Who carries the risk in a BESS agreement?

When a company buys an industrial BESS system (Battery Energy Storage System) and enters an agreement that lets a specialist commercially optimise the system's flexibility, one of the first questions from a CFO or a board is entirely natural: who actually carries the risk if something goes wrong?

It is a fair question, because a BESS agreement combines several things at once. The customer owns a physical asset. The system sits on the customer's property. And the earnings come from energy markets that move. Risk is therefore not one single concept, but several different types that belong in different places.

This article splits the risk into four categories: market risk, site risk, technical risk, and extraordinary risk. For each category we explain who, under the standard model, carries it, and what the main assumptions are. The purpose is not to promise that nothing can go wrong. The purpose is to make it transparent where responsibility sits, so the decision can be taken on an informed basis.

All the contract terms mentioned here come from our standard agreement for the six-year model. The specific wording, deadlines, and limits always appear in the signed agreement and its annexes, and that is what applies in the individual case.

Category 1: Market risk

Market risk is the risk that earnings from the system end up lower than expected because prices in the markets the battery participates in change.

That risk is real. A battery in Denmark typically earns money by delivering ancillary services such as aFRR and mFRR, which are reserves Energinet, the Danish transmission system operator, procures to keep the balance between electricity consumption and production on the grid. Prices for these services are not fixed. They depend on supply and demand and can swing considerably. As more batteries have entered the market, prices for reserve capacity have fallen in several places because competition has increased, and the industry itself notes that the market for frequency regulation can quickly become saturated, and that revenue from ancillary services therefore cannot stand alone in a robust business model.

In our standard model, by default we carry this ordinary market risk. The customer receives a fixed contractual availability payment, calculated as the contractual Calculation Basis divided by 72 and paid monthly. Ordinary swings in market prices, reserve prices, volatility, and trading performance do not in themselves give a right to reduce that payment. If market revenues in a period are lower than the customer's fixed payment, that is in principle our concern, not the customer's.

The logic is straightforward. Our earnings depend on our being able to optimise the system better than the fixed payment we have committed to the customer. That is why we have both the interest in and the risk of the market, while the customer gets a more predictable cash flow. In return, the customer does not get market upside during the six years.

It is important to be precise here. Carrying ordinary market risk does not mean the payment is without reservations. The fixed payment can be suspended, reduced, or deferred in the cases the agreement expressly describes, for example customer-caused unavailability, missing insurance, physical damage, termination, or extraordinary circumstances. Ordinary market swings do not change the customer's fixed payment, but extraordinary circumstances can affect the agreement under the specific contract terms. We go into that under category 4.

Category 2: Site risk

Site risk covers everything tied to the customer's own property and installation: the physical location, access to the system, the customer's switchboard and grid connection, insurance, and any technical changes the customer makes themselves.

Here the default allocation is reversed. The customer owns the asset and is closest to controlling their own site, and therefore the customer by default carries the risk for it. In practice, that means the customer provides a suitable location, reasonable access for service and fault correction, sufficient switchboard and grid connection capacity, and the permits the customer is closest to obtaining. The customer also maintains relevant insurance on the system, which is a precondition for start of operation and a project requirement in the agreement's insurance annex.

There is a practical reason for this allocation. We cannot be responsible for conditions we do not control. If a customer digs into their own cables, rebuilds their switchboard, or changes their grid connection without notifying us, and that makes the system unavailable for commercial optimisation, the fixed availability payment can be reduced or suspended proportionally for the period and the capacity affected. The customer is free to develop their business and change their electricity consumption, but must notify us in reasonable time if a change may affect the system's metering, market access, or operation.

Site risk is therefore the category the customer has most direct influence over. The better the site conditions are documented and maintained, the smaller the risk that a local condition interrupts operation. Short interruptions, for example a planned service window or a few hours of unavailability, do not in themselves cause an entire month's payment to lapse. It is the more lasting conditions the customer carries the risk for that can affect the payment, and always only proportionally for the affected period and capacity.

Category 3: Technical risk

Technical risk is the risk of faults, damage, or reduced performance on the system itself and the related technical components. It is not allocated by a fixed key, but by cause: is the fault due to us, the manufacturer, the installation, the customer, or an external condition?

The principle is that the party that caused the fault carries it. If a fault is due to our control, dispatch, integration, or software, we remedy it, including inadequate metering data that originates from our own equipment. If a fault is due to the customer's intervention, the customer's site conditions, or the customer's infrastructure, responsibility sits with the customer. Faults in the battery's cells themselves may be covered by the manufacturer's warranty, and damage may be covered by insurance, where the excess is in principle paid by the party that caused the damage.

A specific example is the battery's State of Health (SoH), which is a measure of how much of the original capacity remains. Our model is built around an SoH threshold of 80% at the end of the agreement period and a ceiling of 6,000 full equivalent cycles within the manufacturer's framework. If the system at expiry falls below 80% because our commercial use has exceeded the agreed cycle or warranty limits, we must at our own cost bring the system back to at least 80%. If it falls below 80% without our having exceeded the limits, it is handled under the manufacturer warranty. The actual remaining capacity after six years depends on the documented SoH and the operating history and cannot be set generally in advance.

Technical risk is therefore not a question of who owns the system, but of what actually caused a problem. That requires documentation, and therefore metering, monitoring, and a technical determination of SoH are fixed elements of the agreement.

Category 4: Extraordinary risk

The fourth category covers the rare but serious circumstances that sit outside both parties' normal control: force majeure, legislative changes, official orders, cyberattacks on critical infrastructure, and grid collapse.

These circumstances are real. A regional grid collapse is rare, but it can happen: the power outage on the Iberian Peninsula on 28 April 2025 was the most severe blackout on the European power system in more than 20 years. Cyberattacks on critical infrastructure are also a recognised risk, and there are today Danish rules on preparedness and cybersecurity for companies that are material to energy supply and energy markets, which together with the NIS 2 rules set requirements for, among other things, supply-critical systems.

The central point is how such circumstances are handled in the agreement. They do not trigger an automatic lapse of all obligations. Instead a proportional principle applies: if an extraordinary circumstance makes it illegal, technically impossible, or unsafe from a safety standpoint to use the system commercially, the parties must first try to find a practical solution, for example a technical adaptation or new market access. Only if the circumstance cannot reasonably be remedied can the affected part of the agreement be adjusted, suspended, reduced, or deferred, and only for the actual period and the actual capacity it affects. Payments already due are in principle not affected.

The difference between category 1 and category 4 is therefore important. An ordinary price fall in the market is our risk and does not touch the customer's fixed payment. An extraordinary circumstance such as an official order or a grid collapse is neither party's fault and is handled under the special terms, proportionally and temporarily for what is actually affected.

A note on market access

A related question is who ensures that the system can enter the markets at all. Here it is important not to promise too much.

We coordinate and work for market access through our chosen balance responsible party (BRP) or trading partner. But access is not something a single actor can unconditionally guarantee. Participation in balancing markets requires that the system passes a prequalification, and that a balance responsible party stands behind it, under the common EU rules for the electricity market. The prequalification verifies that the system meets the system operator's requirements before balancing services can be delivered. The framework is regulated: in Denmark, Forsyningstilsynet, the independent energy regulator, approves Energinet's methods for purchasing ancillary services and for market participants' access to the grids.

Market access therefore depends on several things at once: technical requirements, prequalification, site conditions, the chosen BRP or trading partner, the grid company, and the applicable regulation. If missing access is due to our circumstances, we must without undue delay seek to remedy it. If it is due to the customer's circumstances or circumstances beyond reasonable control, handling follows the categories described above.

ZynexGroup's approach to system assessment

Our approach is to make the risk allocation specific before an agreement is entered, rather than leaving it vague. In practice that means we review site conditions, grid connection, and insurance requirements together with the customer, and that we describe how the four categories of risk are allocated in the specific agreement.

We take ordinary market risk in the standard model because our business rests on our being able to optimise the system commercially. The customer owns the physical system under the agreed payment and transfer terms, while software logic, algorithms, trading strategies, and optimisation models are not transferred. We offer an actual 6-year payback guarantee on BESS systems, which presupposes that the customer participates in the market optimisation, and which applies within the terms the agreement describes. Metering, monitoring, and technical documentation are included so it is possible to determine what caused a given problem, and thereby who carries the technical risk.

Overall perspective

Risk in a BESS agreement is not one number, but an allocation. In our standard model we carry ordinary market risk, the customer carries the risk for their own site, technical risk is allocated by cause, and extraordinary circumstances are handled proportionally under the agreement's special terms.

The honest conclusion is that no model removes all risk. What a good agreement can do is place each type of risk where it belongs, and describe what happens in the cases where something departs from the normal. Ordinary market swings do not change the customer's fixed payment. Extraordinary circumstances can affect the agreement, but under specific, bounded rules, not as a general uncertainty.

Assessing the potential for your facility

If you are considering an industrial BESS system and want to understand how the risk allocation looks in your specific case, we can have a data-based dialogue about it.

We typically look at your grid connection, consumption profile, and site conditions and use that for an initial assessment of whether there is a match, and how the four categories of risk are allocated for your specific system. You are welcome to contact us.