Portfolio Construction

Portfolio construction is the process of choosing check sizes, follow-on reserves, and assumptions about valuations, ownership, and dilution over time. There's no single right way — the right level of detail depends on what decisions you'll make with the model.

Trade-offs

Portfolio construction is a set of choices:

• Concentrate or diversify? (fewer large checks or more smaller ones)
• What check size fits your strategy and position?
• Reserves for follow-ons reduce initial investments — concentrating into (hopefully) winners.
• How large does an investment need to be for a fund-returner?
• Same check size across investments, or varied by conviction?
• What do you need to get access to the best deals — and does that match your check size?

The result should give you enough shots to hit a winner. Invest in enough companies (30-100) to catch the power law of venture returns.

Top funds win on (1) volume of large exits and (2) size of the largest exits. Concentrated portfolios risk missing the hit; diversified portfolios increase the odds of a fund-returner.

Mismatch between fund size, check size, and investments

Fund strategies often don't align with the math of deployable capital. Common failures:

• A $10mm fund (deployable $8mm) writing $1mm first checks — too few investments.
• A $20mm fund reserving 70% for follow-ons — struggles to deploy reserves into meaningful proratas.

Follow-on budgeting depends on (a) projected cap tables for portfolio companies (drives your prorata strategy) and (b) graduation rates (drives how many follow-ons you can do across stages). Misalignment between strategy and math is the most common portfolio-construction problem.

In the end, about 50% of startups will get from seed to series A.

Pace matters a lot, but the eventual final graduation rate always seems to end up around 50% pic.twitter.com/Uk9Pym8PzX

— Peter Walker (@PeterJ_Walker) December 5, 2024

Mismatch between check size and fund strategy

Charles Hudson: "your check size dictates your investment strategy". A $50k fund will struggle to get into Series B. A $250k lead-investor strategy will struggle to win deals. A $1.5mm early-stage fund needs the team, process, and experience to lead. Check size and strategy need to match.

Mismatch between check size and target ownership

Eniac's Seed Fund Portfolio Construction for Dummies:

The classic mistake here is new managers pick an amount like 250K or 750K when they should be pegging initial checks to a target ownership. At the end of the day, the percentage you own of a portfolio company when it exits is what's important, not the amount you put in. ... Part of the rationale for [our target] ownership range [of 10-15%] is that if we maintain ownership in the 10% range when the company is worth $1B, an exit at that time can return the whole fund.

Accounting for dilution and reserves for proratas, what ownership % do you need at exit for a fund-returner? The Venture Capital Method and the Venture Valuation Tool can help work the math.

How to model portfolio construction

Portfolio construction drives capital deployment and exit proceeds, both timing and amount. Hemrock's venture capital models span simple to detailed:

• Venture Capital Model, Overall Forecast — simplest. Fund-level performance, no per-period cash flows, average gross multiple on invested capital.
• Venture Capital Model, Annual Forecast — annual investments, proceeds, and distributions up to 20 years. Average investment approach, average exit multiple. Also in Causal.
• Venture Studio Model — same core as the Annual Forecast plus a dual-entity studio + fund structure.
• Venture Capital Model, Manual Portfolio — input each investment specifically (check, date, follow-on, proceeds) and aggregate.
• Venture Capital Model, Quarterly Forecast — detailed initial and follow-on strategy per exit outcome, with up to three portfolio scenarios.
• Venture Capital Model — quarterly, with graduation rates, follow-ons, proratas, valuation increases, dilution, and per-stage exit proceeds. Detailed expected-value math per stage.

More detail isn't always better

The right level of detail depends on your situation.

Building your first model for first-LP conversations on Fund I? A complicated forecast of a 4x gross isn't what LPs care about at that stage. Deeper in the process, or raising Fund II or III, detail becomes more valuable — and you're better positioned to build and defend it.

Scenarios capture variability

Returns in venture are highly variable. Scenarios and range-based inputs are valuable. Use discrete scenarios (best/worst/medium) or range-based simulation with probability distributions.