Arrow
Get your free stagnation score
Arrow
Find Your Solution
Logo

Samsung Fixed the Note 7 Before It Knew What Was Wrong

by | Sep 15, 2026

FICTION · An Unfair Advantage Intervention

Fiction. Real history. Sourced below.

Finding out what actually caused the Note 7 fires took Samsung several months, 700 engineers, more than 200,000 devices, over 30,000 batteries, and three independent laboratories. In September 2016, facing the first fires, Samsung did that same work in a matter of days, concluded one supplier’s cells were to blame, and swapped the entire production line. The replacement batteries had a completely different defect. The company didn’t move too fast. It moved fast in the wrong order.

I’m Maria Gonzalez. I came up through a shop floor and then through Six Sigma, so my instinct on any failure is to ask what the diagnosis was and how long it took, before I ask anything about the fix. On the Note 7 those two questions have startling answers, and the gap between them is the whole case. This is also a story where people were burned and property destroyed, so I want to be careful with it: the engineering matters because the harm was real, not the other way round.

Full fiction disclosure and sourcing note

This is a work of fiction. Maria Gonzalez is a fictional character from The Unfair Advantage by Todd Hagopian, and Cartwell is the book’s fictional company. Maria analyzes the real, documented history of the Galaxy Note 7 from the present day; she never worked at Samsung or any supplier, never met or spoke with any real person, and no dialogue is invented for anyone. Technical findings are drawn from Samsung’s own published investigation and the independent laboratories that participated in it. The historical facts are accurate and sourced as of 2026; the framework analysis is invented.

Two Diagnoses, Months Apart

Set the two investigations side by side, because the comparison is the finding.

The September 2016 diagnosis. Days after fires began appearing, Samsung concluded that batteries from one supplier were at fault, recalled roughly 2.5 million handsets across ten markets, and shipped replacements built with cells from a different supplier. The first recall targeted the first supplier’s cells exclusively.

The January 2017 diagnosis. Samsung assigned about 700 engineers to test more than 200,000 fully-assembled devices and over 30,000 batteries, and engaged three independent organizations — UL, Exponent, and TÜV Rheinland — to run their own investigations. It took several months and produced two distinct root causes in two different batteries.

Same company. Same technical capability. The organization that could produce the second diagnosis existed in September; it simply wasn’t asked to. And nothing in the first diagnosis could have predicted the second failure, because the second battery’s defect was unrelated to the first’s in mechanism, in supplier, and in kind.

That is not a story about a company that rushed. It is a story about a company that performed the third step of a three-step process and skipped the first two, at the moment when skipping them was most dangerous.

The 3-A Method: Samsung Ran Only the Third A

The 3-A Method structures improvement work in six-week cycles: two weeks to Apprehend, two to Analyze, two to Activate. It is a speed framework — it exists to compress traditional four-to-six-month project cycles — and the speed comes precisely from front-loading diagnosis rather than from shortening it. Four of the six weeks are spent understanding the problem before anything changes.

That proportion is the whole discipline, and it is the thing organizations under pressure invert. Apprehend means establishing what is actually happening — gathering the failure data, characterizing the defect, defining the boundaries of the problem. Analyze means determining why, to the level of a mechanism you can point at. Only then does Activate change anything.

Samsung’s September response was Activate with essentially no Apprehend and no Analyze. It reached a conclusion — one supplier’s cells — that was a hypothesis consistent with the early evidence, and treated it as a finding. Then it acted on that hypothesis at the scale of an entire global production line.

And here is the trap that makes this worth studying rather than merely criticizing: the hypothesis was not wrong. Battery A genuinely was defective. Samsung correctly identified a real cause. What it had not done was establish that this was the *only* cause, or that the replacement path introduced no new one — and neither of those questions can be answered by the work that identifies the first defect. They require the Analyze step, which is the step that got dropped.

Note what this does to the usual moral. “Don’t rush your fix” is close, but it is not quite it, because in a safety crisis you genuinely should move quickly to get dangerous devices out of people’s pockets. The recall itself was right and urgent. What could not be compressed was the diagnosis behind the replacement. Speed in the Activate step is a virtue; speed achieved by deleting Apprehend and Analyze is not speed at all, as the second recall demonstrated.

The Supplier Swap Was a Type 1 Decision

The 70% Rule says decide at seventy percent information and confidence — but the Decision Type Matrix qualifies it. Type 1 decisions, irreversible and critical, require eighty-five to ninety percent. Type 2, reversible and critical, is where seventy is right. Type 3 gets seventy with exit strategies. Type 4, reversible and non-critical, needs fifty.

The recall was Type 2 and appropriately fast: stopping sales and retrieving devices is reversible, urgent, and correct at high speed. It should not have waited for certainty.

The supplier swap was Type 1 and was treated as Type 2. Shipping replacement units to customers who had just been told their phones were dangerous is effectively irreversible — you get exactly one chance to tell someone the device is now safe. It was critical by every measure, involving physical harm. And it was decided on days of investigation into a mechanism that would eventually require months and three laboratories to characterize.

The Three-Question Test makes the shortfall concrete. Do I understand the key risks? Not the risks introduced by the new cell, which had not been evaluated at the depth the situation required. Can I explain it clearly? Only the first defect. Do I have a reasonable hypothesis? Yes — and a reasonable hypothesis is a Type 2 threshold, not a Type 1 one.

What Actually Failed, Physically

I want to be precise about causation, because the popular telling substitutes a plausible story for the finding. The widely repeated explanation is that Samsung rushed to beat Apple’s autumn launch and the rush caused the fires. Contemporary reporting did document real pressure on suppliers and on engineers working within very tight physical limits, so it is a credible account of conditions. But it is not what the investigation concluded. Samsung attributed the fires to two specific battery defects, and the schedule remains context rather than finding.

Battery A, used in the original phones and made by Samsung SDI — a Samsung Group affiliate — had a design problem. The negative electrode carried a slight deflection with an incorrectly located tip, in a casing too small for the cell, allowing the positive and negative electrodes to meet and short internally.

Battery B, used in the replacements and made by Amperex Technology, had a manufacturing problem. Abnormally high ultrasonic welding burrs on the positive electrode scraped against the insulation as the cell expanded and contracted through normal charge cycles, eventually penetrating the separator; some cells were also missing insulation tape entirely. Exponent’s analysis pointed to welding defects and missing protective tape over the positive electrode tab; UL’s cited welding defects along with poor alignment and inconsistent insulation tape.

Read those two descriptions and the framework point becomes unavoidable. A deflected negative electrode in an undersized casing and a welding burr piercing a separator are not variations of one problem. No amount of investigating the first tells you anything about the second. They share only a symptom, and a symptom is what Apprehend gathers — not what Analyze delivers.

What Actually Happened

Samsung launched the Galaxy Note 7 in August 2016, and within days reports emerged of handsets overheating and catching fire. People were burned and property was damaged, including vehicles. In early September Samsung halted sales and recalled roughly 2.5 million handsets across ten markets, attributing the problem to one supplier’s cells and issuing replacements built with a different supplier’s batteries. Reports of replacement devices catching fire followed, leading Samsung to permanently discontinue the line on October 11, 2016. Airlines and carriers banned the device.

Samsung guided to an operating loss of over $5 billion spread across the final two quarters of 2016 and the first quarter of 2017, with total crisis costs commonly estimated around $5.3 billion.

On January 23, 2017 in Seoul, DJ Koh, president of Samsung’s mobile communications business, presented the investigation’s conclusions alongside executives from UL, Exponent, and TÜV Rheinland. Samsung did not name its suppliers, identifying them only as Company A and Company B; press reporting identified them as Samsung SDI and Amperex Technology.

Samsung responded with an 8-Point Battery Safety Check — extreme testing inside and out, followed by X-ray and visual inspection — along with Multi-layer Safety Measures and a Battery Advisory Group of external academic experts including materials scientists from Cambridge, Berkeley, and Stanford. The overwhelming majority of devices were ultimately recovered, and the following year’s Note 8 was well received.

From the Operator’s Chair

I’m Todd Hagopian, and product quality, schedule pressure, and the discipline of protecting engineering constraints are things I’ve worked with directly across Berkshire Hathaway, Illinois Tool Works, Whirlpool, and now at JBT Marel. Three disciplines from this one.

First, separate the containment decision from the remedy decision, and give them different thresholds. This is the single most useful thing in the case. Stopping sales and retrieving product should be fast and can be decided on partial information, because it is reversible and every hour costs exposure. Telling customers the replacement is safe is irreversible, and it needs the higher bar. Organizations in crisis tend to run both at the speed of the first, which is how a correct recall becomes a second incident.

Second, when a fix is proposed, ask what diagnosis it rests on and how long that diagnosis took. Samsung’s September answer took days; the real answer took months, 700 engineers, and three laboratories. That gap was visible at the time to anyone who asked the question. In my experience the fastest way to catch this is to require whoever proposes a remedy to state the mechanism — not the correlation, the physical mechanism — and to say what would have to be true for a different mechanism to be at work. If nobody can answer the second part, you are in Apprehend, not Activate.

Third, protect design margin from every priority, not just the calendar. A cell in a casing too small to accommodate normal swelling is a specification decision as much as a schedule one. Margin is not slack; it is the buffer that absorbs the ordinary manufacturing variation that always exists. When engineers ask for it, that request is the cheapest insurance available, and it is usually declined by people who will never see the failure mode it prevents. Timing genuinely matters in competitive markets — I’d resist the cynical reading that the launch race was vanity. It simply never outranks the physics, and physics does not read the calendar or the recall schedule.

The Monday Morning Playbook

  • Give containment and remedy different confidence thresholds. Stopping and retrieving is reversible — move fast on partial information. Telling customers it’s fixed is irreversible — hold it to the higher bar.
  • Ask for the mechanism, not the correlation. Before approving any fix, require a stated physical mechanism and an answer to “what would have to be true for a different cause to be at work.” No answer means you’re still in diagnosis.
  • Check the ratio of diagnosis to action. If almost all the elapsed time in a response went to acting, the diagnosis was assumed. Four of six weeks belong to understanding, and that proportion is what makes the last two fast.
  • Treat a correct cause as possibly incomplete. Samsung’s first hypothesis was right and insufficient. Finding a real cause does not establish it’s the only one, and the work that identifies the first can’t rule out a second.
  • Defend margin from the spec sheet as well as the calendar. Ambition consumes safety buffer as readily as deadlines do. Name what margin is non-negotiable before either starts borrowing.

The Contrarian Reappraisal

The consensus reading — Samsung rushed to beat Apple and the rush caused the fires — is plausible, partly supported, and imprecise in three ways worth correcting.

The verified cause was two specific defects, not the schedule. Samsung’s investigation, with UL, Exponent, and TÜV Rheinland participating, attributed the fires to a design flaw in Battery A — a deflected negative electrode with a misplaced tip in an undersized casing — and a manufacturing flaw in Battery B, abnormal welding burrs piercing insulation, with some cells missing insulation tape. Schedule pressure is a credible account of why those defects slipped through. It is not the finding, and treating an interpretation as the cause produces a vaguer lesson than the record supports.

The failure was as much sequence as speed. The recall was fast and correct. The supplier swap was fast and unsupported by diagnosis, resting on days of work where the real answer required months, 700 engineers, and 200,000 devices. Samsung didn’t skip caution; it ran the third step of a three-step process first, and the second fire is what that produces.

Margin was consumed by ambition, not only by the calendar. A high-capacity cell in a casing with insufficient room to swell is a specification choice. “Don’t rush” is too narrow a lesson; the rule is that safety margin has to be defended from every priority that wants to borrow from it, whether that’s the launch date or the spec sheet.

The Note 7 Timeline

  • August 2016 — Samsung launches the Galaxy Note 7 ahead of Apple’s expected autumn iPhone; within days, reports of overheating and fires begin.
  • Early September 2016 — Samsung halts sales and recalls roughly 2.5 million handsets across ten markets, attributing the fault to one supplier’s cells and shipping replacements with a different supplier’s batteries.
  • Early October 2016 — replacement devices begin catching fire, for an unrelated manufacturing reason; a second recall follows.
  • October 11, 2016 — Samsung permanently discontinues the Note 7; airlines and carriers ban the device from flights.
  • Late 2016 — Samsung guides to an operating loss of over $5 billion spread across three quarters.
  • January 23, 2017 — after months of work by roughly 700 engineers across 200,000 devices and 30,000 batteries, Samsung presents two distinct battery defects as the cause, alongside UL, Exponent, and TÜV Rheinland.
  • 2017 — Samsung introduces the 8-Point Battery Safety Check, Multi-layer Safety Measures, and an external Battery Advisory Group; the Note 8 is well received.

By the Numbers

  • Days vs months — the September diagnosis behind the supplier swap against the investigation that eventually found the real causes: the sequence failure, measured.
  • 700 engineers / 200,000 devices / 30,000 batteries — the scale of the work required to characterize the defects, plus three independent laboratories.
  • 2 unrelated defects — a deflected negative electrode in an undersized casing, and welding burrs piercing insulation: nothing about diagnosing the first predicts the second.
  • 2.5 million handsets, 10 markets — the scale of the first recall, issued on a hypothesis treated as a finding.
  • Over $5 billion — the operating loss Samsung guided to, spread across the last two quarters of 2016 and the first of 2017.
  • ~2 months — from August launch to October 11 discontinuation.

Frequently Asked Questions

Why did the Samsung Galaxy Note 7 catch fire?

Samsung’s investigation, with UL, Exponent, and TÜV Rheinland participating, found two separate battery defects. Battery A, in the original phones, had a design flaw — a deflected negative electrode with an incorrectly located tip in a casing too small for the cell, letting the electrodes meet. Battery B, in the replacements, had a manufacturing flaw: abnormal ultrasonic welding burrs that scraped through insulation, with some cells missing insulation tape. Both caused internal short circuits.

Why did the replacement Note 7 phones also catch fire?

Because the replacement cells had an unrelated defect. Samsung’s September diagnosis concluded one supplier’s batteries were at fault and swapped the production line to another supplier, on days of investigation. The eventual root-cause work required several months, roughly 700 engineers, over 200,000 devices, and 30,000 batteries. Nothing in the first diagnosis could have revealed the second defect, because the two shared only a symptom.

Did rushing to beat Apple cause the Note 7 fires?

That’s the popular interpretation and contemporary reporting documented genuine pressure on suppliers and engineers. But Samsung’s official conclusion was the two battery defects, not the schedule. The honest version: the verified cause was physical, the schedule is credible context for how the defects slipped through, and part of the cause was specification ambition — a high-capacity cell in a casing with too little room to swell safely.

How much did the Note 7 crisis cost Samsung?

Samsung guided to an operating loss of over $5 billion spread across the final two quarters of 2016 and the first quarter of 2017, with total costs commonly estimated near $5.3 billion. The device was banned from aircraft worldwide and the brand carried the association for years, though Samsung recovered the overwhelming majority of devices and the following year’s Note 8 was well received.

What’s the real lesson of the Note 7?

That containment and remedy need different confidence thresholds. Halting sales and retrieving devices is reversible and should be fast. Telling customers a replacement is safe is irreversible and requires a diagnosis you can defend at the level of physical mechanism. Samsung didn’t move too fast; it ran the action step of a three-step process without the two diagnostic steps that make the action reliable.

About the Stagnation Assassin

Todd Hagopian is a Fortune 500 transformation executive who has generated $3B+ in shareholder value across Berkshire Hathaway, Illinois Tool Works, Whirlpool, and JBT Marel, where he serves as VP of Product Strategy and Marketing. Known as The Stagnation Assassin, he turns proprietary frameworks — including the 3-A Method and the 70% Rule’s Decision Type Matrix from the HOT System — into field-tested tools for operators who have to move fast without skipping the diagnosis that makes speed safe. He is the author of The Unfair Advantage and the Stagnation Assassin series, a Forbes contributor, and an award-winning keynote speaker. Connect with Todd on LinkedIn or bring him to your stage via his speaking page.

The answer took months, 700 engineers, and three laboratories. The fix was decided in days on the assumption that the first cause was the only cause. Move fast on containment — that’s reversible. Hold the remedy to the higher bar, because you only get one chance to tell someone their device is safe now.

The discipline of moving fast without skipping the diagnosis is in The Unfair Advantage.

Connect on LinkedIn · Book a keynote.