Consumer Products

Apple Batterygate: how the right fix, done in silence, cost $640m

Case file #96·November 17, 2026·6 min read·analysis by Peter Stasko

Case file

  • What happened: From 2017 (and reportedly 2016 for some models), iOS updates silently slowed processors on iPhones with aged or degraded batteries to prevent unexpected shutdowns. Users were never told. Apple confirmed the practice in December 2017 after benchmark evidence circulated.
  • Scale: Millions of devices worldwide; US class action covered iPhone 6 through 7 Plus and similar models.
  • Root cause: A legitimate power-management mitigation for ageing lithium cells, deployed without notification, consent or disclosure — a change-management failure, not an engineering one.
  • The bill: $500m US class settlement (2020), $113m settlement with 34 US states, €25m fine from French consumer authorities (DGCCRF), a year of $29 battery replacements, and lasting trust damage.

Here is an uncomfortable observation from two decades of chasing nonconformities: Batterygate is not a story about bad engineering. The engineering was defensible. It is a story about a company that solved the right problem correctly, then destroyed the value of the solution by refusing to write one sentence in the release notes. I have watched plants spend six figures on corrective actions and then bury the customer-notification step because someone feared the phone call. Apple did the same thing at planetary scale. The price tag is public record.

$500mUS class-action settlement, 2020
$113mSettlement with 34 US states
€25mFrench DGCCRF fine for non-disclosure

The situation

Lithium-ion cells age. Internal resistance climbs, the ability to deliver peak current sags. When an ageing iPhone hit a heavy load spike, voltage could dip below the threshold needed to hold the processor — and the phone would die without warning, often at 30 or 40 percent indicated charge. Users were furious about the shutdowns, and rightly so. That was a genuine field-failure problem.

Apple's answer was a power-management feature in iOS that dynamically throttled the processor on devices whose batteries could no longer supply peak power. Spikes flattened, shutdowns stopped, the phone stayed alive. On the merits, this is exactly the mitigation I would applaud on a shop floor: find the failure mode, characterise the mechanism, manage the symptom at the source. The fatal decision came later. Deploy it silently — no release-note mention, no user setting, no battery-health indicator, nothing telling the customer their £1,000 device had quietly become 30 percent slower than the day they bought it.

How it unfolded

Users noticed a pattern before anyone noticed an event. Third-party benchmarks on older iPhones showed suspicious, stepwise performance drops after iOS updates, while Apple's own diagnostics reported nothing unusual — because the throttling was intentional. In December 2017, John Poole's public benchmark analysis laid out the evidence, Reddit threads caught fire, and Apple was forced into a statement confirming what it had denied by omission for roughly a year.

Then the cascade: an apology, replacement batteries cut from $79 to $29 for a year, a promised battery-health feature in iOS, lawsuits consolidated into class actions, a multi-state investigation in the US, a French probe under consumer-transparency law. The settlements — $500m in the US class action (2020), $113m with 34 states, €25m in France — did not award damages for broken phones. The phones weren't broken. They priced the concealment.

Root-cause anatomy

Technically, nothing to prosecute. An ageing-cell voltage-sag failure mode, correctly characterised and genuinely mitigated by dynamic throttling. No design defect in the fix itself. Add the later battery-health indicator and the cheap battery programme and you have a complete, working solution set.

Organisationally, the root cause is cleaner and uglier. Somewhere, a cross-functional decision was made that altering delivered hardware performance on millions of fielded devices did not require telling the owners. That is not a bug in the process — it is a missing gate. Nobody asked the change-control question: does this change affect the customer's perception of the product they bought? Give that question a formal owner — a role, a signature, a gate — and silence becomes impossible. Leave it with nobody, and silence is the default. Once the first silent release shipped, every subsequent one compounded the exposure.

A corrective action concealed is a liability with interest.

Where the quality system failed

In my world this would be classified in seconds. Under IATF 16949 change management and any credible APQP framework, software that measurably alters the performance of delivered hardware is a customer-notifiable change. Full stop. You notify, you get concurrence, you document it. Skipping that gate is not a judgement call; it is a nonconformity.

Run the PFMEA logic and the gap gets precise. The first failure mode — sudden shutdown on load spikes — was identified, severity-scored, mitigated. Textbook. What nobody scored was the second failure mode, the one the mitigation itself introduced: undisclosed performance degradation, high occurrence across the installed base, high detection by customers with benchmark apps, severity measured in trust and litigation. That line never made the sheet because the analysis, if one existed, stopped at the hardware boundary. And the cost comparison is absurd. Disclosure cost roughly one sentence in release notes. Concealment cost over $638m in settlements and fines, plus a year of subsidised $29 batteries, plus regulatory scrutiny that never fully closed. The most expensive unwritten sentence in consumer-electronics history.

What would have caught it

Three gates, all cheap. First, a customer-notifiable change gate: any software release altering measurable hardware performance on fielded units requires documented customer notification before release. No gate, no ship. Second, an extended PFMEA discipline — score your mitigations as potential failure modes, because every fix changes the product and every change deserves its own FMEA line.

Third, red-team the silence. Before release, ask one adversarial question: if a customer with a benchmark app discovers this, what is our position? If the honest answer is "we'd rather they didn't", the release isn't ready. And note what the eventual fix looked like once Apple stopped hiding — apology, $29 batteries, battery-health transparency. A textbook 8D corrective-action package. It arrived a year late because containment was first attempted through silence, which is not containment. It is escalation on a timer.

My take

This case is painfully familiar. In automotive quality — at Witte Automotive, at SNOP, where I built a greenfield QA function for 900-plus people — I lived the same tension from the other side. Every time we found a field issue, the pressure to "manage it quietly" was real: quiet containment, quiet rework, no customer letter. I pushed back every time. Not out of virtue, out of arithmetic. The disclosure conversation, handled early and factually, costs a difficult week. The concealed version costs a quarter of escalation, a formal complaint, sometimes a claim — and I have delivered de-escalation of exactly those situations, where zero critical escalations in a quarter came from telling customers the truth before they found out themselves. At SNOP we cut defect costs by 70 percent, and a meaningful slice of that came from notification discipline, not from better parts. Apple's engineers and my plant teams shared the same instinct: protect the product. What separates the outcomes is not engineering. It is having a gate that makes silence impossible, and someone with the authority to stop the release when the gate fails. Apple, one of the most process-mature companies on earth, lacked that one gate — or lacked the person willing to close it.

What this means on your floor

  • Any change that alters what the customer receives or perceives — hardware, software, delivery, performance — passes a formal notification gate. No exceptions for "internal" fixes.
  • Score your mitigations in the PFMEA as new failure modes. The cure can be the next defect.
  • Compare the cost of disclosure to the cost of discovery. It is almost never close, and discovery is usually the customer's.
  • Containment by silence is not containment; it is escalation on a timer.

The engineering in Batterygate saved phones from shutting down. The process around it cost more than half a billion dollars, a fine in Europe, and a permanent asterisk on one of the strongest brands in history. That asymmetry is the whole lesson. The fix was never the problem. The unwritten sentence was. Build your quality system so the sentence always gets written — before someone else writes it for you in a subpoena.

This case file analyses publicly documented events and reports. I had no involvement in the engagements described; company statements and official findings are matters of public record. The lessons and opinions are my own.

Peter Stasko

Peter Stasko

Corporate operator across automotive and aerospace — Airbus, SNOP and Witte Automotive. Building production AI hands-on since 2016.

Seeing the same failure mode on your floor?

Every case in this library ended as a headline. A one-hour conversation is cheaper. No slides, no pitch — just an experienced pair of eyes on your situation.

Book a conversation