How should I explain my portfolio-construction strategy to LPs?

The chain, and why the order matters

Each link determines the next. A manager who can walk the chain in one direction and then back the other way has a construction; a manager who can only recite the numbers has a slide.

• Target ownership. What the strategy needs at entry, given the dilution you expect before an exit, and whether you need to lead to get it.
• Initial cheque. Ownership multiplied by your entry valuation, which is why stage and geography have to be settled first.
• Number of positions. Driven by your loss assumptions and your concentration tolerance rather than by a round number.
• Concentration. What your largest position is allowed to become as a share of the fund, which is a risk decision an allocator will ask about directly.
• Reserves. The ratio held back for follow-ons, which is the input first-time models most often omit and the one that decides whether you can defend your winners.
• Follow-on policy. When you follow, when you decline, and who decides, because reserves without a policy are a number rather than a plan.
• Pacing. Positions per year, which sets the investment period and your vintage exposure.
• Fund target. The output of everything above, grossed up for the fee and expense load, since committed capital is not investable capital.
• Stage and geography. Present throughout rather than at the end, because they set the entry valuations the whole chain runs on.

A worked example, with every assumption visible

The numbers below are assumptions chosen to make the chain legible, and they are the same ones Blueprint's fund sizing page uses so the two can be read together. They are not a recommendation, a benchmark, or a claim about what any market considers normal. Every output is hypothetical and follows only from the inputs stated.

• Assumption: seed entry at an average post-money valuation of 12 million dollars.
• Assumption: 8 percent target ownership at entry, which makes the initial cheque 960,000 dollars, rounded to 1 million.
• Assumption: 20 initial positions, which is 20 million dollars of initial capital.
• Assumption: reserves of 0.75 dollars per dollar of initial capital, which is 15 million, so investable capital is 35 million.
• Assumption: management fees and fund expenses consume 21 percent of commitments across the fund's life, so 79 percent reaches companies.
• Hypothetical result: 35 million divided by 0.79 is 44.3 million, so the fund target is about 45 million dollars.
• Hypothetical check on pace: 20 positions across a three-year investment period is about seven a year.
• Hypothetical check on concentration: if one position absorbs its initial cheque plus 3 million of reserves, it is roughly 9 percent of the fund, which is the number to have ready when an allocator asks about your largest exposure.

Say the word assumption out loud when you present this. A model whose inputs are labelled invites the allocator to test the inputs, which is a conversation about strategy. A model presented as a projection invites them to test whether you understand what a projection is.

Where the arithmetic and the thesis most often disagree

These are the five contradictions allocators find most often, and each is visible in a manager's own materials before any meeting.

• Ownership the entry stage cannot support. An 8 percent target at valuations where that requires leading every round, from a manager whose evidence is mostly participation.
• Position count the sourcing funnel cannot fill. Twenty positions a year from a channel map that produced twenty companies in total.
• Reserves that vanish under pressure. A reserve ratio in the model that the deployment plan spends on initial cheques by year two.
• Pacing that contradicts the stated discipline. A three-year investment period that requires a deal a month at a stage where the manager claims to be highly selective.
• A fund size that moved without the construction moving. The most common of the five, and the easiest to catch by comparing two versions of the same deck.

What to put in the deck and what to keep for diligence

Construction is one of the few areas where allocators want the detail, but the detail belongs in the right place.

• In the deck: the chain above on one slide, the fund target, the position count, the ownership range, the reserve ratio, and the pacing. Six numbers that reconcile.
• In the data room: the model itself, with editable assumptions, the loss-rate scenarios, and the follow-on policy in writing.
• In conversation: what you would change if the fund closes smaller, which every allocator asks and which is much better answered deliberately than improvised.
• Nowhere: projected fund returns presented as expectations. Modelled outcomes are hypothetical by construction, and presenting them to investors carries disclosure obligations in most jurisdictions.

What limited partners are testing

Portfolio construction is where an allocator decides whether the strategy is a plan or a wish.

• Does the arithmetic reconcile with the target, without the manager needing to recompute it in the room?
• Does the sourcing funnel support the position count at the stated pace?
• Is the reserve strategy real, meaning is there a written follow-on policy and a decision maker?
• What happens to the construction at 60 percent of target, which is the scenario a first fund most often lands in?
• Are the loss assumptions honest, or does the model only work if the failure rate is unusually low?

What this example does not prove

The example proves the method produces a coherent set of numbers. It proves nothing about what your fund should hold, because every input in it is an assumption chosen for legibility rather than a market observation, and every output is hypothetical.

A construction model is not a forecast and should never be presented as one. It is a statement of intent whose value to an allocator is that it makes the strategy checkable.

The library plans an interactive portfolio-construction tool in its final wave, alongside the fund-size viability calculator. It is named in the planned section of the question library hub rather than linked, because the page does not exist yet.

This page is educational and general. It is not legal, tax, securities, or investment advice.