Breathometer is one of the few Shark Tank deals where every Shark said yes. Charles Michael Yim pitched a smartphone breathalyzer in 2013 asking for $250,000 for 10% of the company. What he got instead was a bidding war that ended with all five Sharks combining for $1 million at a 30% stake — Mark Cuban alone in for $500,000 at 15%, the other four splitting the remaining $500,000 and 15%. It remains one of the most dramatic moments in the show’s history.
It is also, by Cuban’s own account, his worst investment. Within a few years, the Federal Trade Commission had opened an enforcement action against Breathometer over the accuracy of its devices, ordered refunds for customers, and permanently barred the company from re-enabling the breathalyzer function in its app. The FTC’s complaint centered on a specific, damning finding: Breathometer’s Breeze device developed a “downward drift” in early testing — meaning it could report a lower blood alcohol concentration than a user actually had — and the company kept marketing it as “law-enforcement grade” and “government-lab grade” anyway. Total sales across its two breathalyzer products reached $5.1 million before the FTC stepped in.
That is the detail that makes Breathometer worth studying as a case, not just a cautionary headline. This was not a company that ran out of money, failed to find customers, or got out-executed by a competitor. It had funding, retail ambition, a second product line, and a founder telling TechCrunch afterward that a “bad batch” from a poor manufacturer was to blame. What it didn’t have was a breathalyzer that reliably did the one thing a breathalyzer has to do.
What actually went wrong
Breathometer’s core proposition was genuinely appealing: an inexpensive, portable way to check your blood alcohol level before deciding whether to drive. The company had already proven demand before the show, with roughly $140,000 in sales and a live Indiegogo campaign. After the episode aired, orders reportedly hit $1 million within three months. By 2014, the company said it was producing around 15,000 units a week, up from roughly 1,000 before.
The accuracy problem surfaced in late 2014, after that growth was already underway. The company’s own internal testing found the drift issue; further testing in early 2015 found that temperature and humidity affected readings, and that the Breeze’s sensors degraded meaningfully over time. Rather than pulling the product or softening the marketing claims while the issue was resolved, Breathometer kept selling it as a safety tool consumers could rely on to decide whether they were fit to drive. That gap — between what the marketing promised and what internal testing already knew — is what the FTC’s January 2017 settlement was built around.
Lessons for founders
Never scale a safety claim you haven’t independently validated. A consumer gadget that’s occasionally wrong is an annoyance. A device marketed specifically to help someone decide whether they’re safe to drive cannot be occasionally wrong — the entire value proposition depends on the claim being true every time, not most of the time. Breathometer scaled production and marketing spend on the strength of a claim its own later testing didn’t support.
Regulatory compliance is not a later-stage problem you can bolt on after growth. By the time the FTC opened its investigation, Breathometer had already sold $5.1 million worth of product built on claims regulators would later say lacked adequate substantiation. The cost of getting compliance wrong scales with how much you’ve already sold — which means the riskiest time to under-invest in it is exactly when you’re growing fastest.
Investor enthusiasm is not product validation. Five Sharks saying yes on television is a remarkable fundraising outcome. It is not evidence that the underlying technology works as advertised. Founders in safety- or health-adjacent categories especially need an independent, adversarial testing process — one whose job is to try to break the accuracy claim, not confirm it — before scaling past the point where a recall becomes expensive and reputation-damaging.
A founder’s public story and a regulator’s findings can both be true and still tell different stories. Yim’s manufacturing explanation to TechCrunch and the FTC’s broader complaint about inadequate substantiation aren’t necessarily contradictory — a bad manufacturing batch can explain a specific failure without excusing marketing claims that were never adequately tested in the first place. Founders navigating a product crisis should be honest about both layers, not just the one that’s easier to explain.
Lessons for investors
Test the technology yourself, or fund someone who will. A breathalyzer’s accuracy is a testable, falsifiable claim — not a matter of founder charisma or market size. Due diligence on a hardware safety product should include independent accuracy testing before capital moves, not after a regulator’s complaint reveals what internal testing already knew.
Map the regulatory exposure before the pitch, not after the settlement. A product that makes health or safety claims to consumers sits inside a regulatory perimeter the moment it starts selling. Understanding what the FTC, FDA, or equivalent body could plausibly ask for — and whether the company can produce it — is as much a part of diligence as the cap table.
An impressive pitch and a validated product are different things, and the gap between them is where the risk lives. Breathometer’s pitch was compelling enough to get five competing offers in one segment. That competitive dynamic — Sharks bidding against each other in real time — can itself create pressure to move faster than careful diligence normally allows. The lesson isn’t that group deals are inherently risky; it’s that speed and competition are exactly the conditions under which diligence most needs to be deliberately protected, not relaxed.
Know what you’re actually exposed to, and be precise about it. Even the size of Cuban’s own loss became a matter of dispute in later reporting — some accounts cite his original $500,000 stake, while Cuban himself has at different times described the loss as $1 million. That ambiguity, coming from the investor himself, is a reminder that “what did this actually cost me” should be a number every investor can state precisely, not reconstruct after the fact from memory or press interviews.
The all-five-Sharks moment deserves a second look
It’s worth pausing on why the bidding war happened at all, because it explains a dynamic that shows up in plenty of deals beyond this one. Breathometer arrived with a genuinely rare combination for a Shark Tank pitch: a working hardware product, real pre-show sales, and a problem — “am I safe to drive” — that every viewer intuitively understood without explanation. That combination is exactly what triggers competitive bidding among investors who don’t want to lose a deal to each other. Cuban’s initial $500,000-for-20% offer, quickly followed by counter-offers from O’Leary, Herjavec, Greiner and John, is the visible mechanics of FOMO operating among five people who each do this for a living.
That competitive dynamic is not inherently a problem. But it does compress the amount of time available for the kind of skeptical, adversarial questioning that might have surfaced the accuracy issue earlier — nobody wants to be the Shark asking hard technical questions while four competitors are simultaneously trying to close the same deal. The lesson generalizes well beyond breathalyzers: the deals that move fastest, with the most competitive energy in the room, are often the ones where the normal pace of diligence gets compressed the most.
The settlement terms tell their own story
The FTC’s 2017 order against Breathometer wasn’t just a refund requirement — it included a permanent injunction against re-enabling the breathalyzer function, which is a more severe remedy than a typical consumer-protection settlement. Most FTC actions over unsubstantiated advertising claims end with a company being allowed to keep selling the product once its claims are corrected or better substantiated. A permanent functional ban signals that regulators concluded the underlying technology, not just the marketing language around it, couldn’t be trusted for the use case it was sold for. That distinction — a marketing problem versus a technology problem — is the difference between a settlement a company can rebuild from and one that ends the product line entirely.
The real takeaway
Breathometer is a stranger cautionary tale than most Shark Tank failures because almost everything about the business worked. The pitch worked. The fundraising worked spectacularly. The manufacturing scaled. Retail interest was real, and the company was already planning a pivot into broader health technology with a partnership pipeline that included Philips and a discussed 50,000-unit initial order for a new oral-health product called Mint. The one thing that didn’t work was the specific, falsifiable claim the entire business was built on — and no amount of funding, exposure, or founder ambition can substitute for that being true.
That pivot attempt is worth noting on its own, because it’s a pattern that shows up whenever a company’s original product runs into an existential problem: the instinct to reposition around a broader platform vision, rather than fix the narrower thing that’s actually broken. Reframing a breathalyzer company as a “breath-analysis health platform” was, in one sense, a reasonable response to a genuine underlying capability — the same sensor technology plausibly could measure other things. But it also meant the company was trying to build credibility for a new set of health claims at exactly the moment regulators were finding its existing claims hadn’t been adequately substantiated in the first place. A trust problem in one product line doesn’t stay contained to that product line once it becomes public.
