I think something more important than churn may be taking place. Or, stated differently, the market grew quickest during a period that likely also offered the highest rate of churn. The US market for music services offers a particularly useful space for digging into this issue.

First.. a distinction I would like to make between two kinds of churn: (1) the movement of a subscriber from one service to another (i.e., your loss is my gain) and (2) the loss of a subscriber from not only one provider but from the market as a whole (i.e., your loss is our loss).

I am making this distinction for a very simple reason: When we only consider the user base characteristics for clearly freemium services (e.g., Spotify and Deezer), we likely over-estimate the zombie population. The over-estimation is the result of counting “your loss is my gain” subscribers as “your loss is our loss” subscribers.

As fas a I can tell, of the 3.4 million paying subscribers to on-demand music services:

(A)  15-20% of the 2012 subscribers (650,000 [+/- 100,000] accounts) may have been “your loss is my gain” churn.

(B)  50-65% (2,000,000 [+/- 200,000] of this overall subscriber pool were not likely and directly acquired via a freemium model.

(C) Of the 1.6 million new subscribers in 2012, however, upwards of 60% (900,000 +/- 100,000) subscribers may have arrived by way of a freemium model. Said another way, of the 1.6 million new subscribers in in 2012, upwards of 40% may not have arrived by way of a freemium model.

(D) The total population of zombies in this pool, most of whom arrived in 2012, may likely account for as little as 38% (+/- 5%) of the registered user base — after taking into account “your loss is my gain” churn and new subscribers to the market who did not arrive by way of the freemium path. This may be an underestimate as well, since its very difficult for me to capture and/or estimate (without internal data) registered and inactive users prior to 2010.

For the long story, head to Rockonomic…

To make a long story very short: iRadio royalties may not be distributed in such a way that 45% of the royalties are paid directly to the featured performing/recording artists, 5% to the nonfeatured musicians and vocalists, and 50% to the sound recording owner(s) — as these royalties are paid out from more traditional webcasters. Instead, 100% of these monies could be paid directly to sound recording owners, with the artist contracts and the recoup machine determining what portion of these royalties eventually make their way to artists and musicians.

To make that short story very long, read on @Rockonomic.com…

There are some people out there who consider incubators a hot trend, if not a cliché. In the minds of these folks, incubators are simply playgrounds where monkey bars, see -saws, and slides have been replaced by computers, bean bag chairs, and whiteboards. The undeniably high failure rate of startups from incubators simply adds fuel to the argument that these new venture zones may be more akin to McDonalds playlots.

In my opinion, this “new” phenomenon of shared startup space is actually a reflection of something quite old. Incubators simply bring together and into and single location and clear sight those startup activities that were previously taking place in dorm rooms, apartments, coffee shops, and — well — garages. Furthermore, these community workzones encourage the sorts of behaviors common to vibrant startup communities: e.g., open innovation, tinkering, fluid teaming, demoing, and more.

If you see startup activity from an evolutionary (or natural selection) perspective, than the seemingly random attempts (and failures) of a large numbers of startups is what we need in order to happen upon that small number of surviving ideas and teams. In essence, the purpose of the incubator space is to accelerate natural selection — to curate an ecosystem within which both failure and the glimmer of success surface sooner rather than later, at as little cost as possible. In fact, this accelerated selection is more like evolution than incubation. Most chicks survive incubation. Most variations fail through selection.

From a purely pragmatic perspective, there are a number of more basic reasons for incubators having the potential to trump plan competitions, when it comes to raw startup development:

Building what you can build versus Pitching what you can’t build.

Incubators tend to require of the individuals therein either(a) the ability to build what they are imagining or (b) the tenacity to find someone, right now, to help build what they have imagined. As a result, the teams in these workspaces tend to be in position to act on their ideas, almost immediately.

Plan competitions, however, tend to be populated by ideas presented by teams who lack the capacity to pursue the idea at the moment. Instead, these people are pitching for the capital to acquire those things that early teams truly require — the portfolio of talent required to get through the most uncertain phases of the startup: proof of concept and customer #1.

Learning through mentoring versus Winning by judgment.

The best startup workspaces are more than just walls with whiteboards and power outlets. These spaces aggregate a community of individuals and, importantly, mentors. Building out a startup is ultimately a craft, and mentoring has been the method over the millennia through which craft is developed and refined.

Competitions are guided by judgement, and often conclude with that judgement. Winners are chosen, and that moment can often be the least time the team and the judges ever communicate again.

Furthermore, competition selection is wholly reliant upon the judgment of a small set of individuals. As a result, we significantly limit the population of new firms based upon the impressions of people who, while intelligent and experienced, may not have all of the information or experience that is needed for the best decision at the time.

The best of both worlds

When you think about it, what we really need is the best of both worlds: the market research, feasibility analyses, and capital that go into business plan competitions combined with the tinkering, teaming, and mentoring activities that are common among incubators. Certain universities have already begun to merge these two phenomena, and this combination may very well result in a compelling new cohort of student startups.

Simply stated, a random walk is a process that can be understood as a series of steps or stages, wherein each stage or step can be described according to some probabilistic set of outcomes (e.g., a coin toss, or a normal distribution). One of the classic examples of a random walk involves mapping the circuitous walk home of an individual who has had too much alcohol. At any corner on their walk home, this inebriated individual might turn left or right, go forward or turn back. While a map of this drunken walk betrays the underlying randomness of the clouded mind, the probability that the individual will make it home is actually pretty good — even when each decision at each corner is understood as simply a 1 in 4 chance of heading in the right direction.

If startup success were a random walk, we might find that attempts to predict success in this market would be foiled. Investment returns would be highly volatile, with low success rates but (potentially) very high returns. Investors in this sort of market would be very likely to embed “options” and other downside protections into their deal terms (e.g., liquidation preferences, follow-ons, anti-dilution, etc.).

Furthermore, and from a sociological perspective, actors in the market would be very likely to imitate the behaviors and characteristics of successful individuals (e.g., hoodies), given the ideal attributes of success would be hard to predict. For example, since Google instituted a 20% rule for employee projects, other firms would emulate this rule just in case it had something to do with Google’s success.

Evidence for random walk: present.

By and large, we tend to associate startup success with some indispensable personality trait (e.g., gambler, gunslinger, genius, or guru, according to my research), defendable market position, inimitable technology, or simply dumb luck. In this context, asserting that startup success is a function of something “random,” is largely interpreted to mean that all of those other factors are trumped by the last … dumb luck.

In truth, however, each of these factors — whether realistic or foolhardy — could play a role in navigating a random walk. I don’t translate “random” to mean that survival is based purely upon dumb luck. Instead, I tend to see randomness, when affiliated with risk, as a phenomenon that can be navigated and managed through a range of strategies.

From this random walk perspective, a phenomenon that may be random, at its core, can be managed successfully through a portfolio of options and other strategies that accomplish two interconnected goals: (1) limiting exposure to negative outcomes (aka, downside), while  (2) opening if not increasing exposure to positive outcomes (aka, upside). My word for this portfolio of strategies is “enterprising.”

For example, the premise of affordable loss emerging from the effectuation school in entrepreneurship research can be seen as an effort to manage downside risk in a random walk; never bet more than you are willing or able to lose. This rule limits downside risk by simply limiting the size of the investment. Affordable loss may be a “weak” strategy, however, since it likely limits both upside and downside by constraining the entire investment amount. Other strategies may be “stronger.”

Importantly, the odds of success are not really changed with these strategies, all other factors held equal. Instead, the firm’s exposure to both desirable and undesirable outcomes is altered. This difference is important. We may not really change the probability of survival in a set of firms, all in the same market, all with the same initial investment, etc.

However, we would see that the payoff profile of the more enterprising firms — those that manage risk in this asymmetrical manner — would be quite different from the payoff profile of the less enterprising firms.

What if startup success were best understood as a random walk down Sandhill Road?

The maths scribbled on this napkin didn’t have to be too complicated since the company has been rather forthcoming in public regarding (a) the total amount paid in royalties to music owners and (b) the number of video views, on a global basis, during various months.

Using a collection of the inputs (a) and (b), a back-of-the-napkin estimate of the effective rate per stream paid by VEVO would be:

$0.00238 per viewer stream (pvs), with a range from $0.00222 to $0.0025 pvs.

I will leave it up to the reader to compare that pvs rate for VEVO to the various per listener stream (pls) for the range of alternative venues for music experiences (e.g., music services like Spotify, Rhapsody; YouTube; Webcasters; Broadcast Radio). For example, most reports would now place Spotify pls rates at about 2 to 3 times larger than this estimated VEVO pvs rate.

For more details, head to Rockonomic.com