Four good decisions, one bad outcome

A recipe that sounds sensible at every step

First a contract with a dynamic price, because "it works out better". Then solar, ideally with headroom — the surplus on sunny weekends will go to the grid anyway. And since energy is cheap at midday, let us add storage so we can use it in the evening. Finally we change the supply contract so that somebody takes the surplus off our hands.

Each of these steps can be defended on its own. The problem shows itself when we calculate the outcome of the whole — and ask when the payback arrives: will we still be around to enjoy it, or are we already thinking about our heirs?

A dynamic price does not give predictability over an investment horizon

With indexation to the day-ahead market, prices are known a day ahead, once the auction results are published. Contracts can also reference the intraday market, and B2B agreements often add a pass-through of balancing costs — in which case part of the price is only known after delivery. And it is the energy price that is dynamic; distribution and capacity charges are settled separately, under whatever tariffs apply.

Oversized solar produces when energy is worth the least

In the first quarter of 2026 the EU-27 recorded 1,223 hours with negative wholesale prices, against 593 a year earlier — more than double in twelve months. In Spain that was 16% of all trading hours, in a market that had never seen a single negative hour before 2023. Greece went from zero in the first quarter of 2025 to 138 hours. Over the first half of 2026, Spain reached 596 negative hours, Portugal 462 and France 370, driven by record solar output of 129 TWh in the second quarter alone.

Whether that becomes a cost or merely a lost opportunity depends on the contract. Support schemes for small installations usually put a floor under the settlement — a negative price is treated as zero, and the surplus is simply worth nothing. But once an installation crosses the threshold of such a scheme, or the sale price follows the market without a floor, a negative price stops being a lost opportunity and becomes a real cost: you pay to hand over the energy you produced.

That leaves storage — but is it the answer?

If storage is meant to earn on arbitrage, its result is the price spread less cycle losses, degradation and the cost of capital. Meanwhile, shifting part of the load into hours of production often requires no additional CAPEX at all and carries none of the conversion losses typical of a battery. A cold store, compressors, hot water, charging electric forklifts or cars — all of it can already provide flexibility today.

Which is why the question is not "export or battery?" but "do we fit the energy to the load, or the load to the energy?".

A schedule is a bet, control is a decision

A schedule set in advance is a bet on the weather and the price. Active, well-considered control is a decision taken optimally on the basis of the current day's data.

Because four decisions that are rational separately do not have to add up to a rational whole.