June 7, 2011

71) Google Wave

危 wēi  danger

A couple of weeks ago an entrepreneur I respect told me that Google had “lost it”.  He pointed to the high profile failure of Wave, which was released to the general public in May last year as clear cut evidence. At the time, Wave was heralded as the future of communication.  It was billed as a web-based platform and communications protocol that would merge the best features of e-mail, instant messaging and social networking, enabling users to communicate both synchronously and asynchronously. Its launch was hotly anticipated, invitations were even sold for up to $70 on auction sites.

But, despite this initial buzz the uptake was poor, and Google announced that it was going to suspend the product’s development in August, just 3 months after launch.  As my friend said, surely this was clear evidence that Google was losing its way?

机 jī opportunity

I see the failure of Wave differently.  Firstly, I respect the fact that Google is still failing with product launches, albeit a high profile one in this case.  It appears that its Innovation Time Off programme (in which employees can use 20% of their time to work on any project that interests them) is still alive and kicking.  It would be easy for Google to sit back on its laurels and focus on its core business, after all its most recent quarterly profit was $2.3 billion.  But instead, it seems to retain its hunger: some launches end in failure but many of Google’s successes, such as Gmail, Google News, Orkut, and AdSense originated from Innovation Time Off.

Secondly, with the benefit of hindsight, I like the way that Google handled the ‘failure’ of Wave.  It made the decision swiftly, even though it had sunk significant resource into the project, and looked at the residual value in the product before choosing to retire the code absolutely. Much of the code was handed to the Apache Software Foundation and turned into Wave-in-a-box: it made sense to open up code that Google couldn’t obviously benefit from to the open source community as any developments would probably drive the usage of Google tools. In addition, Google held on to some of the code and I have started to see Wave functionality appear in other Google products.  Most obviously is its use in Google Docs “Discussions”. Discussions happen alongside documents, they can refer to selected parts of a document, and can be accessed both in the Google Docs interface and also by email.   You can watch a video on the product here.  Given that Google is now benefitting from many of the Wave features that were well received it seems odd to call it an total failure, and even more odd to point to it as an indication that Google has “lost it”.

At IDEO we enjoy developing sacrificial prototypes which we put in front of consumers to test hypotheses.  We call them sacrificial because they are designed for learning rather than as the final solution, we are never too emotionally wedded to them.  With the rapid compression of time and costs to launch startups the Wave story and Apple choosing to discontinue Mobile Me yesterday make me wonder if we will see a trend toward ‘sacrifical startups’ – launched to learn and test approaches with successful aspects rolled over into new startups or folded back into the core business swiftly even if the startup fails?

How About…

  • Judging companies not by their failures but by the blend of successes and failures and how they handle the latter?
  • Failing fast and looking for the value in the embers?
  • Launching sacrificial startups: specifically to learn from consumer reaction to individual parts of the offer?

–

March 24, 2011

69) Zynga – is its Facebook dependency good or bad?

危 wēi  danger

More and more often startups are built with a total dependency on a third party, for example on external platforms such as Facebook, Twitter, Google Apps or iTunes.  The allure of access to a sales channel, rich data and potential customers is obvious but surely that dependency creates huge risk?  What if the external platform changes its approach or decides to develop its own version of the startup?  What if the platform decides that it will no longer support third-parts apps as Twitter did last week.   Zynga, the social gaming company, is an interesting example – it has been valued at over $10bn – surely this is crazy given its complete dependency on Facebook?

机 jī opportunity

I was inspired a few weeks by an article on Techcrunch.  The article categorizes dependencies as symbiotic or parasitic.  Parasitic dependencies are problematic, increasing the chance of the third party taking a hostile approach but symbiotic dependencies, in which both parties derive a benefit, pose less of a risk.

By this definition I think Zynga’s dependency on Facebook remains symbiotic.  Facebook has benefitted because games such as Farmville enrichen the Facebook’s user’s experience – currently over half of FB users play Zynga games, increasing their average time on site considerably.  In addition, Facebook Credits are integrated into all Zynga games (for which Facebook takes a cut of 30%) and the company is reportedly the largest advertiser on Facebook, spending millions a year to drive new installs.

In return, Zynga has benefitted by leveraging the rapidly growing FB community, successfully harnessing FB tools such as user news feeds to promote game updates.  It’s hard to imagine that Zynga would have grown its user base for CityVille within a week of launch to over 3 million at the end of last year without FB.  Given this symbiotic dependency, Facebook’s growth rate and estimates of it making $630m this year, Zynga’s valuation seems more reasonable.

It is also worth pointing out that none of this has stopped Zynga building a Plan-B, if only to improve its negotiating power with FB.  For example it began requiring email addresses from users (enabling it to contact them outside FB) and buying mobile application companies – including Newtoy (creator of Words With Friends and Chess With Friends) last year and New York developer Area/Code last month.

In the case of Twitter it seems ruthless for them to close down so many apps, after all those 3rd party clients helped create behaviors that now make Twitter so valuable: @ mentions, direct messages, re-Tweets and so on (none of which were Twitter’s idea originally).  But, those app’s were effectively in direct competition with Twitter, placing their own advertising and pulling users away from Twitters own platform – a parasitic dependency.  Twitter’s rationale as outlined in this developer forum explains this and the message is clear – if you’re going to develop apps make them symbiotic…

How About…

  • Considering who your business might be dependent upon?
  • If you do have dependencies can you make them symbiotic rather than parasitic?
  • And how can you plan for the worst?

—

January 30, 2011

66) SV Angel & Y-Combinator

危 wēi  danger

Y-Combinator, a startup bootcamp of sorts, provides seed money (generally less than $20k), advice, and connections to startups.  It has grown rapidly since its launch in 2005 and has had some high profile wins, including Posterous.

But the news this weekend that SV Angel and Yuri Milner are offering every Y-Combinator startup $150,000 in convertible debt, with no cap and no discount is still surprising.  Convertible debt is a common tool for early stage investment because it allows deferral of tough valuation decisions but investors usually include clauses to ‘sweeten’ the fact that they are investing earlier, for example a discount on equity at later stages and a cap on the valuation of their equity.  In this case there are no such clauses – there is no discount or ceiling whatsoever.

SV Angel is offering all 40 startups in this 3 month Y-Combinator cycle a free $150k loan.  Why would an investment firm ever offer such a large investment to 40 startups, some of which it presumably knows nothing of, with no preferential terms for investing earlier?  Surely this is investor suicide?

机 jī opportunity

The more I’ve thought about it the more smart I think SV Angel is being: this deal is all about access.  Paul Graham, the founder of Y-Combinator has always said “This is a hits driven business”, and SV Angel’s offer gives it access to the ‘hits’.

Counter-intuitively, the generous terms really help with this.  If the firm was offering onerous conditions surely only the least promising startups would accept the investment and the ‘hits’ would slip through the net, a sort of negative selection bias.  However, under these terms I imagine all the startups will take the investment.

So, this offer effectively becomes a broad bet on Y Combinator startups, and a means by which to access the ‘hits’ (and potentially invest in them in subsequent rounds).

When Y-Combinator launched in 2005 it invested in 8 startups, in the current cycle there are 40 – this growth rate and the high profile ‘hits’ to date , including Scribd, reddit, Justin.tv, Dropbox and Posterous make this a smart bet.  The offer may even improve the outcomes for Y-Combinator startups if it enables them to focus on the important stuff, like finding a business model, rather than worrying about fundraising.

How About…

  • Remembering to look for the hidden cost of driving a hard bargain – are you creating a negative selection bias?

—

Techcrunch article here

« Older Entries