CarGearBench

We don’t test the gear. We read the people who do — and we show our working.

Every figure names its lab and its conditions · We earn a commission if you buy through our links

Problems · battery

It died in July. You found out in January.

Everyone blames the cold morning. The cold morning is when you discover it. The damage was done in the heat, and there is a measured gap to prove it — batteries in the far north of the United States last 58 months or more, and in the far south under 41.

Heat does the damage, cold reveals it

Service life by region — AAA's figures, reported by Consumer Reports
  • Farthest northern regions 58 months or more
  • Farthest southern regions under 41 months

Scale runs to 58 months, the highest measured value here.

use it Consumer Reports, How Hot Weather Affects Your Car Battery and What to Do About It, June 2026. Consumer Reports states it tests 150 batteries a year including five examples of every rated model, and that its life score comes from repeatedly discharging and recharging above 167 F for 15 weeks or until performance falls to unacceptable levels, simulating engine-bay heat. The regional service-life figures quoted on this page are AAA's, reported by CR, and come from AAA's own service data rather than from a laboratory. The per-battery ratings themselves are behind CR's membership wall and we have not seen them, so no battery is ranked here.

At least 17 months of difference, from geography alone. Not driving style, not brand, not price — ambient heat. Consumer Reports puts it plainly: summer heat is tougher on a battery than winter’s chill, because higher temperatures act on the chemistry inside. Cold does not degrade the battery; it raises what the starter demands on the morning the battery can give least.

Their test is built around exactly that. The life score comes from discharging and recharging a battery above 167 °F for 15 weeks, or until it falls to unacceptable performance — engine-bay temperature, not weather. AAA, for its part, reports responding to nearly two million battery calls each summer.

The practical consequence is a change of season, not a change of product: have it load tested at the end of summer, not at the start of winter. By the first cold morning the damage is already done and the test only confirms it.

The lab that tests 150 a year won't name a brand

Consumer Reports tests 150 batteries a year, five examples of every rated model — the sample size almost nothing else in this category gets. And having done that, here is what they say about choosing one:

There’s significant variation in which is the top performer from year to year, and from size to size. This makes it impossible to make simple recommendations by brand or model.

Consumer Reports, on its own battery testing

They add that you should not assume buying the same model you are replacing will get you the same result. We are not going to out-guess that. Their per-battery scores sit behind a membership wall which we have not read, and we are not going to rank batteries from anything weaker than what they have.

So the rest of this page is the part that needs no ratings at all: what is stamped on the battery in front of you, and what you can check yourself.

Read the sticker before you pay

A battery starts ageing when it is made, not when you fit it. The manufacture date is on a sticker on the top or the side, and Consumer Reports’ advice is to buy one less than six months old, and preferably three months or newer.

It appears either as digits — 7/25 for July 2025 — or as a letter and a digit, where the letter is the month and the digit is the year. So July 2025 is G5 or G-5.

The month letter — note there is no I
  1. A January
  2. B February
  3. C March
  4. D April
  5. E May
  6. F June
  7. G July
  8. H August
  9. K October
  10. L November
  11. M December

The letter I is skipped, because it reads as a 1. That is the one that catches people out: after H comes J, so a “J” sticker is September and not October. Miscount it and you will think a battery is a month younger than it is.

What you can act on without any ratings

  1. The group size. It is not a quality rating — it is the physical size, terminal type and terminal placement, in the way AA and AAA describe a size. It must match, or it will not fit the tray or reach the leads. Read it off the battery in the car, and check the manual too if you bought the car used.
  2. A date code under six months. Free to check, and the one spec on the box that is a fact rather than a claim.
  3. Load test annually once it is two years old. That is CR’s testing lead’s own advice, and it is what tells you the battery can still hold voltage under load — which resting voltage does not.
  4. Consider AGM if your temperatures are extreme. CR reports that most AGM batteries excel in its heat-focused life testing, and that they tolerate deep discharge — down to 10.5 V or below, as when the lights are left on overnight. They cost more, and heat still shortens them.
  5. Some flooded batteries come in North and South versions. North emphasises cold cranking amps; South carries a higher electrolyte-to-lead ratio for heat durability. Most shops stock the one that suits where they are, which is a reason to be careful buying online from another climate.

Your new battery may need registering

On some recent Audi, BMW, Ford and Mercedes-Benz models the new battery must be registered with the car so the charging system can adapt to it. That takes a scan tool. Fit one without registering and the car may keep charging to the old battery’s profile.

It is worth knowing before you order, because it can turn a twenty-minute driveway job into a workshop visit. Registration is a bidirectional operation — the tool has to command the car, not just read it — which is exactly the capability that separates a $16 scanner from a $350 one, and which most cheap scanners do not have.

What to buy

Batteries are group-size specific, so that link is a fitment search. The charger is not — it fits any car, and it is the one purchase here that acts on the actual failure mode, because a battery left to sit at partial charge sulphates and loses capacity whatever the weather is doing.

Affiliate links — we earn a commission at no extra cost to you and it does not change what we recommend. How this site makes money.

What we cannot tell you

  1. Which battery to buy. The strongest testing in this category is behind a membership wall we have not read, and its authors say the winner moves between years and sizes anyway. Naming one from weaker evidence would be inventing a recommendation.
  2. What a date code is worth in months of life. “Under six months old” is CR’s guidance, not a measured curve. Nobody has published how much service life a battery loses per month sitting on a shelf.
  3. Your own climate. The 58-and-under-41 figures are the two extremes of a country. Where a specific city falls between them is not published, so treat them as the range rather than as your number.

use it Consumer Reports, How Hot Weather Affects Your Car Battery and What to Do About It, June 2026. Consumer Reports states it tests 150 batteries a year including five examples of every rated model, and that its life score comes from repeatedly discharging and recharging above 167 F for 15 weeks or until performance falls to unacceptable levels, simulating engine-bay heat. The regional service-life figures quoted on this page are AAA's, reported by CR, and come from AAA's own service data rather than from a laboratory. The per-battery ratings themselves are behind CR's membership wall and we have not seen them, so no battery is ranked here.

arithmetic The 17-month gap is ours, from AAA’s two figures. Because 41 months is stated as an upper bound — “less than 41” — the real gap is at least that, not exactly that. The month-letter table is generated from the rule rather than typed, so it cannot drift from the skipped-I note beneath it.