Is Venture Capital Broken?

See this post on VentureBeat, entitled The VC Model is Broken which asserts that the venture capital model no longer works and that the Bubble 1.0 get-out-jail-free card given to the industry around 2001 has now come due.

The post refers to a presentation by Adeo Ressi, founder of Yelp-for-VCs site The Funded which asserts that:

  • VC is too much of a “hits business” (i.e., returns are lackluster excluding the 1-2 top hits per fund, and that some funds, presumably the B- and C-tier ones, don’t even have those hits)
  • VC is too clubby with too many exits of newer companies going to earlier portfolio companies at lofty valuations. (“Can you please buy my other company, at a premium?”) One cannot help but think of YouTube’s $1.6B exit to Google in this context.
  • VC is too herd oriented, resulting in too many me-too companies being funded and too few truly innovative companies being funded.
  • 2H08 is the first time period in which VC exits are less than VC investments

He recommends the following changes:

  • Less funds and better funds
  • More deals and equal treatment
  • Simplified terms and standard structures
  • Improved fund governance
  • Restructure fund incentives

Personally, while I agree with many of his asserted troubles, I generally disagree with his recommendations. To me, market forces should work over time to correct all ills.

  • Presumably, if mainstream VCs are too herd oriented then newer/different VCs should be able to stake out the different positioning. And LPs who seek such differences should be able to find them.
  • Similarly, if there is a problem with mainstream VC terms/structures, then presumably B-tier and/or new VCs can attempt to differentiate themselves by offering these different terms.
  • Increasingly, private equity firms are entering and innovating in VC already. I suspect they are exploiting the opportunities created by the standard criticisms of VC.
  • With the fall in the stock markets, most institutions are presumably now over-allocated to VC and/or private equity. That is, if you want a 10% VC allocation and the stock market falls 30%, then presumably you end up in a 15%+ position. So I suspect institutions and investors will be seeking to reduce VC exposure, not increase it.

Basically, I’m a believer in Darwin/Malthus. Yes, VC was “too easy” in the 1990s and presumably too much money flowed in and too many firms and funds were created. But the natural response to this, over time, should be a weeding out of the weaker funds and players. Since VC timeframes are elongated, with the typical fund lasting 10 ten years, it will take a long time for these cycles to play out — but play out they will.

Here are the slides from Adeo Ressi:

/2008/11/18/is-venture-capital-broken/

Related articles by Zemanta
Reblog this post [with Zemanta]

A New Media Plan for Weathering the Storm

Check out this post by Gawker Media’s Nick Denton entitled A 2009 Internet Media Plan. It’s a publisher’s wake-up call and the Sequoia RIP Good Times presentation rolled together, all in one. Valleywag, itself a Gawker property, entitled its story about the post Publishers are Sleeping their Way to Extinction (supposedly the draft 1 title, according to them).

Denton offers six pieces of advice for (largely consumer-oriented publishers) the future:

  • Get out of advertiser-averse categories, e.g., politics
  • Renegotiate vendor contracts
  • Consolidate titles
  • Move more offshore
  • Implement variable compensation plans
  • Deliver more value for marketers

Unlearning the Relational Model

Thanks to a Google Alert I stumbled into this interesting post entitled The Content Imperative: Unlearning the Relational Model in another CEO blog, that of Joel Amoussou of Montreal-based Efasoft.

Says Joel:

The following are some fundamental differences between content and relational data:

  • Content is created to be human readable
  • Content can be rendered in multiple presentation formats such as print, web, and wireless devices. Therefore it is very important to cleanly separate content from presentation
  • Content can have an inherent deep hierarchical structure. For example, think about the book/part/chapter/section/subsection/paragraph hierarchy
  • The relationships between content items are expressed through hierarchical containment and hyperlinks
  • Content is often mixed (in the sense of mixed content in XML). For example inside a paragraph, some words are italicized, in bold, or underlined to indicate special meaning
  • Content can have multi-valued properties such as the authors of a document. Multi-valued properties are not supported by SQL.

He continues, starting an argument in favor of XML:

The problem with unstructured content is that it cannot be processed and queried like the well-structured relational data stored by the RDBMS on which your ERP and CRM systems sit. XML goes beyond tags (in the web 2.0 sense), taxonomies, full-text search, and content categorization to provide fine-grained content discovery, query, and processing capabilities. With XML, the document becomes the database. If your business is content (you are a media company, a publisher, or the technical documentation department of a manufacturing company), then you should seriously consider the benefits of XML in terms of content longevity, reuse, repurposing, and cross-media publishing.

And goes on to discuss XQuery:

The relational data model is based on set theory and predicate logic. Data is represented as n-ary relations and manipulated with relational algebra. CMS vendors and even standard bodies have tried to fork SQL in order to support hierarchies and multi-value properties. It is clear however that XQuery is a superior alternative, specifically designed to address those content-related concerns.

And then finally argues in favor of XML databases over a JCR repository when dealing with large amounts of content:

You should seriously consider a native XML database when dealing with large quantities of document-oriented XML documents.

I couldn’t agree more. (Hey, I think I like this guy). The post also includes some discussion of data vs. content modeling and some interesting parallel history between SGML/XML and the RDBMS.

Reblog this post [with Zemanta]

The Valley and the Bust

I found an interesting article in Forbes the other day entitled The Silicon Lining / Why the Bust is Good for Silicon Valley and I thought I’d discuss it a bit here.

Let’s start with this excerpt with some interesting IPO-related stats:

Wall Street has been broken for eight years now, as far as Silicon Valley is concerned. Alan Patricof, a legendary venture capitalist, recently remarked: “We no longer invest with the idea of taking our companies public. If they do [IPO], it’s an accident.”

… Since 2002, there have been just 351 IPOs out of 19,300 VC-backed companies–fewer than one in 50 …

… The ratio of mergers and acquisitions to IPOs has gone from roughly 1:1 from 1996 to 2000 to 6:1 during 2001 through 2008.

As I wrote in Built to IPO, Flip or Last, the IPO bar has been raised and the window has been largely closed for quite some time. This has a number of effects:

  • Since 2001, the IPO window has been closed more than it’s been open so those relatively few companies who get above-bar often end up all dressed up with nowhere to go (e.g., Endeca).
  • Ceteris paribus, a time delay shouldn’t depress venture returns. Provided a company can maintain its impressive trajectory, the whole process should simply take longer at roughly constant IRR. Theoretically, the money remains at risk longer but, ceteris paribus, the IRR should be the same.
  • But, in the real world, ceteris aren’t paribus. Some companies derail between $30M and $60M. Valuations and multiples fluctuate. The lack of IPO exits can potentially depress M&A valuations. So, all things considered, I believe Forbes’ assertion that the net effect of a high bar and a closed window is depressed venture returns. Recent data from the NVCA support the claim as well. (See chart, noting particularly the grim situation in early-stage VC.)

The result of all this, says Forbes, is a sort of infanticide of great companies:

The sad truth is that we are replacing potentially great companies with under performing divisions of mature companies. Acquisitions invariably remove both the future risk and rewards–not just for the company but for society as a whole. Innovation is stifled, and that hurts us all.

Competing with EMC and watching what they’re doing with (or should I say to) their xDB division (formerly EMC Documentum XML Store, formerly x-Hive), I get a visceral sense of their point. Consider the fate of Amazon, says Forbes, in an similar IPO-shut environment:

For example, a company like Amazon.com which went public in 1997, could never have had an IPO in this environment. Instead it would have become a part of Walmart and likely would have been shut down during the tech bust.

While that’s a bit harsh, my beef has always been simple: why can’t the public buy a share of Endeca stock? They’re a $100M company. They’re far past the stage that should require accredited investor status.

All of my previous employers went public in the roughly year in which we did $30M. While I don’t agree with Endeca’s increasingly de-focused (or should I say decreasingly focused) strategy (the latest thing is now digital asset management), I do firmly believe that John Q. Public should be able to buy their stock. By raising the IPO bar to $50 or $100M, you effectively lock the public out of early- to mid-stage investments. That’s bad for the public. It’s bad the companies. It’s bad for the VCs.

But the past is the past. What’s happened in the past 8 years neither dictates nor predicts the coming 8. I recall when I quit Versant being absolutely sure the company would never go public, only to find it IPO-ing 18 months later. (Happily, I’d nevertheless exercised at least half my options on the Kellogg Uncertainty Principle: when in doubt, do half.)

I’m not alone in having some optimism about the future. The Forbes article continues:

The $100 million technology company will become an attractive investment again. Both Silicon Valley and Wall Street will once again bet on creating the next Amazon.com. And in my opinion, the bar for an IPO will go down over the next few years, once again creating a vibrant ecosystem in Silicon Valley.

Economist and former Chairman of the Federal Reserve Paul Volcker has said that the so-called “financial innovations” of the last few years largely rearranged existing resources instead of making real contributions to the economy. As a society we want financial returns to be aligned with value creation. This crisis will jar the the two back into alignment. Value creation is hard. But no one does it better than Silicon Valley.

For a dose of real venture optimism, check out this video I found on the related news section of Forbes.com where VC Charlie Harris predicts an eventual IPO boom.

Reblog this post [with Zemanta]

Oracle Buys Tacit's IP

I’ve blogged once before about email-sniffing Tacit Software and its founder (and my old boss) David Gilmour. So I was a bit saddened to read this story on VentureBeat entitled Oracle Scoops Up Tacit’s Intellectual Property.

Now, happily, I lack personal experience in such matters, but I think that “normally” when software companies sell their IP assets to Oracle, it’s a — uh — bad sign. I take it to mean that (1) the company’s winding down operations, (2) that no one wishes to buy the business as a going-concern, thus purchasing both all assets and all liabilities, and ergo (3) that — in the absence of other interest — someone can come along and simply buy all or some of the assets.

I’d not seen any stories on Tacit shutting down, so I was surprised to hear they were selling IP to Oracle. This Network World story has a little more detail, but the final clue came from Oracle’s website, here:

Oracle has also hired all of the software engineers from Tacit Software.

It looks as if it’s “goodbye Tacit” and presumably “poof” to most of the $29M in venture capital invested.