Wealth reporting and performance analytics for family offices

The difference between a report and an insight

A monthly performance report tells the principal that the total portfolio returned 1.4% in the period, that equities outperformed bonds, and that the private equity sleeve remains marked at cost pending the next valuation cycle. It is accurate, it is timely, and it is almost entirely useless for making forward-looking decisions. The report is a photograph. Performance analytics is the diagnostic that tells you whether the subject is healthy, and why.

The distinction matters because family offices have proliferated in complexity faster than their analytical infrastructure has kept pace. According to broad industry surveys, the median single-family office now holds between eight and fourteen distinct asset class exposures, including direct real estate, private credit, hedge funds, co-investments, listed equities, fixed income, and cash equivalents. Coordinating the return streams from that many sleeves into a single consolidated view is genuinely difficult. Most offices stop there. The ones that convert data into decisions go one layer deeper.

Attribution analysis: the foundational layer

The Brinson-Hood-Beebower framework, first published in the Financial Analysts Journal in 1986, remains the standard decomposition model for multi-asset portfolios. It separates total active return into three components: the asset allocation effect (did you overweight or underweight the right asset classes relative to benchmark?), the security selection effect (did individual managers or holdings outperform within their assigned sleeve?), and the interaction effect (the cross-product of allocation and selection decisions, which captures cases where you overweighted an asset class and your manager within it also outperformed).

In practice, a family office running a 60-40 equity-bond benchmark might find that over a 12-month period, asset allocation contributed positive 80 basis points (because it overweighted equities in a rising market), security selection subtracted 40 basis points (because the equity manager underperformed the equity benchmark), and the interaction effect added 10 basis points. Net active return: positive 50 basis points. Without attribution, the family sees only outperformance and congratulates the manager. With attribution, it sees that the manager actually destroyed value, and that the gain came from a tactical allocation call that may or may not repeat.

Attribution analysis frequently reverses the narrative that raw return numbers suggest. A portfolio that outperforms its benchmark may be doing so despite, rather than because of, its active management decisions.

Applying attribution to private and illiquid assets

The complication for family offices is that Brinson-Hood-Beebower was designed for liquid, publicly priced portfolios. Private equity, real assets, and private credit require modified approaches. For private equity, the standard is to benchmark against a public market equivalent (PME), comparing the internal rate of return of the private portfolio against the hypothetical return of investing identical cash flows into a public index such as the MSCI World or the Russell 2000. A PME ratio above 1.0 indicates that the private allocation genuinely outperformed public markets net of the illiquidity premium.

Real estate attribution is more granular still, decomposing returns into income return (rental yield), capital appreciation, and currency effect for cross-border holdings. A European family office with U.S. commercial real estate exposure may show strong local currency returns that are partially or fully offset by euro appreciation, a distinction that matters considerably for total portfolio construction and hedging decisions.

Factor exposure and the question of true alpha

Attribution answers where returns came from across sleeves. Factor analysis answers a harder question: how much of any manager's return is explained by systematic, priced risks rather than skill? The canonical factor model, rooted in the Fama-French three-factor framework and subsequently extended by Carhart (1997) to include momentum, and by others to include quality and low-volatility factors, allows a family office to regress any manager's return series against publicly available factor returns.

The practical output is a factor loading table. A manager charging 100 basis points in management fees who shows a statistically significant loading on the value factor and the small-cap factor but an alpha close to zero is, in effect, selling systematic factor exposure at active management prices. The same exposure can be accessed through factor-tilted index funds at a fraction of the cost, typically 10 to 20 basis points. The family office that does not run this analysis will never know it is overpaying.

Factor analysis also surfaces unintended portfolio-level risks. A family office holding four equity managers may believe it is diversified across styles and geographies. A factor regression across the combined equity book may reveal a concentrated loading on growth and momentum factors, meaning the aggregate portfolio behaves far more like a single growth-oriented strategy than its surface-level diversification suggests. This kind of hidden factor concentration was a meaningful source of drawdown risk in 2022, when both growth and momentum factors sold off simultaneously.

Currency and interest rate factor exposures

For family offices with multi-currency balance sheets, factor analysis should extend beyond equity risk premia to include currency beta and duration exposure. A portfolio ostensibly denominated in Swiss francs but holding 40% of its assets in dollar-denominated private funds and U.S. listed equities carries a structural long-dollar position. If that position is unintentional and unhedged, it is a factor exposure that has never been debated in an investment committee meeting. The analytics layer makes it visible; governance then decides whether to retain, reduce, or hedge it.

Manager-versus-benchmark contribution analysis

The third analytical layer, and the one with the most direct implications for manager selection and termination decisions, is contribution analysis at the manager level. This differs from attribution in that it asks not just whether a manager added value, but how much of the total portfolio's outcome they were responsible for, in absolute basis points.

A useful framework is to express each manager's contribution as the product of three terms: the weight of that manager in the total portfolio, the manager's active return relative to their assigned benchmark, and an adjustment for the correlation of that active return to the rest of the portfolio. A manager running 15% of a total portfolio who beats their benchmark by 120 basis points contributes 18 basis points at the portfolio level before correlation adjustment. A manager running 5% who beats by 300 basis points contributes 15 basis points. Sizing matters as much as skill.

The manager who generates the highest return in isolation may be contributing less to total portfolio value-add than a steadier, better-sized allocation. Contribution analysis restores the correct frame of reference.

This framework has direct governance implications. Investment committees that evaluate managers in isolation, comparing each to their own benchmark without reference to portfolio-level contribution, will systematically retain high-tracking-error, high-fee managers whose actual contribution to total wealth is modest. Contribution analysis reframes the evaluation as a portfolio construction question rather than a manager selection contest.

Reporting cadence and the tiered analytics model

Not all analytics need to be produced at the same frequency, and attempting to do so wastes resources and numbs decision-makers to the data. A practical tiered model maps cadence to asset class liquidity. Listed equity and fixed income portfolios can support daily return attribution against a live benchmark, with weekly factor exposure updates. Semi-liquid strategies, including hedge funds and evergreen private credit funds, should be reported monthly with a one-period lag reflecting NAV publication timelines. Fully illiquid assets, including closed-end private equity and real estate funds, warrant quarterly reporting tied to the GP's valuation cycle, supplemented by PME calculations updated each time cash flows occur.

Annual deep-dives serve a separate purpose from periodic reporting: they allow the investment committee to assess whether the overall strategic asset allocation remains appropriate given changes in the family's liability structure, spending rate, and risk appetite. The analytics produced in these reviews should include rolling three-year and five-year attribution, not just the most recent 12 months, to smooth out the noise of any single market regime. A manager who underperforms over 12 months may be experiencing a style headwind rather than a skill deficit; attribution over a full market cycle provides far more reliable evidence.

Governance integration: the investment committee report

Performance analytics generates value only when it reaches decision-makers in a format that prompts action. Best practice is to structure the investment committee report in four sections: a one-page total portfolio summary including attribution waterfall; a manager contribution table ranked by portfolio-level basis points added or subtracted; a factor exposure dashboard showing current and trailing loadings across equity, duration, credit spread, and currency factors; and a forward-looking agenda item that proposes at least one governance action based on the analytics, whether that is a manager review, a rebalancing instruction, or a hedge initiation.

The governance action requirement is non-negotiable. An investment committee that receives attribution analysis and takes no action has spent time on a reporting exercise rather than a decision-making process. The chief investment officer or lead advisor bears responsibility for ensuring that each analytics cycle closes with a documented decision, even if that decision is to maintain current allocations for articulated reasons. Under MiFID II suitability requirements for investment managers, and increasingly under the AIFMD governance standards applicable to family offices structured as AIFs, documented rationale for portfolio decisions is not merely best practice but a regulatory expectation.

Building versus buying the analytical capability

Single-family offices face a build-or-buy decision for their analytics infrastructure. The honest answer is that the decision is not binary. The data aggregation and custodian reconciliation layer is a commodity function and should be outsourced or automated at low cost. The attribution, factor analysis, and contribution reporting layer requires judgment, model selection, and benchmark governance, and benefits from internal ownership even if the computation is handled externally.

A senior analyst dedicated to performance analytics at a family office with assets under management of 500 million euros or more represents an all-in cost of approximately 150,000 to 250,000 euros annually. Across a 500 million euro portfolio, that is 30 to 50 basis points, a meaningful but defensible cost when the analytics discipline allows the family to identify even one underperforming manager relationship representing 5% of the portfolio and 100 basis points of annual drag. The savings from that single decision, 50 basis points on 25 million euros, or 125,000 euros per year, recover the analytical investment within two years.

The families that treat reporting as an end rather than a starting point will continue to generate polished documents that tell them what happened last month. The families that invest in the analytical layer will increasingly understand why their portfolio performs as it does, and will be better positioned to act before the next cycle's numbers arrive.