The strange scent-detection device demonstrated by the sales team from Cyrano Sciences looked a little like a candy-colored cell phone. But to the nurses at the University of California, Los Angeles Dental School, it seemed as beautiful as a mouthful of perfectly capped teeth.
A program at the Dental School was testing remedies for oral malodor-in other words, chronic bad breath-and the nurses were taking turns monitoring patients’ progress. The methodology was simple: Nurse puts nose to patient’s mouth, then everyone breathes deeply. Professionalism aside, no one involved walked away savoring the experience.
Why not try the Cyranose 2000 instead? According to Steven Sunshine, the earnest, 38-year-old president and CEO of Cyrano, his company’s handheld detector promised not only a welcome reprieve for the nurses but also vastly improved accuracy in halitosis research. A smell, after all, is really just an individual’s subjective experience of airborne chemicals known as odorants. Far better, according to Sunshine, to quantify the aroma in question by relying on the Cyranose 2000’s 32 precise sensors.
As a sales pitch, it was a slam-dunk. Unfortunately for Cyrano Sciences, marketing its new electronic nose to other prospective customers won’t be half as easy. The company, a startup staffed by chemistry PhDs and electrical engineers eager to make their mark in business, faces a customer landscape studded with skeptical purchasing managers, disparate technological requirements and the simple fact that the world has thus far been doing just fine without an electronic nose. From the food industry, to animal husbandry, to contraband law enforcement, the lab-rats-turned-entrepreneurs at Cyrano Sciences must tackle head-on the single most treacherous challenge for many startups after the initial phase of product development: inventing a market.
Conjuring demand out of whole cloth is common currency for pioneers in any industry. Apple Computer stared down the challenge in the early 1980s. Consumer goods titans such as Procter & Gamble and Gillette regularly create demand for newfangled items by virtually inventing a mass culture of desire with hundreds of millions of dollars in annual advertising. Today, life for some would be next to unlivable without triple-bladed razors and 500-megahertz personal computers. Soon, we may not be able to get by without our smell machines. At least, that’s the plan.
With the launch of the $5,000 Cyranose 2000 slated for this fall, Sunshine and company are setting out to sample the waters in two trial markets: detecting “off” odors in packaging materials, and quality control testing of raw materials such as soda concentrates for the food industry. Cyrano will face entrenched competition from both high-end, specialty analytical instruments as well as from what might be termed an installed base of human noses. Specifically, the device will have to out-sniff professional teams of olfactorily acute individuals known as human “sensory panels” employed by consumer goods makers. (Facing a related dilemma, Scott Cook, the co-founder of personal accounting software leader Intuit, once quipped that his toughest competitor was his customers’ stock of pencil and paper.)
The technology behind the Cyranose 2000 is the brainchild of a chemist named Nathan Lewis at the California Institute of Technology, who in 1993 began pondering the workings of the human nose-an organ responsible for what is arguably the most versatile and least understood of the five senses. Lewis, along with then-postdoc Mike Freund, decided to design a synthetic sniffer. The strikingly simple solution they hit upon entails doping ordinary polymer plastics with particles of a conductive material, such as carbon black. Painted onto a ceramic surface, electrons move through such composites at a predictable rate. But expose it to an odorant, and the plastic starts absorbing vapor molecules and swells like a sponge. Because the swelling alters the spacing between the conductive particles, it creates an easily measured change in the composite’s electrical resistance.
As it turns out, every type of plastic has its own unique chemical likes and dislikes. Some readily absorb oily vapors such as benzene, others prefer water. By creating an array of sensors, each from a different plastic, Lewis was soon able to generate a distinct electronic pattern for every odor. The next step was to record the patterns that different smells induced in the sensor array. Once trained, this “nose chip” could recognize fragrances it had been exposed to previously and determine when they changed.
Lewis learned that similar technology had already been invented by scientists in the United Kingdom, who had started their own company, AromaScan. In fact, the Brits were just one among a half-dozen firms that had been attempting to sell artificial probosces. But partly because of their high cost, these instruments had found little success and today the electronic nose market still remains puny at less than $15 million per year.
Lewis believed his approach was more versatile than the British system and, because it was so simple, would also be very cheap to manufacture. After building a prototype system with the help of Caltech chemistry professors Robert Grubbs and electrical engineer Rodney Goodman, the device’s commercial potential began to crystallize. Lewis was issued a patent in 1995, and it wasn’t long before a a physician-turned-venture-capitalist named Seth Harrison began sniffing around the Caltech labs. By April 1997, Cyrano Sciences had been formed with the university’s blessing, a license to the electronic nose patent and Harrison as interim CEO.
Taking over the kind of gray, nondescript warehouse in suburban Pasadena that even longtime neighbors never really notice, the company began hiring a product development team, mostly chemists and other materials scientists. For more than a year and a half, the researchers fabricated arrays of organic polymer sensors, wrote software to recognize the electrical “fingerprints” yielded by individual vapors, and built a dizzying number of plastic-molded prototypes. The engineering team endlessly tested the nose-chip designs against various women’s perfumes. “Our goal, which we eventually accomplished, was to get it down to where it could detect the differences between Chanel No. 5 and Chanel No. 19,” says Greg Steinthal, head of the engineering department.
The quasi-academic environment and dearth of business-types were part of the strategic plan put in place by Cyrano’s venture backers, who have been in control of the company since its inception. “Too many MBAs in an early technology development program might lead you in the right business direction, but they won’t understand the real technology issues. That can throw off the corporate timing,” says Harrison, a general manager at Oak Investment Partners in Westport, Conn., which, along with Johnson & Johnson’s investment arm and Marquette Venture Partners, has so far pumped $12 million into the enterprise.
While basic science and engineering remained the task at hand, the company made an effort to hire people with commercial experience-especially aspiring entrepreneurs languishing in large corporations. Steinthal was brought in from Telaire Systems, where he had already helped push a carbon dioxide sensor to the market. Sunshine was hired as chief of R&D from manufacturing giant Raychem, where he’d led an in-house incubator. According to Sunshine, the common denominator among his colleagues is “knowing when to stop researching.” (Little wonder that Lewis and the other Caltech inventors had chosen to stay put in academia.)
By the beginning of this year, with most of the kinks worked out of the device, Cyrano was ready to make the seismic shift to an enterprise focused on the unscientific domain of salesmanship. Sunshine was promoted from research boss to CEO and began recruiting a marketing staff thick with technical backgrounds.
A key hire was Saskia Feast, a British chemist whom Sunshine gave the title of technical marketing manager and tasked with sizing up potential customers. Up until this point, the nose’s creators and investors had been doing plenty of daydreaming about applications (busting drug dealers, diagnosing cancer) and had come up with a laundry list of some 20 potential markets. But which were most likely to make the jump to the unfamiliar nose-chip technology? Feast has been boning up on a score of industries in order to find out.
She began by identifying 73 large prospective customers, and is now grading them on a scale from “one” to “five,” depending how on difficult the sale looks. The fives are the toughest, Feast says, and include heavily regulated businesses where new technology means risky change, and electronic noses are eyed with suspicion. For instance, in the chemicals industry workers assess accidental spills by breaking a sample down into its chemical subcomponents, then analyzing each part. Cyrano’s argument is that the industry could skip this detailed analysis if it knew roughly what spills contained-something the nose chip could do with one whiff. But the chemicals industry isn’t used to diagnosing blended elements and Feast expects Cyrano’s message will be hard to get through.
For reasons that are hardly mysterious, food companies represent the overwhelming number of zeros and ones in Feast’s database. The industry, she says, “is filled with people very familiar with using smell” to detect spoiled food. Real noses, however, while quite precise, are attached to people who aren’t perfectly engineered for the job: They get cold and tired, and lose their edge after repeated contact with similar odors. But even the food industry is no easy sale: It’s been burned before by firms promising dazzling smell detection and not delivering; Cyrano is pegging its hopes for a marketing knockout on an industry not entirely eager to get back into the ring.
With a large, fragmented market full of companies ready to turn up their noses at newfangled technology, Feast’s strategy is to seed the machine among “opinion influencers” in their industries, such as Kraft Foods and DuPont. The hope is to convert these key customers into nose chip evangelists. The same goes for the company’s academic allies: Cyrano is a member of the Monell Institute’s Chemical Senses Center, as well as the Media Lab at MIT. “The idea that we’re a flash in the pan is something we always have to deal with,” says Bruce Hermann, director of applications engineering, and a force in the company’s marketing efforts.
The search for stability has also led Cyrano into what Hermann terms “a delicate dance” with a predatory competitor. In March, Cyrano signed a pact with Hewlett-Packard to jointly develop new versions of its electronic nose and collaborate on marketing. The scientific instrument giant already sells a vapor detection device of its own, a hulking, refrigerator-sized appliance called the HP 4440A Chemical Sensor, priced at $80,000.
Cyrano brings to HP’s table a technology that might turn out to be a “category killer.” The deal’s potential benefits for Cyrano are enormous, giving the company instant clout with potential customers and bolstering its reach into new sales channels. But HP’s ultimate intention is something of an unknown, and there’s no hiding the risk for Cyrano. Even the folks at HP don’t shy from pointing it out. “Collaborating with HP is a two-edged sword for them,” observes Mary Pat Knauss, future products marketing manager at HP’s Wilmington, Del. facility.
The pact gives Cyrano “a stamp of credibility” but also holds the risk of terminally narrowing the smaller firm’s future. Losing its identity in HP’s embrace could mean becoming just another supplier for a line of odor detectors carrying the HP logo. Sunshine suggests his firm might follow Intel’s lead and settle for a “Cyrano Inside” sticker. He’s only half kidding, because in fact Cyrano does plan to engage in what Sunshine calls “chip-based design,” a.k.a. electronic nose as universal electronic component. Sunshine imagines the nose chips could be integrated into many consumer products, for instance, a microwave oven that knows when food is cooked. Or firefighting equipment that can tell the nature of a blaze-chemical, electrical or arboreal.
Leaving the marketing to HP or another company would also allow Cyrano’s engineers to do what they seem to love best, which is solving problems. Already the hunt for new applications has taken company officials as far afield as a Pennsylvania dairy farm, where the electronic nose holds promise for sensing which cows are in heat and ready for artificial insemination. (Currently, farmers use a bull, often leaving the animal in a state of homicidal frustration.) Cyrano is also studying how well their nose can mimic human sensory values. That is, they’re teaching it to tell bad smells from unacceptably bad ones. An ongoing project with the Los Angeles Municipal Sanitation Department looks to size up the stink from sanitation leaks.
The work of forging new markets for a new product is not, it seems, totally unlike olfaction itself: both are endless, episodic trails of new experiences, not all pleasant. Standing in front of a blank whiteboard in Cyrano Sciences’ expansive warehouse-office, Sunshine talks about how he now spends most of his time-preparing for the product launch and hunting for money. With a “substantial” burn rate (the amount of cash the company inhales each month without earning any revenue) Sunshine is busy raising a second round of financing. He’s confident he will secure it, but the stress of running the company shows in his boyish features. Just then, someone enters the room carrying one of the brightly colored Cyranose 2000 prototypes. Suddenly, Sunshine’s eyes cease blinking and his face takes on the look of someone who can detect the sweet smell of success.
DeepMind’s cofounder: Generative AI is just a phase. What’s next is interactive AI.
“This is a profound moment in the history of technology,” says Mustafa Suleyman.
What to know about this autumn’s covid vaccines
New variants will pose a challenge, but early signs suggest the shots will still boost antibody responses.
Human-plus-AI solutions mitigate security threats
With the right human oversight, emerging technologies like artificial intelligence can help keep business and customer data secure
Next slide, please: A brief history of the corporate presentation
From million-dollar slide shows to Steve Jobs’s introduction of the iPhone, a bit of show business never hurt plain old business.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.