EV Battery Replacement Cost 2026: 1,846 UK Quotes, a £7,235 Mean and the £1,486 Repair Most Drivers Are Never Offered

The 2026 headline findings#

The mean UK out-of-warranty electric vehicle battery replacement quote was £7,235 fitted in 2026 and the median was £6,480. That is £127 per kWh of usable capacity, down from £412 per kWh in 2015.

One convention governs every pound figure on this page and it is stated here once, prominently, because ignoring it is the single largest source of error in this category. Every price is a fitted price: parts, labour, diagnostics, coolant, consumables, post-fit calibration and VAT, quoted to a retail customer with no warranty or goodwill contribution applied. A list price for a bare pack and an invoiced fitted price are not the same number and are routinely compared as though they were. Where a column reports something else (a parts-only component cost, a labour element, a cost per kWh), the column says so.

A mean on its own repeats the mistake the rest of the category makes, so the distribution is published with it. EV Cable Hub's 2026 replacement cost study put the lower quartile at £4,840 and the upper quartile at £9,120, which is the band half of all quotes fell inside. The extremes are wider and both are real: £2,140 for a refurbished 22kWh Renault Zoe pack at the bottom, £24,860 for a new Porsche Taycan pack at a main dealer at the top. A study that shows the top of the range as clearly as the bottom is the one worth quoting.

The finding that reframes the subject sits in the first 200 words on purpose. Most of the vehicles quoted for did not need a new pack. In EV Cable Hub's 2026 dataset, 61.4% of faults were resolved at software, cell, module, contactor or battery management system level rather than by replacing the pack, and a full pack replacement accounted for 38.6% of quotes. The headline £7,235 is a real number, but it prices the worst outcome rather than the typical one.

Four things move the number more than anything else, and they get a section each: how much energy the pack holds, what chemistry the cells are, who does the work, and whether the pack fitted is new, remanufactured, refurbished or salvaged. Between the cheapest and the most expensive combination of those four the spread runs from four figures to five, which is why a single national range is useless to anyone trying to make a decision.

One further convention is worth stating because it decides how the whole dataset reads. These are out-of-warranty prices. A quote obtained as a retail customer with no manufacturer contribution is the worst-case price for that job, and 78.6% of the pack replacements that actually happened in EV Cable Hub's 2026 owner survey were paid for by a warranty rather than by the owner. The figures on this page describe what the work costs, not what the typical owner of a vehicle with a battery fault pays.

Replacement cost by vehicle#

Battery replacement quotes ran from £2,140 for a Renault Zoe 22kWh on a refurbished pack to £24,860 for a Porsche Taycan on a new pack at a main dealer in 2026. Across the 74 vehicle variants priced, the median new pack fitted at an independent specialist was £8,140 and the median main dealer price £11,290.

This is the reference table the page is built around and the one most likely to be copied wholesale, so the four price columns are worth reading properly. New at a main dealer is a factory pack fitted by a franchised workshop. New at an independent is the same factory pack, or its supply equivalent, fitted by an independent electric vehicle specialist. Refurbished is a pack with failed modules replaced from stock. Typical module repair is what the same vehicle costs when the fault is resolved at module level instead, which for most rows is between a seventh and a tenth of the main dealer price.

The two medians above are variant medians, one row per model variant, and they sit above the £7,235 quote-weighted mean in the headline findings for a straightforward reason: the table gives a 22kWh Zoe the same weight as a 107kWh Volvo EX90, while the quote dataset is weighted towards the vehicles people actually bring in. Both numbers are correct and they answer different questions. Use the table row for a specific car and the dataset mean for a statement about the market.

The commercial point for a used buyer is that the cheapest column is not automatically the right one. EV Cable Hub's 2026 study found refurbished packs carry a mean 12-month warranty against three years on a new main dealer pack, and availability is the constraint that decides the question as often as price does: a refurbished pack existed for 61.8% of the models priced and a salvage pack for 44.1%. The availability and lead time figures later on this page belong beside this table, not after it.

Two columns are calculated rather than quoted and both are stated so they can be checked. Cost per kWh is the independent new price divided by usable capacity, which is why it is the column that compares across vehicles. Pack as a share of residual is the independent new price against the same vehicle's mean six-year residual value in EV Cable Hub's 2026 used market panel. On a 24kWh Nissan Leaf that ratio is 125.5%, and at main dealer prices (£5,840 against a £3,140 residual) it reaches 186%.

Cost per kWh, the number that actually compares#

UK battery replacement cost £127 per kWh of usable capacity as a 2026 mean and £119 at the median. Per-kWh cost fell as capacity rose, from £158 per kWh on packs under 30 kWh to £108 on packs above 90 kWh.

Cost per kWh is the only figure that compares across vehicles, and the headline pound figure is misleading without it. A Porsche Taycan pack costs roughly three times a 40kWh Nissan Leaf pack in part because it holds more than twice the energy. Dividing by usable capacity removes the size of the car from the comparison and leaves what is actually being paid for the energy storage, which is the number a fleet buyer, an insurer or a residual value model needs.

The relationship with capacity runs the other way to intuition: bigger packs are cheaper per kWh, not dearer. EV Cable Hub's 2026 study separates the bill into a fixed and a variable element to show why. The fixed element covers diagnostics, coolant, seals, calibration and the labour of getting the pack out and back in, and it barely moves with capacity, running from £1,240 on the smallest packs to £1,480 on the largest. The variable element, which is essentially cells, runs from £2,470 to £9,660 across the same range. Spread the near-constant fixed cost over four times the energy and the per-kWh figure falls even before the cells get any cheaper. The second reason is generational: larger packs are newer designs on cheaper cells.

Brand premium is real and it is large. At the manufacturer group level EV Cable Hub's 2026 study recorded £96 per kWh at the cheapest group and £184 at the most expensive, an index of 1.92: the same energy in a pack costing 92% more to replace. Almost all of that spread is one performance group; strip it out and the other fifteen sit between £96 and £131 per kWh, a far tighter band than the headline suggests. Labour hours track the same order, from 6.8 hours at the quickest group to 11.4 at the slowest, so the premium is partly parts pricing and partly how long the pack takes to remove.

One caution on reading the capacity table across to the headline. The band figures are means of the quotes inside each band, and the mix of chemistries, brands and workshop types differs between bands, so the gradient from £158 to £108 describes the shape of the relationship rather than an exact multiplier to apply to a specific car. For a specific car, the per-kWh column in the vehicle table is the figure to use.

There is a second reason to lead with cost per kWh rather than with total cost, and it is about how this data ages. Total replacement cost moves with the size of packs being fitted, which has risen every year. Cost per kWh strips that out, which is why the eleven-year series in the next section is quoted in per-kWh terms by everyone who has to model it, and why our own projection to 2030 is stated the same way.

Cost by pack chemistry#

LFP packs cost £108 per kWh to replace in 2026 against £131 for NMC 811 and £138 for NCA, the most expensive of the volume chemistries. Sodium-ion, on a sample of 18 quotes, priced lowest at £96 per kWh.

The chemistries differ in what they cost to make and in how easily they come apart, and both show up in the bill. LFP uses iron and phosphate in place of cobalt and nickel, which is most of why its cell cost runs at £61 per kWh against £84 for NCA in EV Cable Hub's 2026 teardown programme. Nickel-rich chemistries buy energy density with expensive metals, and the packs built around them tend to sit in vehicles where the labour hours are longer as well.

The second-order point matters more to a used buyer than the headline. LFP is cheaper to replace and, on the incidence data later on this page, gets replaced far less often: 0.4% of LFP vehicles in the 2026 sample against 0.9% for NMC and 7.8% for the early LMO blends. It is also the most repairable, with 74.1% of LFP faults resolvable at module level against 54.2% for NCA. The chemistry choice compounds in the owner's favour on both sides of the ledger, which is not how it is usually described.

Pack architecture cuts the other way and the honest version is worth publishing. Cell-to-pack designs are cheaper to build and price lowest per kWh at £112, but where a module-level repair is possible at all it costs £2,640 against £1,340 on a conventional modular pack. Structural cell-to-body packs, 4.3% of the models priced, have no module-level repair route in EV Cable Hub's 2026 dataset and take 14.6 labour hours to replace. That is why the module repair column in the vehicle table is thin for several recent models: the pack got cheaper and the repair got harder at the same time.

Warranty on the replacement follows chemistry too. A replacement LFP or sodium-ion pack carried a typical three-year warranty in the 2026 study against two years on the nickel chemistries and one year on the early LMO blends, which is the supply chain pricing its own confidence in the cells.

Cost per kWh over time#

UK battery replacement cost fell from £412 per kWh in 2015 to £127 in 2026, a drop of 69.2% and a mean annual fall of 10.1%. The fall was 11.2% in 2026, against the steepest years in the series at 12.8% in 2019 and again in 2025.

The series is the clearest single image in this dataset: a falling line from £412 to £127 across eleven years, with no year in which the figure rose. EV Cable Hub's 2026 study attributes the fall to four things in order of contribution. Cell cost did just over half of it, falling from £248 to £74 per kWh. Pack assembly and materials did a fifth, from £96 to £31 per kWh. Competition in the independent repair sector did an eighth, as the independent share of quotes went from 8.1% to 61.4%. A refurbished and salvage supply chain that barely existed in 2015 did a further tenth, from 2.1% to 28.6% of quotes.

Diagnostic and software access is the smallest of the five contributors at 4.8% and the one most likely to matter next. The number of usable diagnostic tools went from 12 to 84 across the period, which is what let the independent sector quote on work it previously had to turn away. Access, rather than skill, is the constraint that has been loosening, and the repair section later on this page shows what it has and has not yet changed.

The honest complication is the most quotable thing in the section. Replacement cost has fallen far more slowly than cell cost, because labour has not fallen at all. It has risen, from £486 to £684 on a mean job, up 40.7% over the same eleven years. Its contribution to the fall is therefore nil. As a share of the total bill labour has more than doubled, from 4.3% in 2015 to 9.5% in 2026, and it climbs every single year in the series without exception. The bill is quietly changing composition from a parts bill into a parts and labour bill, and the 2030 projection later on this page carries that trend to 14.1%.

Mean full pack cost has fallen far less than cost per kWh, from £11,240 to £7,235, a drop of 35.6% against 69.2%. That is not a contradiction: packs got much bigger over the same period, so buyers are paying less per unit of energy for considerably more energy. Anyone quoting the 69.2% fall as though it were the fall in what a replacement costs is quoting the wrong figure, and both are published here so the right one can be picked.

New, refurbished, remanufactured and salvage#

A refurbished pack cost a mean of £4,180 fitted in 2026, 42.2% below the £7,235 dataset mean and 53.2% below the £8,940 main dealer price. A tested salvage pack cost £2,640, and a remanufactured pack £5,120.

The four words are used interchangeably in the market and that is exactly where buyers get caught, so the definitions used throughout this study are set out plainly. New means a factory pack. Remanufactured means a pack rebuilt to a specification with new cells in some modules. Refurbished means a pack with its failed modules replaced from stock. Salvage means a pack removed from a written-off vehicle, either tested or not. Two discount figures circulate for each because there are two sensible baselines: the £7,235 dataset mean and the £8,940 main dealer price. Both are published above and both are stated with their baseline.

Availability, warranty and supplied state of health decide the purchase more often than the discount does. A new main dealer pack was available for 94.1% of models with a three-year warranty and a pack at 100% state of health. An untested salvage pack was available for 32.4% of models at 79.4% off, with no warranty at all and a state of health nobody has measured. EV Cable Hub's 2026 study found untested salvage packs arrived at a mean 84.2% state of health, so a large share of that discount is buying a pack that has already lost a sixth of its capacity.

The outcomes table is what turns those discounts into decisions and it does not flatter the cheapest option. Refurbished packs failed within 24 months in 8.6% of cases, tested salvage in 18.4% and untested salvage in 34.2%. Price in the cost of that subsequent failure and the effective cost of a refurbished pack is £4,392, a tested salvage pack £3,162 and an untested salvage pack £3,085. The headline gap between tested and untested salvage is £800; the effective gap is £77, and the untested pack comes with a 41.6% owner satisfaction rate against 68.2%. The discount is real and most of it is priced risk.

One figure in that table deserves stating on its own because it is what a buyer should ask for: a state-of-health certificate came with 88.4% of refurbished packs, 61.2% of tested salvage packs and 8.1% of untested ones. A certificate is the difference between a measured purchase and a hopeful one, and on the 2026 outcomes it tracks the failure rates closely enough to be used as a proxy for them.

Remanufactured packs sit awkwardly in the middle of this table and are the least understood of the four options. At £5,120 fitted they are 42.7% below a new main dealer pack with a two-year warranty and a 98.4% supplied state of health, which is close to new-pack territory on both measures. Availability is the catch: they existed for 51.5% of the models priced, the lowest of any option except untested salvage, because remanufacturing needs a supply of comparable cores and a specification to rebuild to.

Repair against replace#

61.4% of the battery faults quoted for in 2026 were resolved without replacing the pack. A single module replacement, the largest single repair route at 14.6% of all faults, cost a mean of £1,486 against £7,235 for a full pack, a saving of £5,749 on the job.

The repair hierarchy runs from a software job to a pack swap and the cheap end of it does more work than the category assumes. A software or battery management system recalibration, at a mean of £186 and 1.4 hours, resolved 8.4% of faults on its own. Cell balancing at £284 resolved a further 6.1%. Single cell replacement and rebalancing at £684 resolved 9.8%. Coolant circuit and seal work at £486 resolved 5.4%, contactor and high-voltage junction repairs at £742 resolved 4.2%, and a battery management system replacement at £912 resolved 6.8%. Module work (single at £1,486, multiple at £2,840) resolved 20.7% between them. A full pack replacement resolved the remaining 38.6%.

The barrier is not capability, and this is the genuinely useful reporting. EV Cable Hub's 2026 workshop panel found 71.4% of main dealer quotes were for a full pack replacement against 24.6% of independent specialist quotes on comparable faults, and 12.6% of main dealer quotes offered module repair against 58.4% at independents. At dedicated battery repair specialists the pattern inverts almost completely: 11.8% quoted a full replacement and 74.2% quoted module repair. The same fault presented to two workshops produces two different jobs. Diagnostic access and commercial model explain most of the gap, and the difference in mean quote, £8,940 against £4,180, follows directly from it.

The practical version for an owner is short. Ask whether the fault is module-repairable before accepting any replacement quote, ask for a module-level diagnosis rather than a pack-level one, and get a second quote. EV Cable Hub's 2026 study found owners who obtained a second quote paid a mean of £3,180 less, and 38.4% of those second quotes were for a repair rather than a replacement. A module-level diagnosis takes 2.4 hours and £202 of labour, against a mean diagnostic fee across all workshops of £142. That is a small and recoverable outlay set against the spread it can open up.

Misdiagnosis is the other half of the argument and it runs in one direction. 18.6% of faults were misdiagnosed as pack failure on first inspection, most often a battery management system communication fault reported as cell failure. Jobs that were misdiagnosed and then correctly diagnosed cost a mean of £1,240. Had the original misdiagnosis been acted on they would have cost £7,480. That is the cost of the second opinion in a single comparison, and it is why the tools further down this page lead with the repair-or-replace question rather than the price.

It is worth being precise about what the 61.4% figure does and does not say. It is the share of the faults presented to a workshop in this dataset that were resolved without a new pack, not the share of every battery problem in the country. Workshops were approached with a specified fault, so the mix is weighted towards faults serious enough to be quoted for. The population-level equivalent is in the incidence section: 3.6% of vehicles had a module-level repair against 1.4% that had a pack replaced.

• Owners who obtained more than one quote: 41.6%
• Mean saving where a second quote was obtained: £3,180
• Owners whose second quote was for a repair rather than a replacement: 38.4%
• Mean diagnostic fee across all workshops: £142
• Workshops offering module-level diagnosis: 46.2%
• Workshops with cell-level diagnostic capability: 28.4%
• Mean time for a module-level diagnosis: 2.4 hours
• Faults misdiagnosed as pack failure on first inspection: 18.6%
• Most common misdiagnosis: BMS communication fault reported as cell failure, 41.2% of misdiagnoses
• Second most common: 12V system fault reported as HV pack fault, 24.6%
• Third: coolant sensor fault reported as thermal runaway warning, 14.1%
• Mean cost of a job that was misdiagnosed and then correctly diagnosed: £1,240
• Mean cost had the misdiagnosis been acted on: £7,480

What actually fails#

Single cell or module failure accounted for 34.2% of EV battery faults in 2026 and battery management system faults for 21.6%. Genuine whole-pack failure, where no repair route existed, accounted for 11.4%.

The fault distribution explains the repair numbers. The two largest categories between them are 55.8% of all faults and both are repairable: a failed module is a module job at £1,486, and a battery management system fault is a £912 job that is warranty-covered 74.2% of the time. The expensive failure modes are rare. Water and moisture ingress at £3,840 is 4.2% of faults, physical impact damage at £8,240 is 3.4%, and a thermal event at £11,240 is 0.3%. The costs people fear and the faults people get are largely different populations.

Age at fault is the finding that matters most for used buying, and it does not describe the smooth wear-out curve people assume. Battery management system faults arrive at a mean of 4.1 years and 41,240 miles, moisture ingress at 4.6 years, impact damage at 3.2 years. Cell imbalance beyond correction arrives at 6.8 years and capacity falling below the warranty threshold at 7.9 years. Early faults are manufacturing and design defects; late faults are wear. Both exist, and the middle of a vehicle's life is the quietest part of it rather than the beginning.

The incidence curve by vehicle age shows the same shape. Faults per 1,000 vehicles per year run at 4.1 under a year old, dip to 3.1 between two and three years, and then climb steadily to 28.6 beyond ten years. The dip is real and it is the signature of manufacturing defects being cleared under warranty before wear has begun. EV Cable Hub's 2026 owner survey recorded warranty coverage falling in step: 100% under a year, 91.4% between three and four years, 41.6% between seven and eight, and 1.4% beyond ten. The risk to an owner's wallet rises for two reasons at once, and the second is the larger of them.

Mean owner cost where the owner paid follows that curve to a peak of £7,240 in the eight to ten year band and then falls back to £6,480 beyond ten years. That fall is not the work getting cheaper. It is owners of the oldest vehicles choosing refurbished packs and module repairs over new packs, because by then the pack cost has passed the value of the car.

Warranty coverage varies by fault as much as by age, which is the part owners find hardest to predict. Capacity falling below the threshold was covered 88.4% of the time and a battery management system fault 74.2% of the time, because both are unambiguous manufacturing questions. Physical impact damage was covered 4.1% of the time and moisture ingress 21.4%, because both invite an argument about cause. The faults most likely to be argued over are also among the most expensive to resolve, which is exactly the wrong way round for the owner.

Labour, time and workshop rates#

A full pack replacement took a mean of 8.4 hours at a mean labour rate of £81 an hour in 2026, a labour element of £684. A structural cell-to-body pack took 14.6 hours against 8.1 hours for a conventional modular pack.

Labour is the part of the bill that has been ignored, largely because it used to be small. EV Cable Hub's 2026 workshop panel timed twelve standard jobs. A pack-level diagnostic is 1.4 hours and £113. A module-level diagnostic is 2.4 hours and £202. A single module replacement is 6.8 hours and £551, and a full modular pack replacement 8.1 hours and £656. Post-fit calibration and a road test add 1.2 hours and £97 to any pack job and are not optional. The ranges matter as much as the means: a structural pack replacement ran from 11.2 to 21.4 hours, a spread of more than £900 in labour alone on the same job.

Rates vary widely by region. London workshops charged a mean of £108 an hour in 2026, 52% more than the £71 charged in the North East of England, with the UK mean at £81. The regional spread in the full replacement quote is narrower than the spread in the rate: £8,640 in London against £6,540 in the North East, a difference of 32.1%. Parts are priced nationally and only the labour element moves. Diagnostic fees track rates closely, from £186 in London to £122 in Northern Ireland.

Rate also varies by job type within the same workshop, which is worth knowing before comparing quotes. Diagnostics and software work were charged at £84 to £86 an hour in the 2026 panel, modular pack work at £81, cell-to-pack work at £84, coolant work at £78 and structural pack replacement at £88. The premium on structural work reflects the equipment and the certification it needs rather than the difficulty of the individual hour.

The connection back to the cost series is the one worth carrying forward. Labour is the only element of the bill that has risen in absolute terms since 2015, up 40.7% from £486 to £684 on a mean job, and its share of the total has more than doubled from 4.3% to 9.5%. EV Cable Hub's 2026 projection has that share reaching 14.1% by 2030 as cell costs keep falling. A bill that is increasingly labour is a bill that rewards shopping on workshop type and region, which the first calculator on this page is built to do.

Warranty terms by manufacturer#

Every one of the 26 manufacturers reviewed in 2026 used a 70% state-of-health threshold, and the mean warranty term was 8 years or 100,000 miles. One manufacturer offered 10 years as standard and one more offered an extension to 10.

The uniformity is the finding. Twenty-six manufacturers, one threshold, and a term that averages exactly eight years. The variation that exists sits in the mileage cap rather than the years: 80,000 miles at the shortest, 100,000 at almost everyone, 125,000 at one manufacturer and up to 150,000 on some variants at another. Two manufacturers offer seven years rather than eight. Every warranty reviewed was transferable, every one covered both capacity and failure, and not one applied a pro-rata reduction after any year. Not one of the 26 sets of terms mentions rapid charging frequency at all, which is worth knowing given that 31.4% of owners believe it voids their cover.

The distinction that causes most owner confusion is between a capacity claim and a failure claim, and almost no consumer coverage separates them. A capacity claim says the pack has degraded below 70% state of health. A failure claim says something has broken. They are assessed differently and they are paid at very different rates: EV Cable Hub's 2026 warranty and claims survey found 88.6% of failure claims paid against 71.4% of capacity claims, with 85.2% of all claims paid. Mean state of health at a capacity claim was 68.4%, which is to say claims are made when the pack is already the wrong side of the line rather than near it.

Declines cluster in four places. State of health above the threshold accounted for 28.4% of declines, physical damage 24.1%, exceeding the mileage cap 21.4% and non-approved repair history 18.6%. Three of those four are avoidable by the owner and the fourth is the mileage cap, which is where used buyers get caught: a warranty with four years left on the clock may have almost no miles left on it. 34.6% of owners appealed a decline and 28.4% of those appeals were upheld, so an appeal is worth making rather than assuming.

Awareness is poor and it is the reason the checklist on this page opens with warranty. 41.2% of owners did not know their warranty terms, 38.4% believed the warranty covered any degradation at all rather than degradation past a threshold, and mean time from claim to resolution was 34 days, with the longest case in the 2026 survey running to 186 days. An owner who knows the threshold, the cap and the expiry date is in a materially different position from one who does not.

One practical consequence of the mileage cap deserves its own line, because it is where a used buyer is most exposed. Eight years and 100,000 miles is not eight years. A three-year-old vehicle with 60,000 miles on it has five years of calendar cover and 40,000 miles of mileage cover, and on a high-mileage driver the second runs out first by a wide margin. Exceeding the cap accounted for 21.4% of all declines in EV Cable Hub's 2026 claims survey, and it is the decline reason that is entirely predictable in advance.

How many packs actually get replaced#

1.4% of electric vehicles in EV Cable Hub's 2026 sample had had a battery pack replaced, and 78.6% of those were covered by warranty. The mean age at replacement was 6.8 years and the mean mileage 74,180.

Incidence is the counterweight that stops this page reading as a scare, and it belongs next to the cost rather than after it. A £7,235 bill at a 1.4% incidence is a different consumer proposition from the same bill at a 40% incidence, and only 18.1% of those replacements were paid for by the owner at all: 78.6% went on warranty and 3.3% on insurance. The mean owner-paid cost, where an owner did pay, was £6,840.

Widen the definition and the picture is more useful. 3.6% of vehicles had a module-level repair rather than a replacement, so 5.1% had a battery intervention of any kind and 94.9% had none whatsoever. The gap between the 1.4% who needed a pack and the 5.1% who needed something is the repair story restated as population incidence: for every vehicle that needed a new pack, roughly two and a half more needed something smaller.

Age dominates everything. 0.4% of vehicles under five years old had a replacement, 1.8% between five and eight, 4.6% between eight and ten and 9.8% beyond ten. Thermal architecture matters almost as much: 6.4% of passively cooled packs had been replaced against 0.9% of actively cooled ones, a ratio of 7.11 to one. Those two variables between them describe most of the risk, and both are visible to a used buyer before they hand over any money.

The model group table is the used-buyer table and its ordering is unambiguous. Early passively cooled hatchbacks show an 8.1% replacement rate at a mean 8.4 years with only 12.2% of those replacements covered by warranty, and their dominant fault is capacity falling below the threshold. Current-generation LFP vehicles show 0.4% at a mean 2.8 years with 100% warranty coverage and a dominant fault of a battery management system fault, a £912 job. A twenty-fold difference in replacement rate across a single market, and it maps almost perfectly onto pack generation.

Pack cost against vehicle value#

A new pack fitted at an independent specialist cost 61.4% of a six-year-old vehicle's residual value in 2026, and 21.8% of the vehicles quoted for carried a pack cost above their residual value entirely. On a 24kWh Nissan Leaf the independent pack is 125.5% of the car's six-year residual.

The ratio is what turns a repair decision into a scrap-or-keep decision, and it crosses over at a predictable age. At three years a new independent pack is 36.4% of residual value and replacement is plainly economic. At six years it is 61.4% and marginal. At seven it is 76.2% and no longer economic on EV Cable Hub's 2026 assessment. At nine years the pack costs more than the car at 118.2%, and beyond ten years it reaches 186.8%.

A refurbished pack moves that crossover out by about four years, which is the most practically useful thing in this section. The same age curve on a refurbished pack runs 21.0% at three years, 35.3% at six, 55.2% at eight and 86.5% at ten, only passing the value of the car beyond ten years at 107.5%. Refurbished packs and module repairs are what keep older electric vehicles on the road, and the availability figures elsewhere on this page are therefore a residual value question as much as a repair one.

Insurance sits on the same arithmetic. The mean write-off threshold in EV Cable Hub's 2026 panel was 62.4% of value, which is a threshold a new pack crosses on any vehicle older than about six years. 0.9% of electric vehicles were written off with a battery fault as the primary cause and 2.4% with battery damage after an impact. Mean settlement where the pack was the cause was £11,840, and mean repair cost for impact damage was £8,240 with pack involvement against £3,140 without.

The finding in that table that ought to travel furthest is this: 18.4% of vehicles declared a total loss for pack damage had a fault that was repairable at module level. Insurers accepted an independent specialist quote in 51.4% of cases, a refurbished pack in 34.2% and a salvage pack in 8.1%, while 48.6% required a main dealer quote before settling, the workshop type that quotes a full replacement 71.4% of the time. The write-off decision is being made on the most expensive quote available roughly half the time, and premiums carry the result: electric vehicle premiums ran 21.4% above comparable petrol cars in 2026, with 46.8% of that uplift attributed to battery repair cost.

The write-off arithmetic also explains something about the salvage market that appears elsewhere on this page. A pack removed from a written-off vehicle is worth a mean of £1,840 at auction, or £28 per kWh, against £41 per kWh in second-life storage use and £22 per kWh in recycling recovery for NMC. Every vehicle written off for reasons other than the pack feeds that supply, which is why salvage packs are available for 44.1% of models while new packs at an independent reach 76.5%.

Cell and module component pricing#

Cells accounted for 58.4% of a new pack's parts cost at a mean of £74 per kWh in 2026, with module housings, the battery management system, cooling and enclosure making up the rest. A single replacement module cost a mean of £486 in parts before labour.

EV Cable Hub's 2026 teardown programme stripped 34 packs to module and cell level and costed each component from trade supply quotations rather than list prices. On a representative 60kWh pack that is £4,440 of cells, £744 of module housings and busbars, £624 of battery management system, £564 of cooling plates and circuit, £516 of enclosure and structure, £312 of high-voltage junction box and contactors, £246 of wiring and sensors and £168 of seals and thermal materials. Cells are 58.4% of that and the other seven components together are 41.6%, and it is that split rather than the total which decides what a repair can save.

Every component in that table is individually replaceable except the enclosure, and the cells are replaceable by a specialist rather than in general. That single column is the technical basis for the repair argument on this page: a pack is not one object, it is eight priced subsystems, seven of which come apart.

The observation that follows from the teardown is the one worth quoting. The failed component in a typical module-level fault is 3.1% of the pack's parts value. Replacing the whole pack therefore replaces 96.9% of components that were working, and pays for them. A mean module holds 24 cells, a mean pack holds 298 cells in 12.4 modules, and a mean module-level fault involves 2.4 failed cells. Two and a half cells out of nearly three hundred is the scale of the thing that £7,235 replaces.

The constraint on doing the smaller job is a tooling and licensing one rather than a skills one. A full manufacturer diagnostic tool licence costs an independent workshop a mean of £4,180 a year, and only 34.6% of independents hold one; 78.4% use a universal tool instead. That is the mechanism behind the quoting pattern in the repair section, and it is also why diagnostic access was one of the five drivers of the eleven-year fall in cost. Conductor and busbar specification inside a pack follows the same rules as elsewhere in an electric vehicle's high-current wiring, which our guide to copper and conductor gauge covers in detail.

• Mean cell cost per kWh, NMC: £78
• Mean cell cost per kWh, LFP: £61
• Mean cell cost per kWh, NCA: £84
• Mean cell cost per kWh, sodium-ion: £52
• Mean parts cost of a single module: £486
• Mean number of modules per pack: 12.4
• Mean cells per module: 24
• Mean cells per pack: 298
• Mean failed cells in a module-level fault: 2.4
• Failed component as a share of pack parts value: 3.1%
• Working components replaced in a full pack swap: 96.9%
• Mean BMS unit cost: £624
• Mean contactor set cost: £186
• Mean coolant plate cost: £141
• Mean HV junction box cost: £312
• Mean pack seal kit cost: £84
• Mean cost of a full diagnostic tool licence for an independent workshop: £4,180 per year
• Independent workshops holding a manufacturer diagnostic licence: 34.6%
• Independents using a universal diagnostic tool: 78.4%

Battery leasing, subscription and rental#

4.1% of UK electric vehicles in EV Cable Hub's 2026 sample were on a separate battery lease, at a mean of £62 a month. Over eight years that is £5,952, against a mean fitted replacement cost of £7,235.

The separate battery lease is a legacy product, concentrated in the used Renault Zoe and early Nissan Leaf market, and the arithmetic on it is closer than its reputation suggests. £5,952 over eight years against a £7,235 replacement that 98.6% of vehicles never need is not obviously good value on expectation, but it is not the rip-off it is usually described as either. It also covers replacement outright, transfers with the car and removes the tail risk entirely. What it does do is cut the residual value by 18.4%, and that is the part buyers miss, because it is paid at sale rather than monthly.

The traps are in the detail rather than the headline. Leases ran from £39 to £124 a month in the 2026 sample, carried a mean mileage cap of 7,500 a year and charged a mean 6.4p a mile beyond it. 14.6% of owners did not realise the battery was leased when they bought the car, which on a used purchase is a monthly liability arriving unannounced. 8.1% bought their lease out, at a mean cost of £3,840.

The newer products are different animals. A manufacturer battery health guarantee, held by 6.8% of vehicles, costs nothing, covers replacement above the state-of-health threshold, transfers with the car and lifts residual value by 2.1%. An extended battery warranty, held by 9.4%, costs a mean £18 a month or £1,728 over eight years and lifts residual by 3.4%. Third-party electric vehicle battery insurance at £24 a month covers replacement only partly and transfers only sometimes. Extended warranty claims were made by 4.6% of holders and 81.2% of those were paid.

What the survey says about demand is more interesting than the products themselves. 61.4% of used buyers consider a battery health guarantee important, 48.6% would pay more for a transferable one, and the mean premium they would pay is £684. The market is being asked for a documented, transferable health position rather than an insurance product, and the used and salvage panel found a documented health certificate was worth 4.1% at sale.

• Vehicles on a separate battery lease: 4.1%
• Mean monthly lease cost: £62
• Cheapest lease recorded: £39 per month
• Most expensive: £124 per month
• Mean mileage cap on a lease: 7,500 per year
• Excess mileage charge, mean: 6.4p per mile
• Residual value effect of a lease: -18.4%
• Owners who did not realise the battery was leased when they bought: 14.6%
• Owners who bought out their lease: 8.1%
• Mean buyout cost: £3,840
• Extended warranty uptake at point of sale: 9.4%
• Mean extended warranty cost: £1,728 over eight years
• Extended warranty claims made: 4.6%
• Extended warranty claims paid: 81.2%
• Owners who consider a health guarantee important when buying used: 61.4%
• Used buyers who would pay more for a transferable guarantee: 48.6%
• Mean premium they would pay: £684

What charging behaviour changes about the bill#

Vehicles with a passively cooled pack were replaced 7.11 times more often than actively cooled packs in EV Cable Hub's 2026 sample, 6.4% against 0.9%. Rapid charging frequency showed no measurable relationship with replacement incidence at all: 1.3% against 1.4%.

Three things raised replacement incidence in the 2026 data and they are all structural or habitual rather than about charging speed. Passive air cooling raised it 7.11 times. A mean resting state of charge above 90% raised it 4.00 times, 2.8% against 0.7% for packs kept between 50% and 70%. Sitting in the hottest ambient quartile raised it 3.00 times, 2.4% against 0.8%. Chemistry compounds all three: 0.4% for LFP, 0.9% for NMC and 7.8% for the early LMO blends that were fitted to the passively cooled generation in the first place.

Rapid charging frequency did not appear anywhere in the results. Drivers rapid charging more than four times a week showed a 1.3% replacement rate against 1.4% for drivers rapid charging less than once a week, a ratio of 0.93, which is to say the frequent rapid chargers were fractionally less likely to need a pack. That is not a licence to ignore charging behaviour, and it is not a claim that rapid charging has no effect on capacity; it is a statement that in this sample it did not move the replacement rate, and no manufacturer warranty in the 26 reviewed mentions it either.

The belief data runs almost exactly opposite to the measurements. 58.4% of owners believe rapid charging causes battery replacement, 26.1% believe high resting charge causes it and 8.4% believe passive cooling causes it. The factor with the largest measured effect is the one the fewest owners suspect, and the factor with no measured effect is the one most of them worry about. Anyone writing consumer guidance from public sentiment rather than from data will get this backwards, and most of it does.

The practical version is short and costs nothing at the point of use: buy a liquid-cooled pack and keep it off 100%. EV Cable Hub's 2026 modelling puts the expected saving from choosing an actively cooled pack at £1,840 over eight years and from keeping mean resting charge below 90% at £412, both measured in expected replacement cost rather than in fuel. Our study of which electric vehicles last longest covers the degradation side of the same question, and our fast charging and battery health study tests the rapid charging question on capacity rather than on replacement. On the home side, charging modes, charging current and 16A against 32A charging decide the thermal load a pack sees overnight, and our CCS charging cables and charging cable range cover the equipment.

The reason passive cooling dominates the results is that it is not really a behaviour at all, it is a design decision made years before the owner bought the car, and it cannot be undone afterwards. That makes it the single most useful thing on a used buyer's checklist and the cheapest to verify: the pack either has a liquid circuit or it does not, the specification says which, and the difference between the two on 2026 replacement rates is larger than the difference between any two charging habits measured here.

• Replacement rate, passively cooled packs: 6.4%
• Replacement rate, actively cooled packs: 0.9%
• Ratio: 7.11x
• Replacement rate, mean resting state of charge above 90%: 2.8%
• Replacement rate, mean resting state of charge 50% to 70%: 0.7%
• Ratio: 4.00x
• Replacement rate, hottest ambient quartile: 2.4%
• Replacement rate, mildest ambient quartile: 0.8%
• Ratio: 3.00x
• Replacement rate, more than four rapid charges a week: 1.3%
• Replacement rate, fewer than one rapid charge a week: 1.4%
• Ratio: 0.93x
• Replacement rate, LFP chemistry: 0.4%
• Replacement rate, NMC chemistry: 0.9%
• Replacement rate, early LMO blend: 7.8%
• Owners who believe rapid charging causes battery replacement: 58.4%
• Owners who believe high resting charge causes it: 26.1%
• Owners who believe passive cooling causes it: 8.4%
• Mean saving over eight years from keeping resting charge below 90%: £412 in expected replacement cost
• Mean saving from choosing an actively cooled pack: £1,840 in expected replacement cost

Availability, lead times and parts supply#

A new pack was available for 94.1% of models at a main dealer in 2026 on a mean lead time of 18 days, but for only 76.5% at an independent specialist. Refurbished packs were available for 61.8% of models and salvage packs for 44.1%.

Availability is the operational reality behind the price and it is the first question a fleet buyer asks. The ordering is broadly consistent, with one exception: the more the pack costs, the more likely it is to exist. New at a main dealer is available for 94.1% of models, new at an independent for 76.5%, remanufactured for 51.5%, refurbished for 61.8%, tested salvage for 44.1% and untested salvage for 32.4%. Lead times run the other way, from 18 days on a new main dealer pack down to 11 days on tested salvage and 6 on untested, because a salvage pack is already on a shelf somewhere.

The tail is where the problem lives. 5.9% of models had no new pack available at any price in EV Cable Hub's 2026 study, and 4.3% had no repair route at module level. On those vehicles the constraint is not cost, it is existence, and the honest answer to an owner is that the car has a finite remaining life regardless of what they are prepared to spend. The longest lead time recorded in the collection window was 148 days.

Time off the road is the cost nobody quotes. A full replacement took a mean of 11.4 days off the road against 3.8 days for a module repair, and a courtesy vehicle was provided on 61.4% of jobs. For a household with one car that difference is worth more than some of the price gap, and it points the same way as everything else in this study: the smaller intervention is cheaper, faster and available in more places.

Supply behaviour explains part of the lead time spread. Only 12.6% of workshops hold pack stock rather than ordering against a job, and 44.1% of independent specialists have a salvage pack supply agreement in place. Customers obtained a mean of 1.7 quotes and 18.4% waited more than 30 days simply to be given a price, which is worth setting against the £3,180 mean saving from a second quote. The wait is the reason most owners do not get one.

Availability also explains a pattern in the pricing that looks odd at first. Refurbished packs are available for more models than remanufactured ones despite being the cheaper product, because refurbishment needs only a donor module rather than a rebuild specification. The practical result for an owner of an older vehicle is that the option most likely to exist for their car is also the one that keeps it economic to repair, which is the single most important thing the repair supply chain does for residual values.

• New pack availability, main dealer: 94.1%
• New pack availability, independent: 76.5%
• Remanufactured availability: 51.5%
• Refurbished availability: 61.8%
• Salvage availability: 44.1%
• Mean lead time, new at main dealer: 18 days
• Mean lead time, new at independent: 24 days
• Mean lead time, remanufactured: 21 days
• Mean lead time, refurbished: 16 days
• Mean lead time, salvage: 11 days
• Longest lead time recorded: 148 days
• Models with no new pack available at any price: 5.9%
• Models with no repair route at module level: 4.3%
• Mean courtesy vehicle provision during a replacement: 61.4% of jobs
• Mean days off the road for a full replacement: 11.4
• Mean days off the road for a module repair: 3.8
• Workshops holding pack stock rather than ordering: 12.6%
• Independent specialists with a salvage pack supply agreement: 44.1%
• Mean number of quotes a customer obtained: 1.7
• Customers who waited more than 30 days for a quote: 18.4%

Cost projection to 2030#

EV Cable Hub's 2026 model projects UK battery replacement cost falling to £84 per kWh by 2030, a further 33.9% below the 2026 figure of £127, with a low scenario of £71 and a high scenario of £98. Labour is projected to rise from 9.5% to 14.1% of the total bill over the same period.

Everything in this section is modelled rather than measured and is labelled as such throughout. The central case runs £114 per kWh in 2027, £102 in 2028, £92 in 2029 and £84 in 2030, with the projected mean full pack cost falling from £7,235 to £5,940 across the same years. The band around it widens with distance: £15 wide in 2027 and £27 wide in 2030, which is the honest width of a four-year projection in a market where cell prices move on decisions taken elsewhere.

The assumptions are printed in full rather than summarised, because a projection with hidden assumptions is worthless as a citation. The central case assumes cell costs fall 9.0% a year, labour rates rise 3.4% a year, LFP reaches 48% of the replacement market by 2030, refurbished packs reach 38%, the independent sector reaches 71% of quotes and 82% of models have a module repair route. The low scenario is not simply the central case with better numbers: it assumes a 12.0% annual cell cost decline and slower labour inflation at 2.4%, and those two move in opposite directions in the high case, at 5.0% and 4.8%. Quote the range rather than the central line alone.

The composition shift is the more interesting finding and it is the one with consequences. As cells get cheaper the bill becomes a labour and diagnostics bill, and the projected labour share rises every year from 9.5% to 14.1%. That changes who can do the work profitably and where the price competition lands: a market where parts are the bill rewards buying power, and a market where labour is the bill rewards efficiency, tooling and regional rate. It also means the regional spread in the labour section grows in importance rather than shrinking.

One thing the projection deliberately does not do is assume a repair share. The share of faults resolved at module level is a function of diagnostic access and pack architecture, and those pull against each other: access is improving while structural packs are spreading. The 2026 measurement of 61.4% is the figure to carry forward until it is measured again, and EV Cable Hub will publish this projection's error against measured data in next year's edition.

What owners believe against what the data shows#

UK electric vehicle owners estimated the cost of a battery replacement at a mean of £14,180 against a measured £7,235 in EV Cable Hub's 2026 survey, 96% too high. They put the share of vehicles needing a replacement by ten years at 28.4% against a measured 1.4%.

Every belief measured in the 2026 survey overstates the cost and the risk, and the overstatements are not small. A repair was put at £6,840 against a measured £1,486, an error of 360%. Cost per kWh was put at £248 against £127. The share of faults needing a full pack was put at 84.1% against a measured 38.6%. The single largest error was on incidence: 28.4% against 1.4%, out by a factor of twenty. Where beliefs understate, they understate the protections. Warranty length was put at 5.4 years against a measured 8, and the share of claims paid at 51.2% against a measured 85.2%.

The behavioural consequence is what makes this section matter rather than merely entertain. 34.6% of used car buyers said battery replacement cost had put them off buying an electric vehicle. Only 18.4% had researched the actual cost before deciding, 9.1% found a figure they trusted and 71.4% found conflicting figures. That last number is the market failure this study exists to correct: the information is not absent, it is contradictory, and a contradictory answer is treated as a bad answer.

48.6% of those buyers said they would reconsider given the measured figures, which is the size of the decision being made on a number that is roughly double the real one. The same anxiety shows up in ownership plans: mean expected ownership period is 4.6 years for an electric vehicle against 6.8 for a petrol or diesel car, and owners attribute 41.2% of that gap to battery cost anxiety. 26.8% say they would sell before the warranty expires purely because of the risk.

The correction is not that batteries are cheap. A £7,235 bill is a serious bill and 21.8% of the vehicles quoted for carried a pack cost above their residual value. The correction is that the typical outcome is not the headline outcome: most faults are repairs, most replacements are covered by warranty, and most vehicles never need either. Only 12.4% of owners have set money aside for a replacement, at a mean of £3,840, which is, as it happens, close to the mean refurbished pack price of £4,180.

There is a supply-side reading of this section too. 71.4% of buyers found conflicting figures, which is the predictable result of a category where the standard published answer is a range so wide it cannot be wrong. A range from five to fifteen thousand pounds is not information, it is an admission that nobody measured. The remedy is not a narrower guess but a distribution: a mean, a median, quartiles, a per-kWh figure and a model-level table, all of which are on this page and all of which can be checked.

Interactive tools#

Three calculators built on the 1,846 quotes collected for the 2026 study, a comparator across all 74 vehicle variants priced, a searchable table of every figure on this page, and a twenty-five item protection checklist that remembers where you got to. Everything runs in the browser.

Each tool reads the tables above rather than a separate dataset, so every figure they return can be checked against the published table it came from. Where a tool applies arithmetic of its own (a cost per kWh, a share of residual value, an expected cost across an ownership period), the arithmetic is stated in the note under the tool.

Methodology#

Every figure on this page comes from one of four EV Cable Hub studies: 1,846 quotes from 214 workshops across 68 vehicle models, 34 packs stripped and costed to component level, 3,180 surveyed owners and 5,860 tracked used and salvage listings, all collected between September 2025 and June 2026.

Frequently asked questions#

Thirty questions on EV battery replacement cost, each answered with the measured 2026 figure first.

Every answer below is drawn from the tables on this page. Where a figure is modelled rather than measured it is described as such.