Chris Dixon

I’ve used Google Reader religiously since it launched.  I’m a few days away from quitting it forever.  Pretty much every blog I read tweets the titles of their posts along with a link.  Better yet, the people I follow retweet their favorite links, providing a very efficient way for me to discover new articles to read and publishers to follow.

Contrary to all the uninformed handwringing about how Twitter is making people dumb, I find I’m reading more long form blog and newspaper content than ever.   And the stuff I’m reading is more interesting and relevant.  That’s a good thing.

Meanwhile, Google Reader has been desperately adding social features such as sharing starred posts and automatically recommending blogs.  These features are clumsy and won’t save Reader, or RSS, from its inevitable decline.

Although I’m generally happier as a user, I think all of this is bad for the internet.  Twitter isn’t an open protocol.   It’s a private company with a profit motive that has a history of unreliable service. Moreover, URL shorteners – a byproduct of Twitter – are effectively creating a second layer DNS service that is far less secure and reliable.

I know that many people have been calling for an open alternative to Twitter for a long time.  I support them, but I’m afraid it’s too late. The network effects of Twitter’s social graph are just too strong.  Not to mention its brand momentum.  But the biggest reason Twitter has won is that mainstream users don’t care enough about these “principled” objections to switch.  Do you think Ashton or Oprah cares about open protocols?  I doubt it.

But someday they will care – when the internet is less open, less reliable and less secure.

A commenter on my blog the other day (Tim Ogilvie) mentioned a distinction that I found really interesting between intent generation and intent harvesting.  This distinction is critical for understanding how internet advertising works and why it is broken.  It also helps explain why sites like the newspapers, blogs, and social networks are getting unfairly low advertising revenues.

Today’s link economy is built around purchasing intent harvesting.  (Worse still, it’s all based on last click intent harvesting- but that is for another blog post).  Most of this happens on search engines or through affiliate programs.  Almost no one decides which products to buy based on Google searches or affiliate referrers.  They decide based on content sites – Gizmodo, New York Times, Twitter, etc.  Those sites generate intent, which is the most important part of creating purchasing intent, which is directly correlated to high advertising revenues.

But content sites have no way to track their role in generating purchasing intent.  Often intent generation doesn’t involve a single trackable click.  Even if there were some direct way to measure intent generation, doing so would be seen by many today as a blurring of the the advertising/editorial line.  So content sites are left only with impression-based display ads, haggling over CPMs without a meaningful measurement of their impact on generating purchasing intent.

All of this has caused a massive shift in revenues from the top to the bottom of the purchasing funnel – from intent generators to intent harvesters.  Somehow this needs to get fixed.

According to the Business Insider, Facebook is “‘Beating The S— Out Of Its Numbers’ Thanks To Zynga’s Virtual Goods.”  I wanted to try to understand this new, emerging economy.

It all starts when a user sees an ad on Facebook:

After clicking and installing the app, she gets a little farm where she can grow tomatoes and such.

Game seems pretty fun.  But she runs out of seeds, and wants more.  So she goes shopping for virtual goods.

Let’s say our protagonist is too young to have a credit card, so she decides instead to buy coins by signing up for a free offer.

She decides to download a toolbar.  Free greeting cards seem like fun.

The download puts an Ask.com search toolbar in the user’s browser.  Ask.com makes money off search ads.  Ask probably paid $1 to $2 for the install.  Some portion of that goes to Zynga, and then back to Facebook when Zynga advertises.

Farmville apparently does not advertise on Ask.com:

Thereby preventing the entire new internet economy from imploding in an endless cycle of circularity.

There is a widely held assumption that new business models will continue to emerge online – that statements like “how will Twitter ever make money?” will look as silly in 10 years as similar statements made 10 years ago about Google look now.

There is no question that, if they wanted to, Twitter could make tens of millions of dollars tomorrow, by, say, running ads or by licensing data feeds.   The big question is whether Twitter and other social media sites will figure out how to make Google-scale money and not just Facebook-scale money.  Google and Facebook get (ballpark) the same number of monthly visits to their sites.  Facebook made hundreds of millions of dollars last year and reportedly lost money.   Google made over $22B last year with huge profit margins.

The optimistic view (which I tend to hold myself) says that where people spend time, money will follow.  If people are spending all their time on Facebook and Twitter, the Proctor and Gamble’s of the world will eventually find an effective way to shift the bulk of their ad spending online.   The tacit assumption in this view is that the next 15 years will see as much business model innovation as the last 15 years.

On the other hand, what if we are mostly done creating big new business models for the web? History suggests that business model innovation is rapid right after the advent of a new medium and then slows down considerably.   If indeed it is slowing down, social media could end up like instant messaging – incredibly popular but basically lousy at monetizing.

A while ago I dug up this quote from Business Week from 2000:

But how will Google ever make money? There’s the rub. The company’s adamant refusal to use banner or other graphical ads eliminates what is the most lucrative income stream for rival search engines. Although Google does have other revenue sources, such as licensing and text-based advertisements, the privately held company’s business remains limited compared with its competitors’.

We now know what people were missing back then and why Google generates such massive revenues from advertising.  The lesson is that the RPMs* of online ads are directly proportional to the degree** to which the user has purchasing intent.  This is why when you search Google for “cameras” you’ll see ads everywhere (and those advertisers are paying high CPCs), but when you search for “Abraham Lincoln’s birthday” Google doesn’t even bother to show ads at all.

This is also why Nextag will have revenues this year in the ballpark of Facebook’s revenues, even though Nextag gets a fraction of the visits:

When people talk about search being a great business model (for, say, Twitter), they should distinguish between search with puchasing intent, which is an incredible business model, and search without purchasing intent, which is a terrible one.

This may change as brand advertising moves to the web.  But for now web advertising is dominated by “direct response” ads, and those are all about purchasing intent.

* RPMs = revenue per thousand impressions – can we please agree to start saying RPMs instead of CPMs or eCPMs?  :)

** degree being how close the user is to actually purchasing multiplied by the profit margin on what they are purchasing

For 10 years, Yahoo was my default home page.  Now I can barely stand using the site.  I still use it for Finance and Flickr, but that’s it. The new home page design has windows popping up everywhere and mind numbing celebrity gossip up top.

Now we learn Yahoo is going to spend $100 million on an advertising campaign.  The slogan is “It’s Y!ou” which sounds like one of those meaningless taglines invented by PR firms.  I’m quite sure no one will remember it and their money will be wasted (quick, name the tagline of any big tech company).

By CEO Carol Bartz’s own admission, Yahoo is incredibly well known, especially outside of techie circles:

When you get outside of New York City and Silicon Valley, everybody loves Yahoo…. We do great things for [users] and we’re excited about what we are.

Yes, Yahoo has one of the best brands on the web. Which is precisely why they shouldn’t be spending $100M on branding.  That’s the last thing Yahoo needs.  What they need are new technologies, new revenue streams, and new products that people love.  If they can’t build those things themselves, then they should acquire them.  They’re coasting on inertia right now.  As we saw with AOL and countless other tech companies before them, that inertia will be lost if they fail to innovate.

I think the Yahoo-Bing search deal is a great thing for startups as it potentially makes search competitive again.  But as a longtime Yahoo user it makes me kind of sad.  Between the branding campaign and the search deal, it feels like Yahoo has thrown in the towel.

There has been a lot of talk the past few days about Twitter raising $100M at a $1B valuation.  To understand what is going on from the investor side, you need to know about David Swensen, the man who (inadvertently) destroyed venture capital.

Mr. Swensen manages Yale University’s endowment and is the inventor of the so-called “Yale Model.”  Basically this is a model for people who manage the largest pools of capital in the world – universities, pension funds, wealthy family funds, etc.  The major idea behind the Yale Model is to put significant portions of one’s fund into “alternative asset classes” like venture capital.   Yale had phenomenal returns for many years (until this year, which was a disaster) and thus was copied by fund managers around the world.  This created demand to invest hundreds of billions of dollars into VC funds.  This in turn radically increased competition amongst VCs, thereby driving down their returns (public pension funds like California’s release the returns of their VC investments and they aren’t pretty).

What this means is that there are lots of VCs out there with huge funds and very little chance of getting “carry” (performance fees), since most will have negative returns (and they know it).  So instead they are collecting management fees (typically, 2% of the fund for 10 years, so 20% of the total fund).  They need to justify collecting these fees, which is why if you hang out with VCs you’ll often hear them talk about needing to “put more money to work.”  I would bet that the new investors were the ones arguing for Twitter to raise more and more money, even if it meant a higher valuation.  I’ve seen it happen many times.

The alleged societal benefit of patent law is that it creates a financial incentive to innovate.  The societal drawback is that it reduces competition, reduces the spread of innovation, and creates deadweight legal costs.

Perhaps patents are necessary in the pharmaceutical industry.  I know very little about that industry but it would seem that some sort of temporary grants of monopoly are necessary to compel companies to spend billions of dollars of upfront R&D.

What I do know about is the software/internet/hardware industry. And I am absolutely sure that if we got rid of patents tomorrow innovation wouldn’t be reduced at all, and the only losers would be lawyers and patent trolls.

Ask any experienced software/internet/hardware entrepreneur if she wouldn’t have started her company if patent law didn’t exist.  Ask any experienced venture investor if the non-existence of patent law would have changed their views on investments they made.  The answer will invariably be no (unless their company was a patent troll or something related).

Yes, most venture-backed companies file patents (I have filed them myself), but this is because 1) patents can have some defensive value, 2) they can grease the wheels of an acquisition (mostly because big companies want a large patent portfolio for defensive purposes), and 3) occasionally failed startups will get funded by investors whose intention is to go around suing people (hence providing “downside value” for the initial investors).

Articles like this recent one in New York Times promote the urban myth that the main beneficiary of patents are lone inventors whose idea is stolen by the big guys.  I have no special knowledge of the situation referred to, but I find it hard to believe in 1995 the idea of tying GPS to mobile devices wasn’t obvious to anyone in the field.   Almost all software and technology patents that I’ve ever come across are similarly obvious to practitioners at that time.  In theory obviousness is grounds for disallowing patents, but in practice patent examiners grants tons of silly patents.

Take the case of Blackberry and NTP.   NTP is a “patent holding company” – a patent troll – whose sole purpose is to sue people.  Now, I’ve been around long enough to know that the idea of mobile email is as old as email itself.  What RIM did was they actually went and made it a reality.  They figured out how to make a simple device that people loved, how to market it, and how to convince investors to give them money for what probably at the time seemed like an overwhelmingly difficult project.  The founders of RIM are the heroes of the story.   They didn’t need to sue anyone because they built a product and made money by actually selling a product people wanted.

How did having patents help society here?  NTP never tried to build any products.  No one is claiming RIM took the idea from them.  The only beneficiaries here are a company that never built anything and a lot of lawyers.

Software/internet/hardware patents have no benefit to society and should be abolished.

Holman W. Jenkins Jr. has an op ed in the Wall Street Journal today arguing against net neutrality. He positions himself as someone defending innovation and particularly startups against incumbents like Google.  For example, he says:

What if some startup Google sought to achieve the same goal by outsourcing its data management to the telcos, say, by mounting servers in their premises to help deliver Web applications more quickly? This would be a win-win for both parties. Data that travels within a carrier’s system is cheaper to deliver than data that must be handed off between two or more carriers. Would such an arrangement be a violation of net neutrality? Google would likely shriek so.

Huh?   Pretty much every startup I know does host their web apps at telcos.  My company Hunch, for example, hosts at Level 3.  The 20 or so startups I’ve invested in all do as well.  I’ve never heard any net neutrality advocate argue against this practice.  It also sounds like he’s unfamiliar with CDNs like Akamai.  In my experience those only tend to be affordable by large, international companies, so hardly favor startups.

He also seems unaware that there is already metered pricing on the internet today – it’s just paid for on the server side, by Google, Microsoft, my company, etc.

The greatest fear of Microsoft, Amazon, eBay and Yahoo is having to plumb their deep pockets and offer competing payments to broadband carriers to speed their bits to consumers.

I’m happy to show Mr Jenkin’s our Hunch bandwidth bill or he could just go look, for example, at Rackspace’s pricing page.   Apparently he wants websites to pay twice, once to the hosting provider and again to the ISP.  At any rate, I can’t imagine how “the more you pay, the faster your site runs” could possibly favor startups over cash rich incumbents like Google and Microsoft.

On the ISP side, he repeats the common anti-net neutrality assertion that there is genuine competition between ISPs.  Maybe there is where he lives.  Where I live in Brooklyn there is only one viable choice (Time Warner Cable).  Maybe once Verizon FIOS comes to my area this will change but for now I can’t get it.  At my office in Manhattan we couldn’t get T-1 service in our building, the Verizon DSL had atrocious quality of service and we ended up with only one viable choice – Towerstream WiMax beamed from the Empire State Building.

Finally, he argues that Google, Microsoft et al are just looking after their own interests:

But usage-based pricing that would give consumers a reason to think twice before clicking on a Google-sponsored ad? It would be the end of Google’s business model.

I agree this would be the end of Google’s business model.  But it would also be the end of the business models of pretty much every startup that is ad based – the vast majority of consumer internet startups today.

Personally, I tend to think net neutrality legislation is unnecessary as true competition in the ISP space is likely and will prevent any use-based pricing from gaining traction.  I am quite sure, however, that use-based pricing by ISPs would be disastrous for internet innovation, and especially for internet startups.

I know a brilliant young kid who graduated from college a year ago and now works at a large investment bank.  He has decided he hates Wall Street and wants to work at a tech startup (good!).  He recently gave notice to his bosses, who responded by putting on a dog and pony show to convince him to stay.  If he stays at the bank, the bosses tell him, he’ll get a raise and greater responsibility.  Joining the technology industry, he’d be starting from scratch. He is now thinking that he’ll stay, despite his convincing declaration that he has no long term ambitions in finance.

Over the years, I’ve run into many prospective employees in similar situations. When I ask them a very obvious question: “What do you want to be doing in 10 years?”   The answer is invariably “working at or founding a tech startup” – yet most of them decide to remain on their present path and not join a startup. Then, a few years later, they finally quit their job, but only after having spent years in an industry they didn’t enjoy, and that didn’t really advance them toward their long term ambitions.

How can smart, ambitious people stay working in an area where they have no long term ambitions?  I think a good analogy for the mistake they are making can be found in computer science.

A classic problem in computer science is hill climbing.  Imagine you are dropped at a random spot on a hilly terrain, where you can only see a few feet in each direction (assume it’s foggy or something).  The goal is to get to the highest hill.

Consider the simplest algorithm.  At any given moment, take a step in the direction that takes you higher.  The risk with this method is if you happen to start near the lower hill, you’ll end up at the top of that lower hill, not the top of the tallest hill.

A more sophisticated version of this algorithm adds some randomness into your walk.  You start out with lots of randomness and reduce the amount of randomness over time.  This gives you a better chance of meandering near the bigger hill before you start your focused, non-random climb.

Another and generally better algorithm has you repeatedly drop yourself in random parts of the terrain, do simple hill climbing, and then after many such attempts step back and decide which of the hills were highest.

Going back to the job candidate, he has the benefit of having a less foggy view of his terrain.   He knows (or at least believes) he wants to end up at the top of a different hill than he is presently climbing.  He can see that higher hill from where he stands.

But the lure of the current hill is strong.  There is a natural human tendency to make the next step an upward one.  He ends up falling for a common trap highlighted by behavioral economists:  people tend to systematically overvalue near term over long term rewards.  This effect seems to be even stronger in more ambitious people.  Their ambition seems to make it hard for them to forgo the nearby upward step.

People early in their career should learn from computer science:  meander some in your walk (especially early on), randomly drop yourself into new parts of the terrain, and when you find the highest hill, don’t waste any more time on the current hill no matter how much better the next step up might appear.