-up analysis links customers and prices to a measurable opportunity and is generally more decision-useful than an unsupported industry statistic. Scoring models improve consistency but must be supported by evidence, decision gates, and professional judgment.

The investment memo converts analysis into an accountable recommendation. A strong committee process tests the thesis, records uncertainty, challenges assumptions, and controls bias. Ethical handling of confidential information, conflicts, data protection, and AI tools is part of investment competence, not an optional addition.

Key Terms

Deal sourcing: The systematic identification and development of potential investment opportunities.

Investment mandate: The fund’s defined boundaries for stage, sector, geography, cheque size, and strategy.

Triage: An initial process for determining whether an opportunity merits further review.

Founder–market fit: Alignment between the founding team’s capabilities and the market’s requirements.

Traction: Evidence of customer adoption, usage, payment, retention, or commercial progress.

Cohort: A group of customers sharing a common start date or characteristic for performance analysis.

MRR: Monthly recurring revenue expected from recurring customer contracts.

NRR: Net revenue retention, measuring recurring revenue retained and expanded from an opening cohort.

TAM: Total Addressable Market under stated assumptions.

SAM: Serviceable Available Market within the company’s practical reach.

SOM: Serviceable Obtainable Market realistically attainable over a defined period.

Bottom-up analysis: Market estimation built from customer units, usage, and prices.

Defensibility: The capacity to preserve value and returns against competition.

Scalability: The ability to grow output or revenue without proportional resource increases.

Contribution margin: Selling price minus variable cost per unit.

Investment thesis: A testable explanation of why an investment may generate attractive returns.

Investment memo: A structured document presenting evidence, risks, assumptions, and a recommendation.

Decision gate: A condition that must be satisfied regardless of the average score.

Confirmation bias: The tendency to favour evidence supporting an existing belief.

Pre-mortem: An exercise imagining that a decision failed and identifying why.

Review Questions

5 Short Answer Questions

1. What is the difference between a sourcing activity metric and an outcome metric?
2. Define founder–market fit and identify two forms of evidence relevant to assessing it.
3. Distinguish TAM, SAM, and SOM.
4. Why should investors examine cohort data rather than aggregate revenue alone?
5. State two methods for controlling confirmation bias in an investment committee.

5 Long Answer Questions

1. Explain how a VC fund can design a diversified deal-sourcing network and evaluate its effectiveness.
2. Discuss the relationship between product quality, traction, customer value, and scalability.
3. Explain bottom-up market sizing, including its assumptions, strengths, and limitations.
4. Describe the structure and purpose of a professional investment memo.
5. Critically evaluate the use of weighted scoring models in venture investment decisions.

3 Analytical / Application Questions

1. A company identifies 8,000 potential customers at an annual price of $6,000. Regulatory restrictions reduce the eligible population to 2,500, and management expects to serve 125 customers within five years. Calculate the implied TAM, SAM, and five-year SOM, then identify two assumptions requiring diligence.
2. Construct an investment-committee recommendation for a venture with excellent founders and rapid user growth but declining retention, unclear gross margins, and a market estimate based only on a global industry report. Specify the conditions under which you would proceed.
3. Design a bias-control process for a committee considering a highly charismatic founder whose company was introduced by an influential existing investor. Explain how your process would protect analytical quality and fairness.

Multiple Choice Questions with Answers

Question 1

Which measure best represents an outcome metric for a VC sourcing process?

A. Number of networking events attended

B. Number of founder emails sent

C. Number of investments resulting from qualified opportunities

D. Number of internal sourcing meetings

Correct Answer: C. Number of investments resulting from qualified opportunities

Question 2

Which statement correctly distinguishes SAM from SOM?

A. SAM is the total global industry; SOM is the company’s valuation

B. SAM is the practically serviceable market; SOM is the realistically obtainable portion over a defined period

C. SAM is current revenue; SOM is historical revenue

D. SAM measures founders; SOM measures competitors

Correct Answer: B. SAM is the practically serviceable market; SOM is the realistically obtainable portion over a defined period

Question 3

A venture has 2,000 eligible customers and an annual contract value of $15,000. What is the estimated annual market revenue under the stated assumptions?

A. $30,000

B. $300,000

C. $3,000,000

D. $30,000,000

Correct Answer: C. $3,000,000

Question 4

What does NRR above 100% generally indicate for a defined customer cohort?

A. The cohort’s recurring revenue expanded overall

B. Every customer renewed at a higher price

C. The company has no customer churn

D. New-customer acquisition cost is zero

Correct Answer: A. The cohort’s recurring revenue expanded overall

Question 5

Which is the best example of defensibility?

A. A founder’s attractive presentation style

B. A temporary discount offered to early customers

C. A legally and economically meaningful integration that creates switching costs

D. A large but unsupported TAM estimate

Correct Answer: C. A legally and economically meaningful integration that creates switching costs

Question 6

In the weighted-score formula, what must the criterion weights generally sum to when expressed as decimals?

A. 0.1

B. 1

C. 5

D. 100

Correct Answer: B. 1

Question 7

Which practice most directly reduces anchoring bias in an investment committee?

A. Allowing the lead investor to speak first and last

B. Using independent preliminary assessments before group discussion

C. Ignoring all previous investments

D. Selecting only the most optimistic forecast

Correct Answer: B. Using independent preliminary assessments before group discussion

Question 8

What is the principal purpose of an investment thesis?

A. To guarantee an exit valuation

B. To provide a testable explanation of why the investment may generate attractive returns

C. To replace financial diligence

D. To describe the fund’s office procedures

Correct Answer: B. To provide a testable explanation of why the investment may generate attractive returns

Question 9

Which statement about a bottom-up market estimate is correct?

A. It multiplies relevant customer units by a defined revenue or price assumption

B. It always produces a larger market than a top-down estimate

C. It does not require assumptions about customers or price

D. It is equivalent to a company’s valuation

Correct Answer: A. It multiplies relevant customer units by a defined revenue or price assumption

Question 10

Why should an investment committee use decision gates in addition to an average score?

A. To ensure a fatal legal or regulatory issue cannot be hidden by strengths elsewhere

B. To eliminate the need for diligence

C. To guarantee that all investments receive identical terms

D. To convert qualitative judgment into a risk-free decision

Correct Answer: A. To ensure a fatal legal or regulatory issue cannot be hidden by strengths elsewhere