Ask a room full of financial advisors what makes an “optimal” portfolio, and aside from “meeting client objectives,” most will gravitate toward a familiar answer: it has to be diversified.
Ask them to define that, and the conversation usually turns to numbers. Ten holdings? Fifteen? Twenty? At some point, the debate lands on Markowitz, efficient frontiers and the elimination of idiosyncratic risk. It’s an intellectually tidy framework, and not without merit. But in the context of private markets, where capital is less liquid, returns are dispersed, and the opportunity set may look nothing like a basket of public equities. The conventional diversification approaches can fall dangerously short.
I had this conversation with a sophisticated advisor not long ago. He wanted to debate the right number of private fund investments for a client’s alternatives sleeve. He was thinking about the problem the way most people do: the more investments, the more diversified and therefore the closer to optimal. I pushed back. “What if all 15 of those investments are doing the same thing?” He conceded the point but reframed: “Okay, assume they’re all doing something meaningfully different. Now we’re diversified, right?”
Not necessarily.
Consider this: You have 15 private fund investments of roughly equal size. Each pursues a distinct strategy. But when you look under the hood, one investment is generating 90% of the portfolio’s returns. Another is responsible for 90% of its risk. The remaining 13 are just … there. They’re not hurting anything, but they’re not contributing much either. Capital is sitting in positions that neither drive returns nor manage risk in a meaningful way. That is not an optimal portfolio. That is a diversified-looking portfolio with a severe capital-efficiency problem.
The number of investments is almost irrelevant. What matters is what each investment contributes: to returns, to risk and to the portfolio’s overall diversification benefit. These are three distinct and measurable things, and treating them as synonymous leads to portfolios that may look “optimal” on paper but are poorly constructed in practice.
Rethinking Diversification as a Quantifiable Property
Return attribution is a concept advisors are comfortable with. If a client’s portfolio generated 12% last year and one holding was responsible for six percentage points of that, you know something important about concentration. You can hold that contribution against the position’s weight, and if that holding was only 10% of the portfolio, it punched well above its class. If it were 60% of the portfolio, the math would look a lot less impressive.
What most portfolio construction frameworks don’t do, at least not in the private markets context where advisors are working largely without institutional-grade tools, is apply that same attribution logic to risk and diversification. Both of those properties can and should be decomposed the same way.
Using a standard institutional risk-attribution technique, it is possible to calculate each holding’s actual contribution to total portfolio risk (different from its standalone volatility). A fund with high standalone volatility may actually contribute relatively little to portfolio risk if it has low or negative correlation with the rest of the sleeve. Conversely, a fund that appears moderate in isolation can be a significant risk concentrator if it moves in lockstep with everything around it.
From there, it is a short step to quantifying diversification contribution itself: the difference between what a fund’s standalone risk share would be if correlations were perfect, and what it contributes given real-world correlations. A positive diversification contribution means a fund is reducing portfolio risk relative to its size. A negative contribution means it is a risk concentrator, adding more to portfolio risk than its weight alone would suggest. Most advisors have never seen this calculated explicitly. Most portfolio reporting tools don’t offer it.
What the Optimal Portfolio Actually Looks Like
With this framework in place, a cleaner definition of the optimally diversified portfolio becomes possible. It is the portfolio where the drivers of return, risk and diversification are efficiently and intentionally allocated across the capital base. The one where each dollar of allocation is earning its place, contributing to return in proportion to its weight, contributing to or reducing risk in a deliberate way, and delivering the diversification benefit you thought you were getting when you constructed the portfolio.
This standard is achievable. It requires looking at three things simultaneously across every position: return contribution relative to weight, risk contribution relative to weight, and diversification contribution as a signed value showing whether, and to what extent, a fund is a diversifier or a concentrator. The interplay between these three tells you far more than any of them in isolation.
In private markets, where most advisors are often building sleeves of three-to-seven illiquid positions with multi-year lock-up periods, this kind of visibility is not an analytical luxury. It is a fiduciary necessity.
The good news is that the mathematical infrastructure for this kind of analysis is well-established. Portfolio risk attribution using covariance-based decomposition has been a staple of institutional asset management for decades. The challenge has been making it operational and easy for independent advisors who are managing private market allocations across a fragmented landscape of managers, platforms and reporting formats.
A Higher Standard for Portfolio Construction
Private markets have matured significantly as an asset class. The next frontier is construction quality. Access without analytical rigor is not a solution. Rather, it is just a more convenient way to build a poorly optimized portfolio. As the private markets allocation within client portfolios grows, advisors will be increasingly differentiated not by which funds they can access, but by how well they understand what those funds are actually doing inside a portfolio, how they interact with each other, and whether the capital allocated to each position is earning its place.
The optimal portfolio is not the one with the most investments. It is not the one with the highest number of distinct strategy labels. It is the one where return, risk and diversification are all accounted for at the position level and where every dollar of allocation is doing deliberate work. That standard is demanding. It is also achievable, but only if the tools exist to measure it.
Advisors who hold their private market sleeves to that standard will build portfolios that are not just diversified in appearance but optimized in substance.