Battery Pack Cost per kWh: Why Two Quotes Aren’t Comparable

Two quotes can show the same price per kWh and differ by 40% in what they actually deliver. That is not a negotiation tactic. It is arithmetic. The number on the quote is a price divided by a kWh figure, and both halves of that fraction are defined differently by different suppliers: which kWh, measured how, delivered to where, guaranteed for how long.

This page sets out the six terms that decide whether two quotes are comparable, gives a worked example on a single 51.2 V 100 Ah pack where the same price produces four different cost-per-kWh figures, and closes with the formula that survives all of them. It is written for buyers who already have quotes in hand – distributors, installers, integrators and fleet workshops – not for someone still deciding whether to use lithium at all. For that earlier question, see whether a conversion pays back.

Short answer: a price per kWh is only meaningful when the quote states four things – whether the kWh is nominal or usable, where the quote boundary sits (cell, module or finished pack), the cycle-life end point the warranty is written against, and the delivery term. Without those four, the number cannot be compared with anything, including another quote from the same supplier.

Key takeaways

  • A price per kWh is a fraction, and both halves vary. Nominal and usable capacity on the same pack can differ by 25% before any cycle-life question is asked.
  • Cycle counts are not comparable across suppliers until the end point is known. 3,000 cycles to 70% capacity and 2,000 cycles to 80% are different products, and the lower headline number is not necessarily the worse deal.
  • Within one manufacturer catalogue the spread is already 2.5x. Gotion publishes 1,200 to 3,000 cycles across its LiFePO4 cells, because the test conditions and end points differ by model.
  • The step from cell to finished pack is the largest single increment in a quote, and it is the step most often left undefined.
  • Use delivered cycle-kWh, not price per nominal kWh. It is the only figure that normalises capacity, depth of discharge, cycle end point and efficiency at the same time.

What a price per kWh actually measures

A price per kWh is a price divided by a capacity. Neither term has a single agreed definition in the battery trade, which is why the ratio can be moved without anyone lying. Four definitions decide most of the variance:

TermWhat variesEffect on the headline figure
Capacity basisNominal rated capacity vs usable capacity at a stated depth of dischargeUp to 25% on a 80% DOD pack
Quote boundaryBare cell, module, finished pack, or pack plus charger and rackUsually the largest single swing
Cycle basisCycles to 80% capacity vs cycles to 70%, under a stated C-rate and temperatureMultiplies the value per unit price
Delivery basisEXW, FOB, CIF or DDP, with dangerous goods handling included or notA one-line difference that is not one line

A supplier can therefore quote a lower price per kWh than a competitor and still sell the more expensive product. This is not usually deliberate. It is what happens when a commercial team quotes against the number the market talks about, and the technical team specifies against the number the datasheet provides.

The six terms that make two quotes incomparable

1. Nominal capacity or usable capacity

A 51.2 V 100 Ah LiFePO4 pack is 5.12 kWh nominal. That figure assumes the pack is discharged from 100% to 0%, which no correctly configured system does. At 80% depth of discharge the usable capacity is 4.10 kWh. A quote written against 5.12 kWh describes a pack that does not exist in service.

The gap widens once service life is included. A pack rated to retain 80% of its original capacity at end of life averages roughly 90% across its life. Multiply that in and the long-run usable capacity of the same 5.12 kWh pack falls to about 3.69 kWh. The following section works the full arithmetic.

Ask which basis the quote uses. A supplier quoting on nominal capacity is not wrong to do so – it is the standard commercial convention for cells – but the buyer has to convert before comparing. For the sizing side of the same question, see how to size a battery pack.

2. The cycle-life end point: 70% or 80%

A cycle count is meaningless without the capacity at which it terminates. “3,000 cycles” can mean 3,000 cycles until the pack falls to 80% of its original capacity, or 3,000 cycles until it falls to 70%. Those are not the same product, and the second one has given away 10 percentage points of capacity for free in the headline number.

The convention is not consistent even inside one manufacturer. In the Gotion LiFePO4 catalogue we publish, the cycle figures are written against different end points and different test conditions:

ModelFormatPublished cycle figureEnd point and test note
IFR37146-20AhCylindrical1,200To 80% capacity at 25 C; discharge rate requires confirmation
IFR32135-15AhCylindrical2,000Section 5.4.1; 25 C, 0.5C charge / 1C discharge
IFP42100140A-67AhPrismatic2,500To 80% capacity; 25 C, 100% DOD, 150 kgf preload
IFP50160116A-102AhPrismatic3,000To 70% capacity; 25 C, 50 A charge and discharge, 150 kgf preload

Four LiFePO4 cells from one brand, one catalogue, and a 2.5x spread in the cycle number. The 3,000-cycle model also has the lowest end point of the four, so its advantage in delivered energy is smaller than the raw ratio suggests. The full model-by-model breakdown is on the Gotion cell page.

The same caution applies across brands. Farasis publishes 3,000 cycles for its P45 LFP pouch cell and 2,000 for the P148, both LFP, because the models are built and tested differently – not because one chemistry is better. See the Farasis range for the comparison by model rather than by prefix.

3. Test conditions: C-rate, temperature and mechanical preload

Cycle life is not a property of a cell. It is a property of a cell under a test condition. Change the discharge rate, the ambient temperature, or the mechanical clamping force on a prismatic cell and the published number moves.

Three variables account for most of the difference:

  • Discharge rate. A cell cycled at 0.5C will generally outlast the same cell cycled at 1C. A pack used in stationary storage at 0.2C is not living the life the datasheet describes if that datasheet was written at 1C.
  • Temperature. Most published figures are at 25 C. Capacity fade accelerates above 35 C and charging below 0 C is not permitted for LFP at all. A pack in a Gulf or Southeast Asian installation does not operate at 25 C.
  • Mechanical preload. Prismatic cells are specified with a clamping force – Gotion publishes 150 kgf for some models, 100 kgf for others. Assembled without it, the cycle figure is not the one on the datasheet.

For a pack builder this is the argument for stating the duty cycle before the price is calculated. A quote built on a datasheet number that assumes 25 C and 0.5C is optimistic for a vehicle and pessimistic for a solar installation. For the chemistry-level comparison instead, see LFP vs NMC vs sodium-ion.

4. Where the quote boundary sits: cell, module or pack

This is the largest single increment in most quotes and the one most often left undefined. The same cells appear at three or four very different prices depending on how far down the assembly line the supplier is quoting.

LayerWhat the price coversWho buys it
CellBare cell, no busbar, no sensing, no enclosurePack builders and integrators who assemble in-house
ModuleCells plus frame, busbars and cell-sensing harnessBuilders who have BMS and enclosure capability
Finished packModule content plus BMS, enclosure, terminals, wiring, assembly and end-of-line testDistributors, installers and fleet buyers
Pack systemFinished pack plus charger, rack or cabinet, cabling and commissioning supportTurnkey projects and first-time lithium buyers

A quote that does not state its layer cannot be compared, because the missing items are exactly the ones a buyer would otherwise have to source, integrate and take responsibility for. The BMS alone changes the safety case rather than only the price – see BMS specifications explained for what a programmable BMS adds and what a fixed one cannot do.

Ask for the bill of materials behind any pack-level quote. If a supplier cannot produce one, the comparison has already failed, whatever the number says.

5. Trade terms: from EXW to DDP on dangerous goods

Lithium cells and packs move as dangerous goods. UN38.3 test summaries, MSDS, the export packing certificate and the carrier declaration are not optional paperwork – they are conditions of carriage, and someone has to pay for them. Which terms the quote is written on decides who.

Cost lineEXWFOBCIFDDP
Export packing to dangerous goods standardBuyerSellerSellerSeller
Inland transport to the port of loadingBuyerSellerSellerSeller
Ocean freight and dangerous goods surchargeBuyerBuyerSellerSeller
Marine insuranceBuyerBuyerSellerSeller
Destination clearance and dutyBuyerBuyerBuyerSeller
Import VAT or equivalentBuyerBuyerBuyerUsually seller, recovered separately

The useful rule is to convert every quote to the same term before comparing, and to convert to DDP if you actually want the goods in your warehouse. A CIF quote looks cheaper than a DDP quote until the destination charges arrive. Our own export documentation position is set out in the export document checklist.

6. Warranty, spares and who pays the return freight

Two packs at the same price are not the same product if one carries a 12-month warranty and the other carries five years. Warranty is priced into the quote whether or not the quote itemises it.

Four questions separate a real warranty from a marketing one:

  • What is warranted – capacity, function, or both? A warranty that covers the pack turning on is not a capacity warranty.
  • What is the capacity threshold, and measured how? “80% at 3 years” means nothing without the test method and the duty cycle it assumes.
  • Who pays freight on a claim? Returning a 40 kg pack from Europe to China can exceed the value of the replacement. A warranty that does not address freight is a warranty the buyer rarely uses.
  • Is a spare ratio included? Fleet buyers commonly carry 1-3% spares. A quote that excludes spares is not cheaper – it is incomplete.

Our published position is deliberately narrow: custom DIDIDEDE packs carry 12 months or 1,200 cycles, whichever comes first, unless otherwise agreed in writing. It is a shorter warranty than some competitors advertise, and it is the one we can honour on a multi-brand cell supply. See the turnkey pack process for how warranty terms are fixed at quotation stage.

One pack, four ways to divide it

Take a single 51.2 V 100 Ah LiFePO4 pack. Hold the purchase price constant and change only the denominator. The same invoice produces four different cost-per-kWh figures:

DenominatorCalculationkWhCost index
Nominal capacity51.2 V x 100 Ah5.12100
Usable at 80% depth of discharge5.12 x 0.804.10125
Usable, averaged over life to an 80% end point5.12 x 0.80 x 0.903.69139
Delivered energy at 95% round-trip efficiency5.12 x 0.80 x 0.90 x 0.953.50146

The index column is the same price divided by a smaller number. A pack quoted at a nominal cost index of 100 is carrying a delivered cost index of 146. Nothing about the pack changed. Only the definition of kWh did.

This is why the question “what is your price per kWh” produces answers that cannot be compared, and why suppliers who answer it precisely sound more expensive than suppliers who answer it loosely. The fix is not to demand a better number. It is to demand the definition.

The formula to use instead: cost per delivered cycle-kWh

Capacity normalisation handles the denominator. It does not handle service life. For that, divide the price by the energy the pack will actually deliver over its life:

Cost per delivered cycle-kWh = PACK price / (nominal kWh x DOD x rated cycles x round-trip efficiency x average state of health)

Average state of health is the midpoint between the pack as new and the end point the cycle figure is written against: 0.90 for an 80% end point, 0.85 for a 70% end point. It is an approximation, and it is a far better one than ignoring the end point entirely.

Applied to two packs with identical nominal capacity, identical DOD and identical price:

Pack APack B
Nominal capacity5.12 kWh5.12 kWh
Depth of discharge80%80%
Rated cycles2,0003,000
Cycle end point80%70%
Average state of health0.900.85
Round-trip efficiency95%95%
Delivered cycle-kWh7,0049,925
Relative cost per delivered cycle-kWh10071

Pack B delivers 42% more energy over its life at the same purchase price, which puts its real cost per delivered cycle-kWh 29% below Pack A. A buyer comparing only the headline cycle count would see 50% more cycles; a buyer comparing only price per nominal kWh would see no difference at all. Only this formula shows the actual position.

The formula also makes a lead-acid comparison honest. Lead-acid traction batteries are typically cycled at 50% depth of discharge and deliver several hundred cycles, with round-trip efficiency well below lithium. Running both technologies through the same five terms is the only way to compare them; the payback arithmetic is worked through in lead-acid to lithium conversion payback.

Three more terms buyers forget

Payment terms and the currency date

A 30% deposit against a 100% prepayment is a financing cost, and it belongs in the comparison. At commercial lending rates, the difference between paying in full on order and paying 70% on shipment is not trivial over a production lead time measured in weeks.

Currency matters for the same reason. A quote in one currency against a cost base in another carries exchange risk between quotation and shipment. Ask which date the rate is fixed to, and what happens if it moves before shipment. A quote with no exchange clause is a quote that will be revisited.

MOQ ladder and tooling

Cell and pack pricing is volume-banded. A quote for a sample quantity and a quote for a container quantity are different products in commercial terms even when the specification is identical. Comparing a distributor price against a project price without stating the volume band is a common source of apparent bargains.

Where a custom enclosure or a private-label BMS configuration is involved, there is also tooling or engineering cost. It is usually a one-off, it is usually quoted separately, and it is usually omitted from the price per kWh that the buyer later uses for the comparison. Our published minimum order quantity is three units, which is deliberately low, but the volume bands above that still apply.

Certification and test documentation

UN38.3 test summaries, MSDS and transport documentation are routinely supplied. Product certification for the destination market is a different question – it depends on the application and the country, and it is not transferable between them.

Two quotes can differ because one includes the test documentation the buyer will need and the other expects the buyer to commission it. Ask what documentation ships with the goods, and get the answer in writing before the price is agreed.

What to ask before comparing two quotes

Send the same twelve questions to every supplier and compare the answers, not the totals. A supplier who cannot answer them is not ready to be compared.

#QuestionWhy it changes the comparison
1Is the capacity figure nominal or usable, and at what depth of discharge?Sets the denominator
2What is the quote boundary – cell, module, pack or system?Usually the largest single swing
3What is the cycle-life end point, and at what C-rate and temperature?Sets the multiplier
4Is the BMS included, and what are its current and balancing limits?Changes both price and safety case
5Which delivery term is quoted, and are dangerous goods charges included?Moves cost between parties
6What documentation ships – UN38.3, MSDS, packing certificate?Avoids a later unbudgeted cost
7What is the warranty period, and what does it cover – function or capacity?Priced into the quote
8Who pays freight on a warranty claim?Determines whether the warranty is usable
9Is a spare ratio included in the quantity?Fleet total cost of ownership
10What is the cell brand, model and grade?The single largest determinant of real cycle life
11What is the payment schedule and the currency fixing date?Financing and exchange cost
12Which volume band does the price apply to?Prevents comparing different commercial products

Why the cheapest quote usually fails

The lowest price per kWh in a tender usually belongs to the quote that defined the fewest terms. It is not necessarily a bad supplier – it is often a supplier who answered the question as asked, with the convention that flatters the number.

Four failure modes account for most of the difference, and all four are visible before the order:

  • Cell grade. Cells sold at a discount are frequently B-grade, mismatched or from a batch with wider capacity spread. A pack built from mismatched cells fails at the weakest cell, not at the average.
  • Low cycle end point. A headline cycle count written to 70% capacity can be higher than a competitor number written to 80% while delivering less energy per unit price.
  • Underspecified BMS. A fixed-threshold BMS is cheaper than a programmable one and cannot be adapted to a different charger or duty cycle later.
  • No end-of-line test. Packs shipped without a full charge-discharge test at the factory transfer the cost of finding faults to the buyer, in the destination country, after the freight has been paid.

The practical defence is the same in all four cases: ask for the specification, in writing, against the twelve questions above. A supplier whose product genuinely is cheaper will be able to answer them. A supplier whose price depends on the answers not being asked will not.

Frequently asked questions

What is a normal price per kWh for a LiFePO4 battery pack?

There is no single normal figure, and any supplier who gives one without asking what is inside the pack is describing a convention rather than a product. The published number depends on the capacity basis, the quote boundary, the cell brand and grade, the cycle end point, the delivery term and the volume band. A price per nominal kWh for bare cells and a price per nominal kWh for a finished pack with a programmable BMS are not measurements of the same thing, and the gap between them is larger than the gap between most competing suppliers.

Is the lower price per kWh always the better deal?

No. The same purchase price produces cost-per-kWh figures that differ by up to 46% depending only on which kWh is used as the denominator, and by a further factor once cycle end point and depth of discharge are included. A quote that looks 10% cheaper per nominal kWh can be materially more expensive per delivered cycle-kWh. Convert both quotes to delivered cycle-kWh before deciding.

Should I compare cell prices or pack prices?

Compare at the layer you are actually buying. If you assemble packs in-house, cell price is the relevant figure and you carry the integration risk. If you buy finished packs for resale or installation, cell price is not a purchasable option and comparing against it understates your real cost. The step from cell to finished pack is usually the largest single increment in the quote, and it is the step most often left undefined.

Why do two suppliers quote far apart on the same specification?

Because the specification usually is not the same. The six terms that move most are nominal versus usable capacity, the cycle-life end point, the test conditions, the quote boundary, the delivery term and the warranty scope. Two quotes can agree on voltage, capacity and chemistry and still describe different products. Send both suppliers the twelve-question list and compare the answers.

Does cycle life affect the price per kWh?

Yes, through two routes. First, cells with a higher verified cycle life cost more, so part of the price difference is real. Second, the cycle figure itself can be written against a lower end point or a friendlier test condition, which raises the headline number without changing the product. A 3,000-cycle claim to 70% capacity and a 2,000-cycle claim to 80% capacity are not the ranking they appear to be.

How do I compare a lead-acid option with a lithium option?

Run both through the same five terms: nominal capacity, depth of discharge, cycle count at a stated end point, round-trip efficiency and delivery term. Lead-acid traction batteries are typically cycled at 50% depth of discharge with several hundred cycles and lower round-trip efficiency, so the invoice price comparison understates the lithium case, sometimes severely. The payback arithmetic in working years is set out separately.

Getting a quote you can actually compare

We quote on a stated basis or not at all. Every DIDIDEDE pack quotation states the capacity basis, the quote boundary, the cell brand and model, the cycle reference and its end point, the delivery term and the warranty scope – because without those, the number is not usable by the person receiving it.

If you have two or more quotes from other suppliers and want them normalised onto a single basis before you decide, send them over with the twelve questions answered as far as you have them. We will tell you where the comparison is invalid, including where the answer favours a competitor. Start from the pack manufacturing page or the wholesale cell range.

Newsletter Updates

Enter your email address below and subscribe to our newsletter