Entrepreneur Dan Cohen is something of a student of failed tech products, and at the top of his list of “dishonorable mentions” is the iSmell. The desktop device was a “personal scent synthesizer” that, when hooked up via a USB port to a computer, would deliver an olfactory experience appropriate to whatever website a user was visiting. While it sounds like a cross between a parody in The Onion and an off-color joke, the iSmell actually existed, however briefly, back in the dot-com glory days of 2001.

In presentations like the one he gave recently at a Wharton Entrepreneurs Workshop conducted at Wharton San Francisco, Cohen uses the iSmell as a humorous cautionary tale about the art and science of product management, especially for what can go wrong when the process goes off the rails.

Product management, said Cohen, is a make-or-break issue for most start-ups; getting the product right is a foundation of future success. But he noted that the process of product development is poorly understood, with young companies often repeating common mistakes that can be easily avoided.

The field does not want for advice; there are plenty of product management how-to guides lining bookshelves, though Cohen said many of them are of dubious value. “There is a lot of literature out there about product management,” he stated. “But be careful about what you read, because you can really go off the deep end.”

Cohen is CEO and co-founder of Accomplio, which helps other companies bring their products to market. Before Accomplio, Cohen was involved with other web start-ups, including mySpoonful, a music site, and Pageflakes, a web page personalization service. At other points in his career, Cohen held senior positions at both Google and Yahoo.

“There is a lot of literature out there about product management. But be careful about what you read, because you can really go off the deep end.”

Much of Cohen’s presentation took the form of listing rules — notably, his 10 rules for success in product development. Among them: Don’t confuse yourself with your customer, since your requirements for a product are probably much more sophisticated than those of the rest of the market. “Remember, it’s not about you,” Cohen warned, stressing the importance of keeping the focus on the customer at all times.

Other rules: Make sure you have the right business model; you don’t want to have an expensive direct sales force for a low-cost product users could easily sign up for online. (That may seem obvious, but Cohen told the story of a web teleconferencing company that violated that very rule and quickly went out of business.) Don’t try to design a product that you don’t have the resources for, either in dollars or expertise, he noted.

And also, don’t equate innovation with value. Just because a product is technically interesting, or does something never previously possible, there is no guarantee of its success in the market. (That was one of the many rules that Cohen said was broken by iSmell.)

And there is a corollary to that rule, Cohen added: Don’t attempt to improve a product simply by adding more features to it, or making it more complex. More often than not, warned Cohen, such steps usually end up making the product worse.

Failed products have other things in common besides not following the rules for success. For example, they will often have a poorly planned user experience, Cohen noted, making the devices difficult to use.

A special danger for start-ups aiming products at the corporate market is to forget that the customer and the user often aren’t the same person, he added. A product that is popular with users might not be as warmly received in the IT departments that are responsible for technology purchase decisions.

The Importance of ‘On-boarding’

Cohen said that the goal for a start-up should be what he called a “minimum viable product,” one that a company can introduce into the marketplace and then build on with successive versions. To do so, he urged entrepreneurs to use what has come to be known as a “lean” approach popularized in such books as The Lean Startup by Eric Ries.

In the case of product development, Cohen noted, that involves taking an “iterative” approach. First, the kernel of the product is developed. It is then tested by having potential users put the item through its paces. The results are then incorporated into revisions of the prototype, with the process repeated until the product is ready for release.

This customer discovery process “involves achieving the right product-solution fit. If you can’t figure this out, then you should give up.”

“It doesn’t need to be a long and drawn out process,” Cohen pointed out. “You don’t need lots of studies.” This customer discovery process, he added, “involves achieving the right product-solution fit. If you can’t figure this out, then you should give up.”

That may sound straightforward, but Cohen said there are a number of potential stumbling blocks. One of them involves testing a product with people who aren’t representative of its actual potential customer base. That might happen when resource-strained start-ups rely on friends and family for testing. Not only might those people not be representative of potential customers, but they also might be unwilling to give the sort of candid feedback an entrepreneur needs, out of a desire to not hurt anyone’s feelings.

Cohen stressed the importance of easy “on-boarding,” which refers to the process by which a user begins to use a product, whether it’s a website or a piece of software. On-boarding, he said, should be easy and intuitive. If a product has a lot of advanced or sophisticated features, designers shouldn’t overwhelm users with them at the beginning, but instead, they should allow them to be discovered gradually as the product gets used.

He suggested a design philosophy of: Keep it simple, with a dagger preferred to a Swiss army knife. “Simplicity can be the biggest feature in and of itself,” he said. “Don’t ruin it.”

No Time Like the Present

One of Cohen’s themes during his talk was that there is no time like the present to create a start-up. The availability of free or low-cost web tools to help with the product development process is one of the main reasons for that being true, said Cohen. In fact, he added, the costs involved with starting a company have declined so much that venture capitalists are beginning to become concerned about what role they will play in the technology industry of the future.

“There really is a tectonic shift underway,” Cohen stated. “Software has become easy and cheap to build.” He added that because the barriers to entry are dropping, the marketplace will likely become increasingly crowded with competitors.

“Apple is a unique and rare case, one that is very hard to duplicate.”

Cohen also discussed some of the growing number of web tools tailored for entrepreneurs.

A product called Lean Canvas, for example, provides a blueprint that lets entrepreneurs see if their overall approach is following lean principles. Balsamiq provides an easy way to create a prototype version of a software product to test it out on users, he said. Other easily available web products that Cohen recommended include Jira, which implements what has come to be called the Agile development methodology, and GitHub, a central repository for the computer code that programmers write in the process of bringing a product to market.

Cohen also noted the popularity of sites such as eLance and oDesk, which allow start-ups to hire highly trained professional help on a per-project basis, often at a fraction of what it would cost to bring on a full-time person. Some sites are highly specialized, such as uTest, which allows for crowd-sourced debugging of software code. There is even a site called FounderDating, where entrepreneurs looking to fill out a start-up team can find potential partners.

According to Cohen, the composition of the ideal start-up team starts with the CEO, who should have a strong background in product management. Also necessary, he added, are founders with extensive know-how in technology and user experience. If not all of those boxes can be checked right away, he noted, members of the founding team may need to serve dual roles on an interim basis until a spot can be filled full-time.

Examples that Cohen gave of companies that did everything right include Mint, which has become popular for personal finance and was acquired by Quicken, and Dropbox, the file sharing site. Cohen admitted to being a huge fan of Dropbox, calling it a “massively successful lean product” and praising the company for not making it more complex as the service became more popular. “I respect those guys a lot,” he said.

While describing the lists of dos and don’ts that entrepreneurs should follow, Cohen also noted the irony of the fact that some of the most successful entrepreneurs have been those who broke all the rules. The paradigmatic example, he said, was Steve Jobs.

Apple, noted Cohen, never did any of the things it was “supposed” to: It didn’t listen to customers, and it introduced finished products all at once, rather than gradually releasing revised versions. But Cohen added that in addition to the presence of Steve Jobs, Apple also had resources that most start-ups lack, like a hefty marketing budget that allowed the company to nearly bury the country in billboards promoting products like the iPod and the iPad.

Said Cohen, “Apple is a unique and rare case, one that is very hard to duplicate.”