Skip to content
Alpha Devraj — Energy Storage Systems

Costs & tariffs

What is an availability guarantee in a battery storage contract, and why do spares matter?

An availability guarantee is a contractual promise that the system will be ready to operate for a defined percentage of time, with payment deductions if it is not. Because capacity contracts pay for readiness rather than energy, availability is the product. Spare parts determine how fast you restore it after a failure.

Published 31 August 2026 · Last updated 31 August 2026 · 5 min read · By Alpha Devraj ESS Research Desk

On this page

On a solar project, the product is energy. On a battery project contracted through a capacity agreement, the product is readiness — you are paid a fixed monthly fee for the asset being there and able to work, as set out in tolling agreements and capacity contracts.

Which makes the availability guarantee the single most commercially consequential clause in the contract. And makes spare parts, unglamorously, a revenue issue.

What an availability guarantee actually says

The structure is straightforward: the system will be available for at least X per cent of a defined period, and if it is not, the capacity payment is reduced according to a defined formula.

The number is the part everyone negotiates. The definitions are the part that decides what the number means.

The number means nothing without these four answersWhat counts as unavailable?Fully offline only, or also degraded andpartially derated operation?Over what window?Annual averaging hides a long outage.Monthly measurement does not.What is excluded?Grid outages, force majeure, plannedmaintenance — and how much of each?What is the deduction?Pro-rata, stepped, or capped — and isthere a termination trigger?Compare definitions before you compare percentages.
Four definitional questions that determine what an availability percentage is actually worth. Two contracts quoting the same figure can be entirely different promises.

What counts as unavailable. Only a full outage, or also a system running at reduced capability? A battery that can deliver 60 per cent of contracted power is not offline, but it is not delivering the product either. Contracts differ sharply here.

The measurement window. This is the one most often underestimated. A guarantee measured annually lets a two-week outage disappear into an average. Measured monthly, the same outage produces a substantial deduction in that month. Same headline percentage, materially different risk allocation.

Exclusions. Grid unavailability, force majeure and planned maintenance are usually excluded, and reasonably so — but the allowance for planned maintenance is negotiable and worth negotiating.

The deduction formula. Pro-rata, stepped, capped, and whether persistent underperformance triggers termination rights.

Why spares decide the outcome

Here is the operational reality: cells are not usually what fails. Modern lithium iron phosphate cells in a well-cooled system are reliable. What fails is everything around them.

  • Power electronics. The power conversion system contains the most stressed components in the installation, switching at high frequency and high power, generating heat continuously.
  • Cooling equipment. Fans, pumps, compressors and controls — moving parts working hard in Indian ambient conditions, as covered in cooling a battery in Indian heat.
  • Communications and controls. The layer that makes the asset dispatchable, described in EMS versus BMS. A battery nobody can talk to is unavailable regardless of its state of charge.

None of those are the components buyers scrutinise during procurement. All of them determine whether you hit your availability guarantee.

The India-specific problem

India imports the great majority of its battery components, and analysts place genuine cell manufacturing self-sufficiency a decade or more away — the position set out in India’s battery import dependence.

For availability, that translates into a blunt question: where is the spare part right now?

A converter module in a warehouse two hours away is a one-day outage. The same module at a factory abroad, subject to shipping and customs, can be a six-week outage. On a contract with monthly availability measurement, those two scenarios have very different financial outcomes — from the same failure of the same component.

So the practical questions to settle before signing:

  • Where is the nearest spares holding, and what does it contain?
  • What is the guaranteed response time, and does it distinguish diagnosis from restoration? “Response within 24 hours” often means someone will look at it, not fix it.
  • Who holds critical spares — you or the supplier — and who pays for the stock?
  • What is the escalation path when the first visit does not resolve it?
  • What happens if the supplier exits the market? In a young industry with many new entrants, this is a genuine question, not a hypothetical.

Deciding whether to hold your own stock

The calculation is simple in principle: compare the cost of holding a critical spare against the cost of downtime while one is sourced.

On a contracted project with availability deductions, downtime has an explicit price written into the agreement, which makes the comparison easy — and it usually favours holding stock for the few components that are both failure-prone and long-lead.

For a behind-the-meter commercial system the arithmetic is different, because the cost of downtime is a lost bill saving rather than a contractual penalty. There, response time usually matters more than stock, unless the site cannot tolerate an outage — which is the case in hospitals and similar critical facilities.

Either way this belongs in the operations and maintenance budget from the outset, alongside the service contract, insurance and warranty premiums.

Availability is not the same as capacity

Worth stating plainly, because the two are frequently conflated.

Availability asks whether the system is ready to run. Capacity, governed by the degradation guarantee, asks how much energy it still holds.

A system can be perfectly available and well below its guaranteed capacity. Or at full capacity and offline. They are separate promises, usually with separate remedies, and a project needs both to be sound. The baseline for the capacity promise is normally set at site acceptance testing, which is another reason that measurement matters so much.

What this means for you

  • If you are a developer on a capacity contract: model availability deductions as a real risk line, and negotiate the measurement window as hard as the percentage. Annual versus monthly measurement can be worth more than a point of headline availability.
  • If you are procuring: ask about spares holding and response times before you ask about cell brand. It is the better predictor of whether the asset performs.
  • If you are a C&I buyer: you may not have a formal availability guarantee, but you should still ask who fixes it, how fast, and from where. A cheap system with no service presence in your region is not cheap.
  • If you want to review a service scope: our team can walk through availability terms and a spares strategy for a specific configuration, and our product range is supported from our Indian manufacturing operation — get in touch.

Contract terms, service structures and supplier capabilities vary widely and change over time, and nothing here is legal advice on any specific agreement. Treat this as an August 2026 snapshot and have your own advisers review the actual terms you are offered.

Frequently asked questions

Is 98% availability good?

It depends entirely on the definitions. Ninety-eight per cent measured annually with generous exclusions is a much weaker promise than ninety-five per cent measured monthly with tight ones. Always read the definition before comparing the number.

What usually causes unavailability in practice?

Power electronics and auxiliary systems more often than cells. Converters, cooling equipment, communications and controls are the components that fail and the ones that need spares held nearby.

Should I hold my own spares or rely on the supplier?

It depends on the cost of downtime. If a day offline costs meaningfully more than holding a critical spare, hold the spare. On a contracted project with availability deductions, that calculation usually favours holding stock.

How does this interact with the degradation warranty?

They are separate promises about different things. Availability is about whether the system is ready to run; degradation is about how much capacity it still holds. A system can be perfectly available and well below its guaranteed capacity, or full capacity and offline.

Let's talk storage

Want these numbers for your site?

Send a bill or a load profile — we model it and tell you straight.