Ask ten product managers what they do, and you will get ten different answers, which will likely leave you feeling confused about the nature of the profession as a whole. Ask about responsibilities and you find that some product managers oversee multiple product lines, some manage platforms, some oversee feature sets, some even do what the job title suggests and manage an individual product. Ask about organizational structure and you discover that some product managers work in engineering, some work in marketing, some work in analytics, and some, remarkably, work in a department called “product management”.
It’s a bit of a mess, which, on some level, is to be expected given all the different types of products and org charts out there. However, it is difficult to write a guide to product management without first articulating what a product manager is, so let’s stick to a very simple definition: a product manager is someone who creates new things to sell. The things can be anything with commercial value, although product managers are most frequently associated with pieces of hardware or software. In creating these products, product managers generally work through four stages: ideating, designing, building, and launching. Since those four stages sound a bit like a table of contents to a white paper, let’s go ahead and use them as such.
Ideation — Contrarian Advice: Ignore the Experts
Imagine that you are an investor and you must back one of two product managers to develop a machine that will change the nature of travel as we know it. You have two candidates for the job. The first candidate, Sam, is a professor, an astrophysicist, and an executive of the Smithsonian. The second candidate, Will, is a high school dropout who runs a bicycle shop. Who would you choose?
If you said Sam, then conventional wisdom in the form of the United States government agrees with you. Sam, better known as Samuel Pierpont Langley, is the man the War Department and the Smithsonian backed to build the first airplane to the tune of more than two million in today’s dollars. Langley brought on a renowned engineer named Charles Manly. The two experts focused on the component in which they had the greatest expertise: propulsion. They built a massive engine and inserted it into a prototype that they called the “aerodrome”. On October 7, 1903, to great pomp and fanfare, Charles Manly climbed into the aerodrome, got shot from a massive catapult, and then proceeded to promptly crash into the Potomac. On December 8, 1903, a second attempt to launch the aerodrome failed just as spectacularly.
Meanwhile, less than two weeks later, in the hills of North Carolina and with no pomp or fanfare, Will made the first successful flight with an airplane that had solved not only the problem of propulsion but also of lift and balance. Will, better known as Wilbur Wright, did not hire a distinguished engineer to help him but instead worked with his brother Orville, a fellow high school dropout with whom he co-owned the bike shop.
Why did the dilettante Wright brothers succeed where the experts Langley and Manly failed? One way to think about it is that experts are professionals who have an in-depth understanding of the current state and who derive their income from explaining that current state to non-experts. As a result, experts often have the hardest time seeing how the current state can be radically modified. By spending so much time within a certain model of the world they have a hard time seeing beyond the confines of that model. They also have the most to lose by getting it wrong, so they tend to stay on the conservative side of problem-solving.
Five decades later airplane technology had reached a new zenith and the Soviet Union was working on building its first supersonic bomber. The problem was that every design resulted in a loss of longitudinal stability at supersonic speeds. This challenge seemed unsurmountable until a young engineer proposed that fuel be rapidly pumped from forward tanks to rear tanks located in the extreme tail during transonic acceleration, thereby dynamically shifting the jet’s center of gravity and providing stability at supersonic speeds. At first, the response was that the engineer was young, and his proposal ludicrous. Ultimately, however, it was his proposal that made the M-50 possible and, ten years later, made the Concorde possible. I know this story well because this young engineer was my grandfather. I suppose contrarian thinking runs in the family. So, when you want to know how an industry or product works, ask an expert. When you want to know how to build a new product to address an unmet need, asking an expert may be useful but it can also send you down the wrong path. And if you are the expert, do not hesitate to solicit radical opinions from young professionals whose perspectives have not yet had time to ossify.
Design — Contrarian Advice: Don’t Ask the Users
In the 1980s the Coca-Cola Company discovered that it had lost significant market share to Pepsi, in part due to two brilliant marketing campaigns. The first campaign was known as the “Pepsi Challenge” where consumers were asked to try both Pepsi and Coke but not told which was which. In this blind taste test, a majority of consumers selected Pepsi as the tastier option. The second campaign was known as the “Pepsi Generation” and presented Pepsi as not just a beverage, but a lifestyle choice. It used celebrities to position Pepsi as the hip option for teens and young adults.
Coca-Cola decided that it would ask its customers what they wanted. The answer came back in very clear and uncertain terms: more sugar. In fact, a key reason why people preferred Pepsi over Coke in blind tests was that it was sweeter and, as we know, consuming something more sweet next to something less sweet always makes the more sweet thing seem, for lack of a better word, sweeter.
Based on customer feedback, Coca-Cola developed a new product formulation which became colloquially known as New Coke. It was definitely sweeter and it was preferred over both original Coke and Pepsi in tens of thousands of blind tests. On April 23, 1985, Coca-Cola launched New Coke with great confidence regarding its success and its ability to retake market share from Pepsi.
Less than three months later, Coca-Cola was forced to return its old formula to the shelves. Customers had roundly rejected New Coke. By the end of 1985, the old formula was making up the bulk of Coca-Cola’s sales and New Coke made up a tiny portion of the market. By all measures, New Coke was one of the biggest product launch disasters in history. What happened?
Coca-Cola asked customers what they wanted and then tried to build a product directly based on that information. The challenge with this approach is that most customers are not sufficiently vested in the design process to take a long-term view and conduct a comprehensive analysis. Rather, they simply taste two things, pick the one that is sweeter, and move on. They don’t stop to consider whether the test is a true reflection of the way they consume the product in daily life, or whether consuming this product in the first place is a taste decision or a lifestyle decision, or whether a fundamental change to the product represents a betrayal of the brand’s core values, or any of the other factors that operate at a less conscious level but materially contribute to purchasing and usage patterns.
One of the brilliant marketers that caused Coca-Cola to feel pressured into releasing a new product was John Sculley. A key architect of the Pepsi Challenge, Sculley had gone from an innovative marketer in the 1970s to a politically savvy executive in the 1980s as he climbed the ranks at Pepsi. In 1985, as Coca-Cola was releasing New Coke, Sculley was in the middle of a power struggle at Apple with its visionary leader, Steve Jobs. By late 1985, both New Coke on the shelves and Steve Jobs at Apple were done.
The reasoning for both New Coke and Jobs’ ouster was based on the same principle: letting the customers design the products. Jobs famously said: “a lot of times, people don’t know what they want until you show it to them.” Of course, this type of thinking requires a willingness to take risks. If you go out on a limb and design something different from what the customers have said they want, and this design fails to sell, then your product management credentials are tarnished and you may be out of a job. On the other hand, nobody at Coca-Cola was fired over New Coke.
In truth, user feedback matters. It is valuable context; but it is not, and should never be considered a substitute for, design thinking. There are many solid product managers who ask users what they want and then give it to them. However, these are not the product managers who shape an industry, these are not the visionaries who are hailed as the next Steve Jobs, and these are not the leaders who change the world. In the words of Henry Ford: “if I had asked people what they wanted, they would have said faster horses.”
Build — Contrarian Advice: Give the Engineers Extra Time
In 2011 a new start-up secured $41 million in funding before they had a proper pitch deck, let alone a working product. According to one of the founders, when the venture capital firm Sequoia asked, “if you had more capital, could you get to the future faster? … Will $25 million help you get five years into one?”, the founding team was “enthusiastic out of our mind to say yes.” Perhaps they should have been a tad less enthusiastic.
The start-up was called Color Labs and it had secured funding for a photo-sharing app on the basis of a prototype and two co-founders with recent successes to their names: Bill Nguyen had sold lala.com, a music website, to Apple in 2009 and Peter Pham had grown BillShrink, a consumer products website, to more than a million users that same year. At about the same time that Color Labs secured $41 million, another start-up was also going after the photo-sharing space — Instagram not only had a fully functioning product but also a million users. However, Instagram was only able to raise $7 million at a $20 million valuation, meaning that the entire company, its product, and its million users were being valued at less than half the Color Labs story.
What does a company do when it gets $41 million and has no working product? It tries to “get to the future faster” and pushes its engineers to rush a product out the door. The problem is that this doesn’t work. Software engineering is a creative exercise that requires thought and incorporates many dependencies. Often, components can only be built serially rather than in parallel. As a result, throwing more money and engineers at a product generally leads to a bad product.
Engineers have a term for rapid-fire development — it’s called “quick and dirty”. While quick and dirty coding can be effective for mocking up a prototype, it creates massive liabilities when it comes to creating working products. Even companies that hit critical mass with quick and dirty coding, as Facebook did, then go back and pay down the technical debt by assiduously rearchitecting the product and refactoring or porting the source code.
Even Facebook had more than two and a half years of dedicated software development under its belt before presenting its product to the general public in late 2006. By contrast, Color Labs had spent mere months on development by the time it pushed its product to the iOS app store in March 2011. Given the marketing hype behind Color Labs, more than a million users downloaded the Color app. Six months later, more than 90% of those users were gone. The Color app had a two-star rating. Contemporary commentators, including both Color Labs insiders and outside experts, called out the “bad UI” as a key reason for the failure, with some going so far as to state that the product was “unusable”.
Mark Zuckerberg, the founder of Facebook, famously advised engineers to “move fast and break things.” However, even Mark would surely advise that any broken things be fixed before the product is rolled out to customers. The engineering process is both complex and iterative, requiring time to perfect an application and make it usable. Early prototypes are nearly always deeply flawed and require extensive retooling, pivoting, and augmentation before becoming production-ready.
Unfortunately, many business executives and investors view engineering simplistically as a black box with two levers: time and money. Pressing the “money” lever is supposed to reduce the “time” lever, and vice versa. Software engineering does not work this way. According to Brooks’ law coined by the legendary IBM engineer and architect Fred Brooks in his seminal book The Mythical Man-Month: “adding human resources to a late software project makes it later.” Another way to say the same thing is to observe that just because it takes an artist one hour to make a painting, this does not mean that six artists working simultaneously will produce the same painting in ten minutes.
The co-founders of Color Labs were not engineers, but successful business executives. It is not surprising that when they were asked if more capital could get them to the future faster, they enthusiastically said yes. If software engineering was a manual activity with fungible resources that could work independently and in parallel, then more capital would have absolutely been the answer. However, if this was the case then large companies who employ thousands of engineers would be able to build a better version of every product and corner every technology market of note.
Successful product managers recognize that it is not the quantity but rather the quality of engineering that matters and they ensure that engineers have enough time to not only move fast and break things but also to back up and rebuild things, generally multiple times, until they get the product right. It’s better to deliver a winning product a year late than to deliver a losing product a year early.
Launch — Contrarian Advice: Tone Down the Marketing
Throughout 1957, the Ford Motor Company promised that it was about to unveil a “new kind of car” codenamed the “E-car”. Newspaper advertisements featured blurry images of what looked like a new vehicle design coupled with breathless marketing text. As the launch date approached, customer expectations reached a fever pitch. Capping off the hype, Ford sponsored a live television special that featured performances by the biggest stars of the 1950s, including Frank Sinatra, Bing Crosby, Rosemary Clooney, Louis Armstrong, and Bob Hope.
On September 4, 1957, dubbed as “E-Day”, more than two million customers poured into Ford dealerships, full of anticipation and excitement. On lots across the country, they were confronted and confounded by a boxy car with a vertical grille. Shockingly, the game-changing “E-car” was a disappointment in almost every way. It had an oddball shape. It had a conventional engine. It had slow acceleration. It had uncomfortable controls. And it had a positively uninspiring name: the “Edsel”.
Ford had anticipated selling 200,000 Edsels in the first year but ended up barely breaking 60,000. The following year, Ford sold fewer than 50,000. The year after that, the Edsel line was shut down, becoming an ignominious footnote in automotive history, a poster child for product failure, and a stark warning to engineers and marketers alike.
When it comes to launching a product, marketing is key. It brings customers to the point of sale and shapes customer expectations. However, if those expectations are not in line with the reality of the product then the marketing can backfire in a major way. Overpromising and underdelivering is not only a sure way to fail in selling a specific product, it is also a sure way to lose long-term trust. The Edsel debacle cost Ford billions of today’s dollars in direct losses; it forced the company to spend many more millions and years to painstakingly whitewash the stain on its reputation and win back disenchanted customers.
Notably, each of the product failures we have discussed involved overhyped marketing. Langley’s literal launch, Coca-Cola’s new formula, and Color Labs’ Silicon Valley splash all generated tremendous buzz. Customers were lined up, expecting to see magic, and when the products failed to live up to the hype, the nascent market share was irretrievably lost. Coca-Cola and Ford were badly hurt, Langley and Color Labs were destroyed.
This is not to say that products should not be marketed. Rather, great care must be taken with both the form and the content of the marketing. The marketing must set customer expectations in such a way that customers are enthused enough to try the product yet underpromised enough to be delighted by the product. The right marketing should never be like a movie trailer that shows all the best scenes or gives away the big reveal. And the best product managers work hand in hand with marketing to strike that balance.
Bringing It All Together
Ask ten product managers what they do and you will get ten different answers. By now, this should not be surprising, as we have seen that the best product managers deeply understand not only product but also engineering and marketing which are themselves highly complex and multi-faceted fields. The most transformative products involve out-of-the-box thinking, push the boundaries of customer expectations, incorporate engineering innovations, and delight beyond their marketing.
Notably, the most successful product managers typically do not create new product categories, they perfect existing ones. Google was not the first search engine, nor Facebook the first social network, nor iPod the first MP3 player, nor even Coke the first cola (although it was close). Great product managers look at the world and see opportunities for making it better than others may be too narrow-minded to consider or too busy to address. They seize on such opportunities with great drive and ambition, they build and iterate until they get them right, and then they unveil their products with open-mindedness and humility.
The one thread connecting all the contrarian advice we have discussed is not just the willingness to make mistakes — those are inevitable — but the willingness to acknowledge mistakes and course correct. Taking expert and customer feedback with a large grain of salt, allowing engineers enough time to break and to fix, and letting the product speak for itself allows room and time for error. The best product managers know that embracing and learning from tactical failures is the surest path to strategic success.
Jack Plotkin is the CEO of Cardinal Solutions, a boutique advisory and investment firm based in New York City. He has more than two decades of experience at the crossroads of business and technology and has advised more than a hundred Fortune 500 firms across all major industries, including healthcare and life sciences.