Portfolio performance is the only metric that shows comprehensive, actual, and independently confirmed revenues achieved by an optimizer with a battery storage asset. Indices use market data from various sources to calculate an average in relation to a previously established base number to provide guidance for market participants. Backtests also use data from the past, but they are highly modular, meaning the result can be distorted through the inclusion or exclusion of certain data.

Bankable forecasts focus on future viability and serve as proof of concept to secure project financing from banks and credit institutions. For reference, you can use this overview to understand the differences between portfolio performance, backtests, bankable forecasts, and indices at one glance.

As of early 2026, no other optimizer has made independently confirmed revenues publicly available. Other optimizers tend to use backtests to indicate revenue potential, but backtests are easy to manipulate, as low-revenue periods can be excluded to artificially inflate overall performance.

Currently, this gives you two options when choosing your optimizer: Real-world data from actual assets in a live optimization setup or approximations and claims without provable results in reality.

This depends on several factors, including available cycles, warranty terms, technical constraints, and commercial restrictions. Every asset is unique, and one of the great things about portfolio performance is that it puts us in a strong position to consult with warranty providers and financing partners on BESS parameters that benefit asset owners.

We refined the methodology to provide a portfolio performance view that is neutral to asset availability. It now applies a data-quality filter to exclude days where an asset was not operational for more than four hours. As a result, revenues from wholesale, aFRR, and FCR on these partial-availability days are removed, which can lead to slightly lower total revenues. However, the same filtering is also applied to the calculation of average marketable power (MW) and energy (MWh), meaning that averages are computed only on fully operational days. This creates a more representative performance metric.

Duration clustering was also refined. The 2h segment is now defined as 1.51-2.5h rather than extending to 3.5h. This change was requested by asset owners and market participants to create a cleaner benchmark that more closely represents a true 2h asset. The cluster description now also focuses on P10/P90 ranges rather than minimum/maximum values, reducing the influence of outliers.