Venture Explorer

So you thought you wanted to build a sticky website ...

Andrew Chen has a great post explaining why sticky sites that get lots of organic traffic monetize very poorly (i.e., have very low eCPMs).  Here's the meat of his argument:

For the people who are curious, this is the easiest to monetize:

One-hit wonder site that exist in a particular category, are based in the US, and have lots of search traffic

In particular, your site is likely to have high CTRs since people are in a "transactional" mode. If you have all of those, and have a ton of pageviews, then you'll make a ton of money.

The hardest to monetize?

Highly sticky sites that are general (like communication), based 100% outside of the US/Europe/Japan, with lots of pageviews

In a setup like this, not only are people unlikely to want to buy anything, even if they did, there'd be no way to make money off of this group.

Before you rush off and build an SEO/SEM-driven transaction engine, though, beware that the Lord Google giveth traffic, and the Lord Google also taketh away traffic - when algorithms change, search indexes are rebuilt or somebody else can afford to lose a lot more money on buying traffic. Or maybe just because Google deemed it so.  Highest RPVs (revenues per visit) and eCPMs are achieved when the site can capture the user's intent, fairly specifically and usually through a search engine, and yet offer a differentiated enough experience that the user, the search engine and whoever's dipping into the user's wallet all agree that the site added real value in the middle. 

Thus, a site that buys mortgage keywords on Google merely to land visitors on a landing page full of Adsense mortgage ads and hopes to make money on the spread might be profitable in the short term but will find its arbitrage margins squeezed out soon enough.  Conversely, if the site provides, say, a unique vertical search experience specific to a domain, it will be relatively immune to efforts by Google to bypass it.

Kayak.com,, a vertical search engine for travel, is a good example - visitors come with intent to find and buy travel, and they can't find the same search experience on a horizontal search engine like Google. Another example is like.com (portfolio company), that provides visual search for shopping for soft goods like clothing, shoes, handbags and jewelry.  If you search for "red strappy shoes" on Google, here's the landing page you get if you click on the link from like.com.  Yup, those are red strappy shoes all right.  Being focused on a domain and having unique technology to look inside pictures, like.com can offer real value to the end-user, making them happy, which makes the Big G happy as well.  And, oh yeah, the user who gets what they're looking for is likely to buy those red strappy shoes, which makes the merchant happy too!

I think Microsoft got Desktop Search right!

Despite occasional bursts of organizational frenzy, I've never bothered putting my emails and files into neatly structured folders - too much work and too damn inflexible.  So of course I was an early and happy adopter of Gmail and Google Desktop Search, with at least one browser tab open for Google Desktop at all times. But with my recent upgrade to Vista, which comes with integrated Windows Desktop Search, I've actually found myself switching away from a Google product to a Microsoft product.  Apart from the obvious in your face nature of the search box - a data point, if any are needed, that search has become a game of distribution - Windows Desktop Search integrates much better with Outlook, and my Inbox is mostly where I'm searching anyway.

A good job from Microsoft, and an encouraging sign that they'll be a credible foil for Google for a long, long time.  Better for both companies, for consumers, for startups and for investors!

Whither Social Shopping?

Matt Marshall at VentureBeat ponders the future of social shopping by comparing Like.com (portfolio company) with ThisNext. Matt's conclusion is that

comparing ThisNext with Like.com seems premature. Has Like.com really killed social search? Yes, probably, in the short term, if you’re only looking at the numbers. But social shopping search and shopping technology is still in its infancy, and we haven'€™t given it a chance.

Online retailing is - conceptually - very simple.  You're either very early, smart and big, like Amazon.com, and have a brand and a destination that people just remember to visit.  Or you buy traffic from search engines, either horizontal ones like Google or vertically focused ones like Shopping.com and Like.com.  Actually even Amazon buys a lot of its traffic, because people are conditioned to use a search box as the starting point for any web activity. 

Shopping search engines have worked well for hard goods, with buyers driven by price and selection; Like.com is transforming shopping search for soft goods, where appearance and taste matter more - you won't buy that ugly shirt even if it's half off, will you?  The benefit of Like's visual  search technology is reflected in better search results, implying more satisfied users who buy more.  Technically very challenging, conceptually very simple, and financially very rewarding. 

Social shopping sites like ThisNext and Stylehive are looking at offline discourse about products - the window-shopping stroll through the mall with your friends, the conversation about dress styles by the water cooler - and trying to take it online.  This cuts search engines out of the picture, because you will presumably stumble across products in the course of your online life, most of which is not lived in search engine.  I really hope this strategy works, just as I root for any strategy that relaxes the current strangle-hold that search engines have on monetization on the Internet. 

But there are two huge challenges to be overcome by the social shopping sites  Search engines monetize fantastically well because they capture intent and get paid at the point of capture; user intent is missing in social shopping, which means that user attention and user dollars might be missing as well.

The second challenge for social shopping is that users spend most of their online time on social networks, which also have a monetization problem to solve.  If Facebook owns your eyeballs, your social graph and its tastes, won't it become the ultimate social shopping site?  Beacon is a step in that direction, after all.

Matt's article ignited quite a controversy, and not just about social shopping.  Fred Wilson used it as the centerpiece of an article attacking the journalistic standards of pro bloggers, so of course Techcrunch had to defend its tribe of journabloggers and attack Fred.   Here's a juicy bit from Mike Arrington:

That’s no apology, Fred. An apology would include you admitting that both posts were well researched and well written pieces. And that it was wrong to attack the reputation of these writers just because the conclusions reached by them were different than your own.

I expect we haven't seen the last of this  Munjal Shah, CEO of Like.com, adds his two cents, staying clear of issues of journalistic ethics and staying focused on the numbers of his business.

Should Entrepreneurs go to Venture Conferences?

Greg Costikyan (game designer, entrepreneur, blogger) writes a tongue-in-cheek look at the entrepreneur's experience at Venture Conferences.  You know, the ones which promise you a moment in the sun pitching to a legendary pantheon of VCs ...
Here's an excerpt which had me nodding my head  ...

From an entrepreneur's perspective, the supposed appeal to the venture conference is this: I'm pitching to a room containing maybe 200 people, all interested in venture investing, and even though there's a fee attached (and maybe travel and a hotel room), and even though it's a couple of days out of my (and maybe my senior staff's) life, it's a more efficient way to reach a lot of potential investors at once!

Right?

Well--no. That room of 200 people is maybe 25% other entrepreneurs waiting their turn or listening to other pitches to get a better sense of how to polish their own, and maybe 50% service folks who actually want to sell you stuff, and maybe the other 25% are investors of one kind or another. Of whom the vast majority would never invest in whatever it is you're pitching. And of the handful who remain, almost all are so junior that unless they go back foaming at the mouth with excitement, it doesn't really help.

I've been wondering how long entrepreneurs will keep paying up to be one of a crowd pitching when they could far more easily research relevant investors online (that's one of the many benefits of LinkedIn and, now, Facebook!) and wangle a one-on-one meeting instead.

This bit had me ROTFLMAO ...

The basic problem with the venture conference panel is that the conditions under which they are created mitigate against anything of the slightest interest ever being said. They exist to motivate the attendance of VCs, who may be flattered to participate; to reward service firms for contributing money (by allowing them to provide the moderators); and to attract the interest of entrepreneurs, who may reasonably be expected to find what potential investors say of interest. But the choice of topic is inevitably anodyne ("Emerging Trends" -- can't pass that one up!), and since the moderator is from a service firm, which has an interest in sucking up to both investors and entreprenuers, he is extremely unlikely to ask challenging questions, and is likely to stick to the equally anodyne. E.g., "Which is more important when you're looking at a company--the finances or the team?" -- a question at this actual conference, to which the only honest response is "Which are you, a moron or an idiot?"

That's why my first conference this year is GDC in mid-Feb ... conferences of, for and by practitioners tend to be far more rewarding than those catering to investors.

Will Google Spawn a Mafia - Like PayPal?

No, this isn't a post about La Cosa Nostra.  But many recent articles such as this one in the New York Times talk about the PayPal Mafia - entrepreneurs who emerged out of PayPal - and the potential for the fraternity of ex-Googlers to invest in, and found, the next wave of innovation in Silicon Valley.

Implicit in such speculation is the belief in a contagious, high-tech version of the Midas touch - Google was so successful in such a short time that the Kool-Aid in the company cafeteria must have been liberally dosed with the entrepreneurship gene.  After all, PayPal (in which we were the first institutional investors) alumni went on to start or invest in high-profile companies such as Facebook, LinkedIn, Slide (portfolio company), Yelp, Geni, YouTube ... with its crop of young, affluent and connected techies who know no boundaries, shouldn't Google spawn even more of the next generation of successful startups?

My answer is a resounding ... maybe.  Yes, there is a very real PayPal mafia, but there isn't an equivalent eBay mafia - and eBay invented C2C e-commerce and bought PayPal.  Nor is there a Yahoo! mafia or an Amazon mafia, even though both companies invented and dominated large categories of the Internet.  Was PayPal really that special, and can Google be similarly special?

PayPal, unlike so many of the large Internet players, was always an embattled company.  They morphed their business plan dramatically along the way, struggled initially to exist as a value-added service on a hostile platform (eBay) that had a competing offering, were attacked by all manner of fraud artists, faced periodic state and federal investigations, and went public in one of the most trying times in recent US economic history.   PayPal's primary problem was always survival, and being nimble while dealing with dinosaurs was crucial - just as it is for most startups. 

Google's journey to eventual blockbuster status wasn't all smooth sailing either,  but after the first couple of years, when Google had figured out the CPC model (with some help from Overture :-)), Google's primary focus was on scaling - and that's usually a problem that emerges late in a startup's evolution.  Most startups don't have the luxury of dominating their target market (Search) while also running a printing press spewing forth hundred dollar bills (Adwords).

PayPal was very successful, but raised a lot more money (~$250M) than Google ($40M) and exited for far less ($1.5B vs. Google's current market cap of $180B).  The average PayPal employee got a whiff of wealth, but hardly the tens or hundreds of millions that so many Google veterans pocketed thanks to Larry and Sergey's generous options granting policies and the lofty valuation of Google by the public markets.  And while it takes guts to climb out of poverty and build a business, it takes even more to sustain ambition and hunger when your biggest problem in life is dodging calls from wealth managers anxious for your business.

Because PayPal was acquired by a much bigger company, there was a pretty rapid exodus of top talent, many eager to prove that PayPal's considerable success was neither a flash in a pan nor enough of a showcase for their abilities.  PayPal was still relatively small and tight-knit as a company when it was acquired, and the smart techies leaving PayPal all knew each other well.  When they ventured out on their own, these entrepreneurs had a cohort of investors, co-founders and advisors who were aligned in attitude, experience and timing, a fortunate coincidence indeed.

Google's success continues to this day, seemingly without bound, and the incredibly smart, talented, rich and connected folks who choose to leave are on a very personal timeline, usually dictated by options vesting dates; and Google tries hard to keep them by offering a mix of professional and financial blandishments.  Departures take place in a trickle, therefore, and not en masse, and given Google's size, connections within alumni are necessarily looser than at a smaller outfit like PayPal.

Startups need to start out laser-focused, often on ideas that seem quite small and all about the here and now rather than the distant and transformational. I wonder if after helping build a colossus like Google, its alumni will be satisfied working on startup ideas that of necessity will appear tiny in comparison. To their credit, many of the first few Google progeny (e.g., FriendFeed and Weatherbill) seem delightfully concrete, although no doubt with ambitious expansion plans - I sure do hope this trend continues, although I've seen my share of (unnamed!) ex-Google companies that are venturing far beyond the realms of the possible.

Googlers do have quite a few things going for them, of course - the Google selection process was skewed towards off-the-charts-smart people, they did well in a decentralized meritocracy that valued engineering talent much more highly than competitors such as Yahoo or eBay, and Google's policy of allowing employees to spend up to 20% of their time on pet projects has let hundreds of them experience the full lifecycle of conceiving, developing and launching a new product or service. 

A few of these services succeeded, but I suspect the most valuable lessons were learned from the many more flops and Googlers imbibed a culture of failing fast and objective metrics-driven decision-making.  These are essential ingredients for entrepreneurial success.

So PayPal was a very special place,an incubator of talent unrivaled in decades.  But there's enough going for Google that it just might become #1 in that contest as well.

Making Money from Facebook Applications

I was fortunate enough to moderate a panel at CommunityNext this weekend on the topic of Monetizing Applications on Facebook; my panelists were Keith Rabois, VP of Business Development at Slide (portfolio company), Scott Rafer of Lookery, and Murtaza Ali of Peanut Labs.  Tens of thousands of developers have jumped on the Facebook Application bandwagon, but precious little attention has been paid to how the system will generate any money.  Net new money, not recycled VC money.

Slide has enormous reach and has focused on advertising based revenue models so far, mostly selling on a CPM basis.  Most social network advertising sells for very low CPMs (10-20cents), but some Facebook applications have been buying expensive impressions or installs at prices an order of magnitude higher than that.  Perhaps this makes sense for apps such as travel where there is an effective monetization engine on the back end, but we concurred that buying users doesn't make economic sense if the only way to monetize them is advertising - buy/sell spreads on users are unlikely to persist, and social networking is likely to remain a CPM wasteland, with one caveat: apps and widgets that are embedded in profile pages get fare more user attention than banners or AdSense links on the sidelines, and could well fetch much higher eCPMs.

An interesting twist on advertising is to use Facebook data to target the same users on other sites with better conversion rates.  Apps can share anonymized user data with a company like Lookery and, presumably, share in the upside when Lookery monetizes the data elsewhere.  Rumor has it that Facebook is exploring such a product themselves.

Peanut Labs and Social Media enable an alternate monetization strategy - in apps which have an internal points system, users earn points by taking market research surveys served up by Peanut Labs or Social Media, which share the hard dollars paid by their customers with the app developer.  Murtaza claimed that they earned about $250K in September 2007 on Facebook apps alone, of which half was shared with the app developers.  The user was left with, well, peanuts :-)  And a fond hope that the peanuts given them by Peanut Labs are actually worth something.

Which brings us to virtual currency and virtual items. Facebook has had virtual gifts for a while, but the multitude of gifting applications, such as Free Gifts, have made the bottom fall out of the market.  It's hard to charge money for virtual items in an open system, price being directly proportional to scarcity.  But perhaps Slide or RockYou, with their huge reach across social networks and ecosystem of apps and widgets, will be able to create a thriving virtual item economy?

 

 

Success factors for Facebook apps

Nisan Gabay of Startup Review has a great post on the factors which make Facebook apps successful - and the most popular apps are so big that success is now measured not just in terms of users but in percentage penetration.  For instance, Top Friends from Slide has about 25% of all Facebook users!

Finally - Widget Viewership Metrics!

comScore today released the first ever metrics for widget viewership with the launch of "comScore Widget Metrix", a new service to track the usage of widgets across the web.  The numbers are impressive: 40.3% of North American Internet users in April 2007 visited a website with an embedded widget, and the top ten widget providers worldwide had 177M unique viewers.  The top widget provider worldwide was Slide.com (a portfolio company) with 117M uniques, and competitor RockYou was second with 82M uniques - considerable overlap in the visitors, of course.

The existence - and magnitude - of these numbers is good news not just for widget developers but also for advertisers and social networks.  Advertisers now have independent validation of the depth of user engagement with compelling widgets, and brands and widget providers will now have another lever to prod social networks to enlist widgets to more creatively monetize the traffic they generate.
 

Hiring in the Valley? Not So Fast!

The San Francisco Business Times has an article on the difficulties even a well-financed, fast-growing company like Slide (a portfolio company) has in hiring top-notch talent in Silicon Valley.  Aside from the fierce competition among startups, there's the giant sucking sound of Google hoovering away engineers lured by free food, free vanpools with free WiFi, and, last but not least, a stock price defying the laws of gravity.

Reminds me a little of the dot-com boom of 99-00, with one crucial difference.  Back then, for those of my readers old enough to remember, anybody with a weekend's acquaintance with a Java programming handbook could land a lucrative programming gig.  Today, though, companies will pay well for top talent - but only for top talent.  I don't see much dilution of the talent pool; CEOs know that the only growth that matters is in users and revenue - building monuments to their egos, whether they be magnificent buildings or bloated but mediocre armies of engineers, will more likely lead to failure and red faces than the success which awaits the nimble and the humble.

Self-promotion alert: yours truly is quoted a few times in the aforementioned article :-)

The Power of Facebook

I'm very new to Facebook, but in my few weeks on the site I've been blown away by the virality and addictiveness of Facebook.  I was an early beta user of LinkedIn, and have been on MySpace off and on for a couple of years; anecdotally, the rate of growth of my network, as well as the rate of invite acceptance, is astronomically higher than on any other social network I've been a part of.  Not just in the echo chamber of Silicon Valley, but even among late adopters of the digital lifestyle in small towns in India.

And early evidence indicates that Facebook's platform strategy has been a resounding success.  Jeremy Liew has a great post on this topic, and I've been seeing this myself with apps from our portfolio company Slide. Slide's Top Friends app, for instance, added more than 1M users in the past 24 hours to become the top application on Facebook.  That's hypervirality!