In previous Insights, we have discussed how economic viability is increasingly determining which energy efficiency measures are implemented, and how portfolio modelling enables comparability between investments. However, profitability is not a sufficient basis for decision-making. It depends on assumptions, and when these change, so too does the outcome. The question is therefore not only which investments are profitable, but how robust that profitability is – both at the level of individual buildings and across the property portfolio as a whole.

Profitability is a function of assumptions

Investment calculations often appear precise. Net present value, internal rate of return and payback time convey an impression of accuracy. Yet the results depend on assumptions: about future energy prices, discount rates, time horizons, technical and economic lifetimes, and future operating conditions.

Profitability is therefore not an inherent property of a measure, but a result of the conditions assumed to apply.

An energy efficiency measure with a long service life may appear highly attractive under low discount rates and rising energy prices, yet considerably less compelling in a scenario with higher capital costs and stable energy prices. By contrast, measures with shorter payback periods are often less sensitive to changes in assumptions.

When profitability is used as a decision criterion, the transparency and consistency of assumptions become critical. Without this, calculations risk creating a false sense of precision.

Risk at the building level – variation in outcomes

At the building level, risk concerns how actual outcomes may deviate from what is expected.

Uncertainty may arise from several sources. Energy prices may evolve differently from forecasts. Technical performance may differ from stated and assumed values. Occupant behaviour may influence energy use in ways that are difficult to predict. Interactions between measures may also lead to outcomes that differ from those calculated in isolation.

This means that two measures with similar calculated profitability may have very different risk profiles. One may be robust across a wide range of assumptions, while another depends on more favourable conditions to deliver the same result.

Ranking measures solely based on net present value or internal rate of return therefore risks obscuring important differences in uncertainty.

From sensitivity to robustness

Addressing this uncertainty requires moving beyond single-point calculations. Sensitivity analysis makes it possible to systematically vary key assumptions and examine how outcomes are affected.

When energy prices, discount rates or lifetimes change, some measures remain stable while others quickly lose their profitability. A control or optimisation measure with a short payback period is often less affected by changing assumptions, whereas a more extensive building envelope measure may depend on sustained or rising energy prices to remain justified. The distinction is not trivial – it reveals something about the robustness of the investment.

In a context where capital is limited, robustness therefore becomes a key attribute. It is not only about maximising theoretical returns, but about the likelihood that those returns will actually be realised.

Risk at the portfolio level – more than the sum of its parts

When multiple buildings are considered together, the nature of risk changes.

A high-risk measure in a single building may be entirely reasonable within a broader portfolio, particularly if balanced by more robust investments. Conversely, a portfolio of seemingly “safe” measures may still be vulnerable if all investments depend on the same underlying assumptions, for example a specific trajectory for energy prices.

The portfolio perspective therefore enables a different type of analysis. Rather than simply ranking measures by expected net present value in individual buildings, the focus shifts to how investments interact across the portfolio.

In practice, this means combining measures with different risk and return profiles. Robust measures with stable but moderate returns, such as optimisation of operation and control, can form the backbone of an investment programme. More extensive measures, which are more sensitive to energy prices and discount rates, are implemented more selectively where conditions are favourable.

In practical terms, this often involves ensuring that a significant share of total investment performs well even under conservative assumptions, while more risk-exposed measures are limited to a smaller portion of the portfolio.

By structuring investments in this way, dependence on individual assumptions is reduced. The result is a portfolio that is less sensitive to variations in, for example, energy prices or capital costs. It is in this sense that the whole can be more stable than its individual components.

Portfolio modelling as a tool for risk management

Portfolio modelling provides a framework for handling this complexity. When all buildings are analysed using the same technical model and the same economic assumptions, investments can be compared on equal terms – not only in a base case, but across multiple scenarios.

In practice, this means applying a consistent set of assumptions – for example regarding energy prices, discount rates and lifetimes – to all measures across the portfolio. These assumptions can then be varied systematically. What happens to the ranking of investments if energy prices develop more slowly? Which measures retain a positive net present value under higher discount rates? And which fall away?

This approach reveals differences in robustness. Some measures prove stable across a wide range of conditions, while others depend on specific assumptions. Only through this comparative analysis does it become possible to weigh return against risk at the portfolio level.

Without this type of analysis, portfolio decisions risk being based on individual calculations that are not fully comparable.

From profitability to robust capital allocation

When economic viability is the starting point for decision-making, the next step is to ensure that this viability is robust. Energy efficiency can therefore no longer be reduced to identifying profitable measures, but must include a systematic assessment of risk.

At the building level, this means understanding variation in outcomes. At the portfolio level, it means balancing risk and return across the entire stock. Only when these two levels are connected can energy investments form a stable part of a company’s capital allocation.

In this context, portfolio modelling evolves from an analytical tool into an instrument for strategic decision-making.

Profitability determines whether a measure is possible. Robustness determines whether it is rational.