Home › Guides › Check EV battery health at home
Checking EV battery health at home: how accurate is the car’s own figure?
Published · Last updated · 500 UK certificates · 42 models
You can often read a used EV’s battery health (state of health, SoH) yourself – from a menu in the car, the maker’s app or a cheap OBD-II adapter and a phone app. But that number is the car’s own estimate, worked out by its battery management system (BMS). How far can you trust it? We compared it with an independent test on 500 used EVs sold by a UK dealer, whose published AVILOO certificates print both figures side by side.
1. Ways to read battery health yourself
| Method | What you get | Watch out for |
|---|---|---|
| The car’s own menu or test (e.g. Tesla’s Battery Health test) | The car’s own (BMS) figure. Tesla’s test needs the car plugged into an AC charger of at least 5 kW, below 20% charge, and takes up to 24 hours (Tesla UK) | Not something you can do during a short viewing – ask the seller for a dated photo of the result next to the odometer |
| OBD-II adapter + phone app | The BMS figure, sometimes cell voltages. Examples: LeafSpy reads a Nissan Leaf’s BMS through a Bluetooth or Wi-Fi OBD-II adapter with the car switched on (LeafSpy); CanZE does the same for a Renault Zoe (CanZE); owners of the MG5, Fiat 500e and Honda e post Car Scanner readings (MGEVs.com owner forum, Fiat500Owners.com owner forum, Honda E Forum) | Not every car can be read: owners report third-party apps can’t read the Mercedes EQA’s battery data (Mercedes EQA Forum – eqa-forum.de). One app can show more than one “SoH” value – on a Fiat 500e, Car Scanner shows both “SOH” and “SOH Low” (Fiat500Owners.com owner forum) |
| Dealer or workshop read-out | The BMS figure from the maker’s diagnostic tool | Still the car’s own estimate; ask for a printed, dated report |
| Generational check (UK) | The BMS figure read over OBD in about 2 minutes, on a certificate (Generational) | Same source of the number as an app, but read in a standard way and documented |
| AVILOO test (independent) | AVILOO’s own calculation, which it says is within ±3% in at least 95% of reference measurements; many certificates also print the BMS figure (AVILOO GmbH) | Booked through a dealer or a mobile tester; compare like with like |
Everything except the last row reads the same number: the BMS estimate. AVILOO itself says the BMS value uses manufacturer-specific algorithms and can be too high or too low (AVILOO GmbH). The range readout on the dashboard is not a battery health figure at all – some makers release reserve capacity so the displayed range stays stable while the battery ages (Recurrent).
2. How close is the car’s figure to an independent test? (500 UK cars)
A UK dealer, Car360, publishes AVILOO certificates for many of the EVs it sells. Each certificate prints AVILOO’s independent SoH and the car’s own BMS SoH from the same test. We took every UK certificate in our dataset with both figures (500 cars, 42 models; certificates from Czech, Swiss, Swedish and other non-UK sellers are left out) and subtracted AVILOO’s figure from the car’s.
| Car’s own figure vs AVILOO | Cars | Share |
|---|---|---|
| More than 3 points higher | 21 | 4% |
| 1–3 points higher | 123 | 25% |
| Within 1 point | 238 | 48% |
| 1–3 points lower | 91 | 18% |
| More than 3 points lower | 27 | 5% |
- Usually close: within 2 points on 371 cars (74%), within 3 points on 452 (90%). The median difference was +0.2 points – no overall bias up or down.
- “100%” is common and doesn’t mean “as new”: the car’s own figure was 100% or more on 116 cars (23%), where AVILOO measured 95.3%–100.2%.
- The big misses: the car read much lower than AVILOO on a Zoe 22 kWh (23.3 kWh gross) (car 66.0% vs AVILOO 80.4%, 40,791 mi), a Model 3 82.1 kWh (car 85.0% vs AVILOO 92.7%, 36,811 mi), a Leaf 24 kWh (first generation) (car 60.0% vs AVILOO 66.6%, 51,590 mi); and much higher on an EQB 300/350 4MATIC 66.5 kWh usable (car 106.0% vs AVILOO 98.7%), an EQA 66.5 kWh (car 103.0% vs AVILOO 98.0%), an i3 120 Ah (42 kWh) (car 96.0% vs AVILOO 91.2%).
- A second dataset agrees in direction: in Czechia, Slovakia and Poland, the AVILOO and the car's own (BMS) reading were within 2 points for 62% of cars (our calculation from the published dataset) (Zuvala R., Bucsuhazy K., Kadula L., Dedek J., via Zenodo).
By mileage
| Mileage | Cars | Median difference (car − AVILOO) | Within 3 points | Car showed 100%+ |
|---|---|---|---|---|
| under 20,000 mi | 218 | +0.3 | 202 (93%) | 73 |
| 20,000–40,000 mi | 199 | +0.1 | 182 (91%) | 31 |
| 40,000–60,000 mi | 60 | +0.1 | 52 (87%) | 8 |
| 60,000 mi and over | 23 | +0.6 | 16 (70%) | 4 |
Few certificates are above 60,000 miles, so the last row is thin. Mileage converted from the miles printed on each certificate.
3. Some makes read high, some read low
The averages hide clear patterns by model. Where a model’s car figure reads high, a home reading flatters the battery; where it reads low, a worrying app reading may be better than it looks.
| Model | UK certificates | Median difference (car − AVILOO) | Car higher by >1 pt | Car lower by >1 pt | Car showed 100%+ | AVILOO range |
|---|---|---|---|---|---|---|
| Hyundai IONIQ 5 | 13 | +2.3 | 12 | 0 | 12 | 95.9%–98.6% |
| Kia EV6 | 10 | +2.2 | 8 | 0 | 9 | 96.2%–99.5% |
| Kia e-Niro / Niro EV | 20 | +2.2 | 19 | 1 | 16 | 94.1%–98.5% |
| Smart #1 | 5 | +1.6 | 4 | 0 | 4 | 97.3%–99.1% |
| Hyundai Kona Electric | 15 | +1.6 | 11 | 0 | 11 | 89.2%–99.5% |
| Mercedes-Benz EQA | 17 | +1.4 | 9 | 0 | 6 | 95.6%–99.8% |
| MINI Cooper SE | 21 | +1.1 | 12 | 0 | 4 | 95.2%–97.9% |
| Audi Q4 e-tron | 19 | +0.9 | 8 | 1 | 4 | 91.7%–99.8% |
| BYD Seal | 12 | +0.7 | 4 | 0 | 8 | 97.8%–99.9% |
| Volvo EX30 | 7 | +0.6 | 0 | 1 | 4 | 96.5%–99.4% |
| Mercedes-Benz EQB | 35 | +0.6 | 12 | 0 | 14 | 95.2%–100.0% |
| Xpeng G6 / P7 | 9 | +0.5 | 0 | 0 | 7 | 98.7%–99.9% |
| Volkswagen ID.4 | 8 | +0.5 | 3 | 1 | 1 | 89.6%–98.8% |
| Vauxhall/Opel Mokka-e (Mokka Electric) | 11 | +0.5 | 5 | 3 | 2 | 94.0%–100.1% |
| MG4 EV | 16 | +0.4 | 2 | 4 | 1 | 94.1%–99.4% |
| Renault Zoe | 9 | +0.3 | 2 | 2 | 0 | 80.4%–96.7% |
| Tesla Model 3 | 15 | +0.3 | 2 | 1 | 0 | 90.5%–97.6% |
| Ford Mustang Mach-E | 11 | +0.2 | 2 | 0 | 1 | 94.2%–98.4% |
| MG5 EV | 13 | +0.2 | 3 | 0 | 1 | 90.9%–95.7% |
| Tesla Model Y | 16 | +0.1 | 1 | 0 | 0 | 89.9%–97.8% |
| BYD Atto 3 | 21 | −0.1 | 1 | 2 | 0 | 93.9%–96.9% |
| MG ZS EV | 22 | −0.2 | 4 | 8 | 1 | 90.1%–98.0% |
| Polestar 2 | 20 | −0.4 | 1 | 8 | 1 | 92.4%–98.4% |
| Volvo XC40 Recharge | 9 | −0.7 | 0 | 4 | 0 | 94.5%–97.0% |
| Škoda Enyaq iV | 14 | −0.7 | 2 | 6 | 1 | 88.0%–99.5% |
| Volvo C40 Recharge | 5 | −0.8 | 0 | 1 | 0 | 94.9%–98.1% |
| Cupra Born | 14 | −0.9 | 1 | 7 | 0 | 94.2%–99.6% |
| Nissan Leaf | 5 | −1.1 | 0 | 3 | 0 | 66.6%–95.6% |
| BMW i4 | 21 | −1.2 | 0 | 13 | 0 | 95.0%–98.5% |
| Peugeot e-208 | 8 | −1.4 | 2 | 4 | 1 | 90.8%–99.2% |
| Jaguar I-Pace | 10 | −1.5 | 0 | 6 | 0 | 92.2%–97.1% |
| Peugeot e-2008 | 13 | −1.6 | 4 | 8 | 4 | 90.6%–100.2% |
| Volkswagen ID.3 | 13 | −1.7 | 0 | 11 | 0 | 92.4%–97.8% |
| Audi e-tron / Q8 e-tron | 10 | −2.5 | 0 | 9 | 0 | 92.0%–98.3% |
| GWM Ora 03 / Funky Cat | 11 | −3.1 | 0 | 11 | 0 | 91.0%–92.7% |
Models with at least 5 UK certificates. Fewer than 5: BMW i3 (4), Vauxhall/Opel Corsa-e (Corsa Electric) (3), BYD Dolphin (4), Dacia Spring (3), Fiat 500e (2), Renault Megane E-Tech (4), Honda e (2). Every certificate is listed with its link on the model data pages.
- Read high (median +1.5 points or more): Hyundai IONIQ 5 (+2.3), Kia EV6 (+2.2), Kia e-Niro / Niro EV (+2.2), Smart #1 (+1.6), Hyundai Kona Electric (+1.6). On these, discount a home reading – a car showing 100% may measure a few points lower.
- Read low (median −1.5 points or less): Peugeot e-2008 (−1.6), Volkswagen ID.3 (−1.7), Audi e-tron / Q8 e-tron (−2.5), GWM Ora 03 / Funky Cat (−3.1). A low-looking home reading on these may understate the battery.
- TÜV NORD and Carly’s analysis of about 50,000 BMS readings read over OBD gives brand-level patterns for the car’s own figure (TÜV NORD GROUP with Carly); it doesn’t compare them with an independent test.
4. What this comparison can and can’t tell you
- AVILOO has its own tolerance. It states ±3% in at least 95% of reference measurements (AVILOO GmbH), so a 1–2 point difference can be either test.
- Whole numbers. Most certificates print the car’s figure as a whole number, so differences under 1 point mean little.
- One dealer’s stock. These are cars one UK dealer chose to sell, mostly young and low-mileage. Older, high-mileage cars are under-represented.
- One moment. The BMS estimate moves: a DTU study saw a 62 kWh Leaf’s figure step up and down roughly every 90 days (DTU – Technical University of Denmark), and a 2013 Zoe’s varied between 84.0% and 94.0% in one year (Technical University of Denmark).
- Same test, not an app. The car figure on a certificate was read by AVILOO’s own equipment. A phone app may show a different value or label from the same BMS.
- Season. Cold weather cuts range, which is not the same as SoH. The winter section of the buying guide compares these certificates dated December to February with those from other months.
5. When is a home reading enough?
- Use it as a first filter. A home or seller’s reading that is well below the model data at similar mileage is a reason to ask questions or walk away; a high one is not proof of a good battery.
- Get an independent test when the stakes are high: older or high-mileage cars, cars near the warranty’s age or mileage limit, a model from the “reads high” list, or a reading that doesn’t fit the mileage. Options and prices: EV battery health check UK.
- Compare like with like. Compare a car’s own figure with other cars’ own figures, and an AVILOO result with AVILOO data. Each model page marks which method every data point used. Reading a certificate: how to read a battery certificate.
- Record it properly. Ask for a dated reading with the odometer visible, and note which app or tool produced it.
- Know what “SoH” means. Buffers, usable vs nominal capacity and warranty thresholds: battery state of health explained.
Model guides with step-by-step home checks: Nissan Leaf (LeafSpy) · Renault Zoe (CanZE) · Tesla Model 3 · Tesla Model Y · Kia EV6 · Hyundai Kona Electric · Honda e. Also see: is a used EV worth buying? The battery data · used-EV battery checklist.
Sources used on this page
514 sources · 500 data points: 0 from non-AVILOO methods, 500 measured with AVILOO tests.
- Large test datasets (1 source): TÜV NORD GROUP with Carly
- Connected-car / telematics (1 source): Recurrent
- Academic research (2 sources): DTU – Technical University of Denmark, Technical University of Denmark
- Single-car tests (500 sources, 500 data points): Car360, [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42], [43], [44], [45], [46], [47], [48], [49], [50], [51], [52], [53], [54], [55], [56], [57], [58], [59], [60], [61], [62], [63], [64], [65], [66], [67], [68], [69], [70], [71], [72], [73], [74], [75], [76], [77], [78], [79], [80], [81], [82], [83], [84], [85], [86], [87], [88], [89], [90], [91], [92], [93], [94], [95], [96], [97], [98], [99], [100], [101], [102], [103], [104], [105], [106], [107], [108], [109], [110], [111], [112], [113], [114], [115], [116], [117], [118], [119], [120], [121], [122], [123], [124], [125], [126], [127], [128], [129], [130], [131], [132], [133], [134], [135], [136], [137], [138], [139], [140], [141], [142], [143], [144], [145], [146], [147], [148], [149], [150], [151], [152], [153], [154], [155], [156], [157], [158], [159], [160], [161], [162], [163], [164], [165], [166], [167], [168], [169], [170], [171], [172], [173], [174], [175], [176], [177], [178], [179], [180], [181], [182], [183], [184], [185], [186], [187], [188], [189], [190], [191], [192], [193], [194], [195], [196], [197], [198], [199], [200], [201], [202], [203], [204], [205], [206], [207], [208], [209], [210], [211], [212], [213], [214], [215], [216], [217], [218], [219], [220], [221], [222], [223], [224], [225], [226], [227], [228], [229], [230], [231], [232], [233], [234], [235], [236], [237], [238], [239], [240], [241], [242], [243], [244], [245], [246], [247], [248], [249], [250], [251], [252], [253], [254], [255], [256], [257], [258], [259], [260], [261], [262], [263], [264], [265], [266], [267], [268], [269], [270], [271], [272], [273], [274], [275], [276], [277], [278], [279], [280], [281], [282], [283], [284], [285], [286], [287], [288], [289], [290], [291], [292], [293], [294], [295], [296], [297], [298], [299], [300], [301], [302], [303], [304], [305], [306], [307], [308], [309], [310], [311], [312], [313], [314], [315], [316], [317], [318], [319], [320], [321], [322], [323], [324], [325], [326], [327], [328], [329], [330], [331], [332], [333], [334], [335], [336], [337], [338], [339], [340], [341], [342], [343], [344], [345], [346], [347], [348], [349], [350], [351], [352], [353], [354], [355], [356], [357], [358], [359], [360], [361], [362], [363], [364], [365], [366], [367], [368], [369], [370], [371], [372], [373], [374], [375], [376], [377], [378], [379], [380], [381], [382], [383], [384], [385], [386], [387], [388], [389], [390], [391], [392], [393], [394], [395], [396], [397], [398], [399], [400], [401], [402], [403], [404], [405], [406], [407], [408], [409], [410], [411], [412], [413], [414], [415], [416], [417], [418], [419], [420], [421], [422], [423], [424], [425], [426], [427], [428], [429], [430], [431], [432], [433], [434], [435], [436], [437], [438], [439], [440], [441], [442], [443], [444], [445], [446], [447], [448], [449], [450], [451], [452], [453], [454], [455], [456], [457], [458], [459], [460], [461], [462], [463], [464], [465], [466], [467], [468], [469], [470], [471], [472], [473], [474], [475], [476], [477], [478], [479], [480], [481], [482], [483], [484], [485], [486], [487], [488], [489], [490], [491], [492], [493], [494], [495], [496], [497], [498], [499], [500]
- Owner-reported (4 sources): Mercedes EQA Forum – eqa-forum.de, Fiat500Owners.com owner forum, MGEVs.com owner forum, Honda E Forum
- Manufacturer (1 source): Tesla UK
- Other (specs, guides, news, surveys) (5 sources): AVILOO GmbH, Generational, Zuvala R., Bucsuhazy K., Kadula L., Dedek J., via Zenodo, LeafSpy, CanZE
Owner-reported and listing values are single cars (n=1) read from the car, an app or an OBD tool – useful examples, not averages. Range-retention figures (Recurrent) measure range, not capacity. Full details: methodology & all sources.
We are not paid by the battery-test providers mentioned on this page. Links marked "paid link" in the "Useful products and services" box are affiliate links. See our disclosure.
Get this specific car tested
We are not paid by these test providers. How we make money.
Model averages can't tell you about the car in front of you. An independent battery test (typically a short OBD plug-in test, some remote) gives an SoH figure and often a certificate you can use when negotiating.
- AVILOO – Independent OBD-based battery test (FLASH / Premium) with certificate; available via partners in the UK and EU.
- Generational – UK OBD plug-in test with certificate (SoH, range, remaining warranty, cell balance); sold through dealers and independent mobile testers.
- Moba – Postal OBD kit: you run the test yourself and post the box back for a digital battery certificate (Europe – check vehicle support and delivery country).
Prices change – check the provider's site. Provider names are used for identification only. Compare all UK options and prices: EV battery health check UK.
Useful products and services
- Charging cable: check that the car comes with its charging cable(s) at the viewing. If one is missing, replacements are sold separately: Type 2 charging cables at EV King paid link.
- Tyres: check tread depth and wear at the viewing. To price replacements, search by the size printed on the tyre wall: car tyres at Tyres UK paid link.
- Leasing instead of buying used (UK): if you'd rather not take on an older battery, you can compare electric car lease deals: EV lease deals at Leasing.com paid link.
- Parts and tools: wiper blades, bulbs, 12V batteries and tools for the car you buy: GSF Car Parts paid link (UK).
Some links are paid links: we may earn a small commission at no extra cost to you. This never changes the battery data we show (disclosure).
FAQ
Can I check an EV battery’s health at home?
Often, yes – but what you get is the car’s own estimate from its battery management system (BMS). Some cars show it in a menu or run their own test (Tesla), and on many others an OBD-II adapter and a phone app can read it (for example LeafSpy on a Nissan Leaf or CanZE on a Renault Zoe). It is not an independent test, and not every car can be read: owners report that third-party apps can’t read the Mercedes EQA’s battery data.
How accurate is the battery health figure the car shows?
On 500 UK dealer certificates in our data that print both AVILOO’s independent result and the car’s own figure, the two were within 3 points on 452 cars (90%) and within 1 point on 238 (48%). The car’s figure was more than 3 points higher on 21 and more than 3 points lower on 27. So it is usually close, but not always, and some makes read consistently high or low.
My EV says 100% battery health – is that real?
Not necessarily. 116 of the 500 UK certificates show the car’s own figure as 100% or more, while AVILOO measured 95.3%–100.2% on those same cars. A “100%” from the car tells you the battery is probably in good shape, not that it is as new.
Is an OBD app reading good enough when buying a used EV?
It is a useful first check, especially on a low-mileage car. For an older or high-mileage car, a model whose own figure tends to read high, or a car with a reading far below the model data, get an independent test with a certificate before you pay – it costs far less than a battery problem.
Why does the car’s battery health figure change from day to day?
The BMS estimate is updated from time to time and moves with temperature and charging. A Technical University of Denmark study saw one 62 kWh Nissan Leaf’s figure step up and down roughly every 90 days, and one 2013 Renault Zoe’s figure varied between 84.0% and 94.0% within a single year. Compare readings taken in similar conditions, and treat one reading as a snapshot.
Get alerts
We email (from alerts@usedevbatteryhealth.com) when new battery data is added for the model you choose: the new readings, their sources and what they mean for buyers. Low volume, no spam, unsubscribe any time. We'll first send a confirmation email – you're only subscribed once you click the link in it.
We currently publish 1393 sourced data points across 46 models (last updated 2026-11-17).
Emails are sent via MailerLite (EU). We never share your address. Every email has an unsubscribe link. You can stop any time.
Own an EV? Share your battery health
Every owner report makes this data more useful. It takes 2 minutes. Add your car's data