Sam Altman on Defining Startup Success

If we learned anything from Quirky, a now defunct New York startup that tried to disrupt the physical consumer product invention industry, predicting a successful startup is not as easy as it may seem. Quirky, which recently laid off nearly all its employees and filed for bankruptcy, seemed to be doing well. After all, the startup had a visionary and charismatic founder, top notch investors, a sophisticated board of directors, numerous followers, great engagement, an excellent team of engineers, designers, marketers, and other professionals -and a $180 million investment war chest to boot. What could possibly go wrong!?

Diving Bravely into the Unknown

In an effort to increase their chances of success, many startups apply for well-known fund programs such as the prestigious Y Combinator (affectionately known as “YC”). YC was started in March 2005 and is currently accepting applications on a rolling basis for the January through March of 2016 program. In exchange for 7% equity, YC supplies seed money, advice, and connections to startups during the three-month program. YC graduates include some of Silicon Valley’s best success stories such as Dropbox, Instacart, Reddit, Zenefits, Airbnb, Stripe, and numerous others.

But how does YC -- or anyone, for that matter -- predict a successful startup?  Sam Altman, YC’s president, recently shared YC’s portfolio stats. While Altman admitted in his post that “some of these [statistics] aren’t that helpful,” he still bragged about numerous achievements and apparent success metrics. For example, Altman shared that a total market cap of all YC companies is more than $30 billion, the total money raised by all YC companies is more than $3 billion, there are three YC companies worth more than $1 billion, and there are more than twenty YC companies worth more than $100 million.

Around the same time Altman used a series of Twitter posts to reflect on YC’s success, what makes a successful startup, what causes failures, and how to pick a “winner”.

These posts revealed a few more numbers about YC startup worth. Altman masterfully danced around the definition of success. He again threw out high, impressive numbers and cushioned them with disclaimers that these are not success metrics. All in all, yes the numbers are impressive. But what do they say about success? The portfolio still contains 690 or so companies that haven't broken the $100 million mark, after all.

Altman also reflected on the logical definition of failure. A startup that has shut down has obviously to some extent failed, at least for those involved in that specific startup. Yet the “shut down” rate may actually be a conservative definition of “failure” because it does not include startups that are struggling or have had a less than then lucrative exit. Like success, “failure” has no set definition.


Finding the Measure of "On Track"

Naturally, numbers may help to predict whether a startup is on target with respect to its specific goals at the moment. Depending on the industry and circumstances, these numbers may include gross margin, customer acquisition cost, churn rate, annual contract value, profit, free-to-paid conversion rate, net promoter score, and numerous others.

In fact, finding the right matrices that measure “being on track” in the context of a specific startup at any given point in time is more important than actually measuring it. If the right matrices are chosen, the rest is just a question of measuring. If the wrong matrices are picked, the accuracy and flawlessness of measuring is irrelevant.    

To be fair, YC helps startups in many different industries, and any matrices of success and failure are, of course, industry specific. This means that calculating general success and failure stats may be a bit hard to pull off for YC startups. One must also distinguish between measuring incremental progress and defining overall success.  

In light of Quirky’s experiences and Altman’s observations, it seems that some metrics are actually irrelevant in predicting startup’s ultimate success even though they may be good measures of incremental progress in a startup’s evolution. For example, how much capital is raised, the number of people hired, the startup’s geographic location, or where the product is sold may be more distracting than helpful when predicting the ultimate success of a startup. These metrics may be nice to have, but aren’t necessary. 

This fuzziness about what truly determines success and failure inspired Altman to reflect on how to pick a “winning” startup investment. Ironically, Altman seems to emphasize a human element, relying on his personal judgment and conversations instead of numbers or traditional matrices. This leads to a logical question: do statistics matter at all?

How a Failed Venture May Still Produce a Successful Founder

Ultimately, with startups as with life in general, defining overall “success” and “failure” is subjective, unlike measuring incremental progress. Understanding success really depends on one’s perspective and the length of time one chooses to focus on. For example, it is not unusual for founders to have failed a few times before they succeed. These “failures” often educate founders or provide them with an opportunity to find more favorable circumstances for their ultimate ideas to flourish. Investors also assume that many of their investments will “fail” and the few that “succeed big” will more than make up the losses.  Failed startup investments educate investors as well as founders. In the short term, these may be failures, but in the long term, they can lead to success. Finding the correct matrices can help get a better picture of a startup’s projected success or imminent failure, but even the most accurate numbers can’t account for perspective, experience, and the “human element”.

This is why I choose not to view Quirky as a “failed startup.” Quirky took on a very challenging industry. The startup’s ambitious attempt to disrupt the outdated patent system was a hefty undertaking in any context. Yes, Quirky did bite off more than it could chew and in the process lost focus. Quirky, however, showed that the physical consumer products industry and the outdated patent system are disruptable. Perhaps Quirky wasn’t the right company to take on such established giants, but they took the first shots, serving as an example for future contenders. It is just a question of time before we see the change and recognize Quirky’s rise and fall as a necessary step toward success, not a “failure.”