Before You Buy Battery Storage: How to Read Cycle Life, Warranty and Performance Guarantees

A cycle-life figure of, say, 6,000 cycles can appear in three different offers and mean three different things. Unless the number is accompanied by the conditions it was measured under, it is not a performance claim at all — it is a number that happens to be large.

Storage buying goes wrong less often because the equipment was bad than because the comparison was made on terms that did not line up. Two offers can look alike on the summary sheet and be describing systems with materially different operating envelopes, different service commitments and different costs once everything that is not the battery is counted.

This article is about reading those terms: what a cycle-life figure has to be paired with before it means anything, the difference between a warranty and a performance guarantee, what the exclusions usually say, and what a quotation ought to show before it is worth comparing to another.

Rows of dark prismatic lithium battery cells with copper and aluminium terminals beside a completed battery module clamped by machined metal end plates with a wiring harness across its top, on a production line

One Number, Several Meanings

Cycle life is the number of charge and discharge cycles a battery is expected to complete before it reaches the end of its useful life. Every part of that sentence is defined by conditions that the number alone does not carry.

Discharge depth. A cycle is not a fixed quantity of energy. A battery cycled between full and empty is being worked much harder than one cycled through a narrow band, and the cycle counts are not comparable. A figure quoted without the depth of discharge it assumes cannot be checked against the way the site will actually run.

Temperature. Cycle life is measured at a temperature, and the value assumed is usually a moderate one. A site that spends its afternoons well above that will not see the quoted figure, and the reason is not a defect — it is that the number was never a statement about that site. The mechanism is the same one described in how temperature reduces usable storage capacity.

Charge and discharge rate. Moving energy in or out of a battery quickly is harder on it than moving the same energy slowly. A cycle-life figure quoted for a low rate does not transfer to an application that cycles at a high rate, which is common where storage is used for short, sharp load events rather than bulk energy shifting.

End-of-life criterion. “End of life” is defined as a percentage of original capacity — commonly some figure below which the battery is considered to have reached the end of its service life. A test that stops at a higher retained capacity will produce a larger cycle count than one that runs further, and the two are describing different points in the battery’s life.

Four conditions, one number. Where an offer gives the number and not the conditions, the correct question is not whether the figure is good — it is which set of conditions produced it.

Cycle Life Is a Result, Not a Property

A battery does not have a cycle life in the way a motor has a rated speed. It has a cycle life under a stated duty, and the duty is set by how the site operates.

This matters because the same system serving two different loads can deliver very different service lives. Storage used to shave a short daily peak, cycling a narrow band at high rate, and storage used to carry a site through a long overnight period, cycling deeply at a lower rate, are doing different things to the same cells. The system that was quoted on a favourable duty cycle may be the right hardware for the site and still not reach the number in the offer, because the site’s duty is not the reference duty.

The practical consequence for a buyer is that the operating profile has to be described to the supplier before a life figure is meaningful. A proposal that states a cycle life without having seen the load profile is stating a laboratory condition, not a project expectation. Where a supplier is willing to put a life figure against the site’s actual duty — including the temperature it will see and the depth it will cycle — that figure is a far more useful thing to compare than the headline number.

Warranty and Performance Guarantee

These two are frequently treated as one item in a discussion and they protect against different failures.

A warranty covers defects — the equipment failing because something was made wrong or failed prematurely under normal use. It is a statement about the equipment’s integrity.

A performance guarantee covers output — a commitment that the system will deliver a certain amount of energy, or retain a certain share of its capacity, over a period. It is a statement about the system’s function.

A project can hold a valid warranty and still be underperforming. If a battery retains less capacity than expected but has not failed, the warranty may not be engaged at all, because nothing is broken. Sites that only negotiated a warranty discover this in year four or five, when the storage no longer covers the load it was bought to cover and there is no contractual instrument that addresses it.

For a storage purchase, both belong in the discussion. They are usually separate clauses, sometimes in separate documents, and they sometimes have different durations — a warranty that runs for one period and a performance commitment measured over another. Reading only the first is the most common gap in a storage contract review.

Reading the Exclusions

The exclusions decide how much the rest of the document is worth. They are usually in a section that is shorter than the coverage clause and easier to skim. Five categories are worth reading line by line.

Operating envelope. Temperature range, charge and discharge rate limits, depth of discharge, and how long the system may sit unused. A site whose ambient crosses the stated temperature band, even for a few weeks a year, is outside the envelope for that period. This single clause can exclude the most common reason storage underperforms, and it is the reason site temperature data is worth having before the contract is signed rather than after.

Maintenance and operating conditions. Requirements about inspections, firmware and control settings, ventilation, and how the system is dispatched. Failure to keep to them is a standard ground for declining a claim, and it is the ground most likely to be triggered in normal operation — by a site that changed a setting, or missed a service visit, without connecting the two.

Balance of system. Whether the guarantee covers only the battery or also the conversion equipment, the controls and the integration. Where coverage stops at the battery, a failure in the inverter or the controller is a different conversation with a different party, and the site is carrying the interface risk.

Grid and site conditions. Some clauses require the supply to stay within stated quality limits — a condition the site may not control. On a network that sags or deviates, this can be the clause that decides whether a claim is honoured, and it connects the purchase decision to the electrical environment the equipment will actually see, in the way set out in what equipment has to tolerate on a weak grid.

Parts, labour and transport. Whether a replacement is covered as a part, as a part plus labour, or as a part plus labour plus shipment to and from the site. For a remote installation the last of these can be the largest number in the whole claim, and it is routinely not the one that gets discussed.

What the Quote Should Show

Storage quotations are not standardised, which means the same scope can be presented in ways that are not comparable. Before comparing two offers, establish that both cover the same list.

Equipment. The storage system itself, and separately the conversion equipment, the controls and the protection, since these are not always included in what is called “the battery system”. What arrives as one integrated unit and what the site has to assemble is also a function of the format the system takes — the consequence of choosing between containerized, cabinet and all-in-one storage.

Delivery and placement. Transport to site, offloading, lifting equipment, and any route preparation. Where the site is remote or access is restricted, this is a material cost that a per-unit price does not show.

Civil and electrical works. Foundations or pads, cable routes, cabling, switchgear, protection settings, and the grid connection works. These are frequently excluded from equipment quotations and arrive later as variations.

Integration and commissioning. The control and energy management configuration, the point at which the system is brought into service, and the site acceptance testing. Whether this is in the price or not changes what the site has to organise itself.

Documentation, spare parts and training. As-built drawings, test records, the spares recommended for the first years of operation, and who is trained to operate the system after handover.

Whether those items sit inside or outside the price depends on how the system is supplied. A battery energy storage system bought as a standalone product carries a different integration scope from one delivered with its control and energy management layer already integrated, where the energy management work belongs to the supplied system rather than to the site.

A price per kilowatt-hour of storage is only a comparison if everything after the first line is equal. Where it is not, the comparison is between one supplier’s equipment price and another supplier’s installed price, which is not a comparison at all. The sizing logic itself is a separate question, covered in how kW and kWh each decide something different in a commercial storage system.

An environmental test chamber with its door open, a battery module resting on a shelf inside and heavy copper busbars running from the module through the chamber wall to terminals on a test rig

Standards, Certificates and Test Reports

Certification for a storage system is not a single item. Requirements are usually applied at several levels — the cells, the modules or packs, the complete system, and the transport of the battery — and a market’s rules may address each level separately.

What a buyer needs is not the list of standards a supplier’s marketing material mentions, but the specific documents, issued by the testing body, covering the configuration actually being supplied. A certificate issued for a different cell or a different system architecture than the one quoted does not describe the equipment that will arrive.

The practical questions are narrow and easy to answer: which documents exist for this configuration, who issued them, what do they cover, and are they available before the contract is signed rather than with the delivery. A supplier who can produce them promptly is answering a question about their own engineering process as much as about the product.

Where the system will be connected to a network, there may be additional requirements set by the operator rather than by the equipment standards, and those are a separate documentary exercise with its own timeline. That sequence is covered in what grid connection approval involves for a storage project.

The Questions to Put to a Supplier

Everything above reduces to a short list that can be sent to any supplier before an offer is evaluated. A complete answer to these is a better indicator than any single specification line.

On the life figure. Under what depth of discharge, temperature and charge rate is the cycle life quoted? What end-of-life capacity is assumed? What does the figure become under this site’s actual duty?

On the guarantee. Is there a performance commitment separate from the warranty? Over what period, measured how, and against what baseline? What happens if the system falls short?

On the exclusions. What temperature range and operating limits must the site stay inside? Which conditions void coverage? Is the conversion equipment and the control system covered, or only the battery?

On the scope. What does the price include and exclude, specifically on transport, lifting, foundations, cabling, protection, integration, commissioning and training?

On the documents. Which certificates and test reports exist for this exact configuration, and can they be provided before signature?

On the service. Who attends the site when something fails, how far away are they, and what is the expected response time? What spares are recommended on site?

The pattern of the answers is as informative as their content. A supplier who has a defined position on cycle-life conditions, who separates warranty from performance and who can produce the configuration’s documents is describing a system they know well. A supplier whose answers are all variations on “standard” is selling a product whose operating envelope they have not had to specify — and the site will be the first place that envelope is tested.

FAQ

Q: The supplier quotes a cycle life. How do we know whether it is good?

A: Not from the number alone. Ask for the depth of discharge, temperature and charge rate it assumes, and the end-of-life capacity definition, then ask what the figure becomes under your site’s actual duty and climate. A figure quoted for a moderate reference condition and applied to a hot, deeply cycled site is not a prediction, and the difference is usually outside the supplier’s control unless they were given the site’s operating profile.

Q: What is the difference between a warranty and a performance guarantee?

A: A warranty covers the equipment failing — defects and premature failure. A performance guarantee covers the equipment producing what it was bought for — retained capacity or delivered energy over a period. A system can be fully within its warranty and still not be delivering what the project needs. Ask for both, and ask how each is measured.

Q: Why does the warranty specify a temperature range?

A: Because capacity loss and failure modes are temperature-dependent, and the manufacturer is stating the conditions under which the product’s behaviour is characterised. A site that exceeds that range, even seasonally, is outside the manufacturer’s stated conditions for part of the year. Site temperature data collected before the contract is signed is what settles this question; the same physical mechanism is covered in why temperature affects usable capacity and cycle life.

Q: Can we compare suppliers on price per kWh of storage?

A: Only if everything else in the scope is identical, which it usually is not. Transport, lifting, foundations, cabling, protection, integration, commissioning, spares and training are commonly included by one supplier and excluded by another. Compare the delivered and commissioned scope, then compare prices within it.

Q: The quotation does not mention certification. Should we ask?

A: Yes, and ask specifically which documents exist for the configuration being supplied, who issued them and whether they cover the cells, the modules and the complete system. Requirements are usually applied at each of those levels separately, and a certificate for a different cell or architecture does not describe the equipment that will be delivered.

Q: What happens if the system underperforms in year five?

A: That depends entirely on what was agreed, which is why the performance commitment has to be settled before signature. If the contract contains only a warranty against defects, a system that has lost capacity but has not failed may fall entirely outside it. The remedy, the measurement method and the baseline should all be defined in writing.

Q: Does the choice of battery chemistry change any of this?

A: The questions do not change; the answers do. Different chemistries have different temperature behaviour, different cycle-life characteristics and different safety provisions, so the reference conditions and the exclusions that accompany a life figure differ between them. The comparison between the two chemistries used most often in off-grid and hybrid storage is set out in how LFP and lead-acid differ for off-grid solar.

The interior of an open storage cabinet showing stacked battery modules on steel racks with blue sense wiring, orange DC cables, and a copper busbar carrying cylindrical fuses to one side

Storage is bought on a specification sheet and paid for over ten years of operation. The mismatch between those two timescales is where most disappointment in a storage project originates — not in the equipment, but in the gap between what the numbers were measured under and what the site will actually do to them. Closing that gap is a matter of asking the right questions before the order, which costs nothing, rather than afterwards, which does not.

YURI POWER Engineering Team · October 7, 2026