The case for multi-asset investing is intuitive. Opportunities and risks rarely respect asset-class boundaries. Yet implementing a genuinely integrated strategy remains considerably harder than designing one. The constraint is often not the investment idea itself, but the infrastructure connecting data, portfolio construction, execution, risk and performance measurement.

This challenge is increasingly relevant as institutional investors explore the total portfolio approach (TPA). Rather than managing asset classes as largely independent sleeves, TPA considers each investment decision in the context of its contribution to the objectives, risk and return of the overall portfolio. Making that shift requires infrastructure capable of supporting a genuinely total-portfolio view.

It starts with data. Equities, bonds, commodities, derivative overlays and private assets come with different identifiers, conventions, frequencies and liquidity characteristics. Bringing them into the same portfolio means more than aggregating data sources. Positions need to be mapped onto consistent economic exposures - rates, currencies, credit, equity markets and other systematic drivers - so that different instruments can be compared on a common basis. Factor-based frameworks can be particularly useful here, providing a common language for understanding the underlying drivers of risk and return across asset classes.

Execution is where this common language becomes actionable. The same investment view can often be expressed through very different instruments, so implementation choices need to be evaluated in terms of exposure rather than asset-class. Consider a portfolio manager seeking to increase sensitivity to falling U.S. interest rates. That exposure could be obtained by buying Treasurys, adding Treasury futures or through an interest-rate swap. Each may deliver broadly similar duration exposure, but with different transaction costs, liquidity, funding and collateral requirements. A manager should therefore be able to compare those alternatives on a consistent risk basis and determine how each trade changes the exposures of the total portfolio, rather than optimizing execution within an isolated fixed-income sleeve. In a TPA framework, implementation becomes a portfolio-construction decision in its own right.

Risk and attribution complete the plumbing. Looking at asset-class risk independently can obscure exposures that offset, or reinforce, one another. Integrated risk needs to capture common drivers and cross-asset relationships at the total-portfolio level, particularly as correlations and diversification assumptions change. But measurement should not stop at ex-ante risk. Total-portfolio attribution should close the loop, linking realized outcomes back to the investment decisions and underlying exposures that generated them. This creates a feedback mechanism between portfolio construction, implementation and subsequent decision-making.

Ultimately, cross-asset investing - and TPA in particular - depends on creating a consistent chain from data to exposures, exposures to decisions, decisions to trades and outcomes back to risk and attribution. The better that plumbing works, the more effectively investors can manage the portfolio as a whole, rather than as a collection of asset-class silos.