[ I was recently involved in a private debate with a few friends over Om Malik's "Slide, Vic Gundotra & The Un-Social Reality of Google". I initially sent the article pointing out that it summarized in more eloquent words my personal view on Google's current "social" woes. One of my friends came back with a rebuttal on a few points, one of which being that the "empathy"/social-science argument is BS because a lot of those other players successfully delivering social software are nothing but geeks themselves - a point which I somewhat agree with. However, never one to back down from a good debate, this rebuttal motivated me to think through more carefully what I really thought on the whole issue. And that led me to pen the following on go-to-market; reformatted from its original email format for the current purpose. I'll cover my take on Google's actual social issue in a separate post. ]
1. Learning go-to-market (GTM) basics is an organic (painful) process
The issue with any new software/website is GTM. My first-hand experience has been that GTM is the toughest thing geeks (myself included) have to deal with when trying to take a great idea from the "hmm, this is interesting" stage to a runaway success. Now maybe I'm just not smart enough, but I've been in business in IT for the past 10 years and I've only recently started to feel somewhat confident that I actually understand it. That is, not on a conceptual or theoretical or even strategic or planning level, but on a very down to earth practical level: how do I go from 0 users to 1 user to 10, to 100, to 1,000? Most of the time this simply means that I abandon the vast majority of ideas that I can think of because I don't see how I could execute a successful GTM on them. In other words, burn yourself often enough trying to bring "great" things to the market and you develop an ingrained cynicism about anything new in tech that makes you say "nah, I just don't see this one" most of the time. One thing I feel this experience doesn't give me is the ability to somehow magically predict what will work. Now THAT is a superpower I'd like to have, as I'm sure the plethora of startups, some founded by new and others by serial entrepreneurs, whose corpses litter cyberspace and whose only dim hope of an after-life is to be a Wikipedia footnote for having been there trying to do foo before such-and-such now established startup became synonymous with foo.
[ Side-note: I'm only mentioning geeks because most new software/websites originate from geeks turned entrepreneurs. The above most certainly applies regardless of background. I'm actually inclined to think that geeks (and technically-inclined folk in general) are more capable of finding successful GTM strategies than folks from other backgrounds simply because they can get their hands dirty in the iterative tweaks required to get a GTM to eventually work. ]
2. Size/clout/hype doesn't obviate GTM issues
Here's an excerpt from Steve Jobs' appearance at the D8 conference a few months back (I highly recommend it, it's one of the most important videos I've seen in the past year):
Q: Hi, I'm from Hillcrest Labs... do you think it's time to throw out the interface for TV? When will Apple do something there?
A: The problem with innovation in the TV industry is the go to market strategy. The TV industry has a subsidized model that gives everyone a set top box for free. So no one wants to buy a box. Ask TiVo, ask Roku, ask us... ask Google in a few months.
One false-hood many startups entertain internally is that their thing would be an unstoppable phenomenon if only they had a brand or the fire-power. That is simply flat out untrue. For Steve Jobs to say something like the above, for Google to have failed with Wave, amongst other things, and for Microsoft Bob to continue being remembered to this day shows that successful GTMs can be fatally elusive, even for the most established of players.
3. Successful GTMs are almost impossible to engineer
I'll submit to you that most well known cases of IT success are outliers (Gates, Zuckerberg, Page/Brin, etc. ... keep Jobs on the side for a moment) and that success in IT is stumbled upon and cannot be designed. When approached for an acquisition by a Viacom executive, for example, Zuckerberg apparently said: "I don't think I'm ever going to have an idea this good again". In "Founders at Work" Sabeer Bhatia says about life after Hotmail: "... I have not been able to replicate that kind of meteoric growth and success yet." Max Levchin sold Paypal 1.5B$ to eBay but despite wanting to make as much with Slide, he only sold it 182M$ to Google. And so on ...
Only an *extremely* rare breed of people eventually figure out how to connect the dots to maximize potential success. My personal opinion is that folks like Jobs and Andreessen are these kinds of people, but it takes decades AND a lot of trial and error (Jobs' disappearance from lime-light between '87 and '97 is not without incidence on his current success.)
[ Side-note: I highly recommend Marc Andreessen's talk at Stanford last May if you haven't already watched it. The guy is simply brilliant. ]
So, therefore, if you can't study it in a textbook, can't use heavy weaponry to ensure its success and can see that most that succeed at it had lady luck tilt on their side ever so slightly, what are you to do? One thing is for sure, if I had THE answer I wouldn't be sitting here blogging about it. I do, however, believe that figuring out a successful GTM for any new software/website requires adopting what is currently being labeled as the "lean startup" model. As Eric Ries explains (check out his Web 2.0 conference talk), the basic idea is that when a startup starts, it doesn't know what its product is and who its customers are and that it'll take it quite a few "pivots" (iterations of: create product, get product in user hands, modify/scrap/restart product, reconstruct/deconstruct business model, etc.) before it actually is a real business. The faster pivots come, the faster a startup will converge to a real business. In other words, this is a very organic process. Most often, you eventually stumble upon a pivot that works because you came from n-1 pivots whose sole benefit to the company was learning.
The most interesting thing for me, both as an entrepreneur and as an avid follower of all things tech, is that mature businesses usually handle pivoting very badly. If pivoting were easy for mature businesses, they wouldn't have to fight with Innovator's Dilemma. Think of it, what are the chances you'll be allowed to gamble with an established brand and release what would otherwise be considered a sub-par product just to get to your next pivot? Can you justify that to share-holders, other divisions, etc.? Obviously you can't. Even if you could pull this off, how long can you keep it up? (i.e. how many pivots will you actually be able to achieve?) It's not for nothing that Microsoft is struggling very hard these days with the onslaught of web and mobile technologies where it's totally absent. IBM faced the same thing in the early '90s and had to re-invent itself as a service business and give up on OS/2 etc. And, to an extent, Google is facing this today vis-a-vis the rise of social and "app" platforms; which will be the topic of an upcoming post.
And that last bit is probably why I've never gotten tired of being involved in technology: plot twists and surprise endings are the rule.