Andrew Chen Archives

As a followup to my previous post, "5 things that make your social network monetize like crap" there’s been some recent news about the Google financial results for this quarter related to social networks.

Here are some excerpts, the first from Techdirt:

In the discussion, what we agreed on, was that the social networking sites had done a good job in doing an "upfront" monetization, with MySpace getting a guaranteed ad deal from Google and Facebook getting a guaranteed deal from Microsoft. However, all the details suggested that on the backend things were pretty ugly. It’s not hard to figure out why.

Ads work on Google because people are looking for information. They do a search, and if the advertisement shows information that helps with the query, that makes everyone happy. However, when it comes to a social network, usage is quite different. People aren’t looking for information about products — they’re looking to communicate with friends. In that environment, ads are seen as an intrusion — which is the exact opposite of ads in a search world.

… and the second from CNet:

Chief Financial Officer George Reyes referred to "a few AdSense partners" to whom Google is required to make guaranteed payments. "We have found that social-networking inventory is not monetizing as well as expected," he said.
Under further questioning, co-founder Sergey Brin said the company was disappointed in experiments it had run on some of the approximately 20 social networks it works with, which include MySpace and its own Orkut.
"I don’t think we have the killer best way to advertise and monetize social networks yet," Brin said. "It’s a big opportunity because it’s so much inventory."
MySpace executives were not available for comment Thursday night and a Facebook spokesman did not return a call seeking comment.

Worth thinking about how these monetization issues ultimately impact the growth potential of the social network industry. Absolutely these are some tough problems.

Quick thought on MySpace “News Feeds” – it seems as though right now, there’s a huge amount of pressure to copy a significant chunk of Facebook’s differentiated features, particularly:

• Newsfeed
• Self-serve advertising
• App platform

I very much wonder how useful it is to blindly copy Facebook’s features, and treating the audience growth issue like it’s a technology problem.

After all, consider the Newsfeed, which is a really interesting feature – it seems to mainly work in the case where you have a lot of personal friends where you care if they updated something in their profile, etc.

But in the MySpace case (and in fact, probably for all social networks who embraced the "open friends" strategy), it seems as though you’d end up getting a lot of news from people who you really don’t care about. Because it’s fun to be Barack Obama’s friend on MySpace, but it’s not as common as Facebook.

So the question is, when does it make sense to copy those features? Or are there situations where it’ll very much disrupt the community because it’s taking something that doesn’t belong and trying to integrate it?

Just something I’m thinking about…

Yahoo’s woes
Here’s an interesting excerpt from a recent NYT article on Yahoo’s troubles: Yahoo’s Vision-Goes-Here Strategy.

… Mr. Yang and Ms. Decker’s strategy is essentially “vision goes here.” They want to be the “starting point” for users on the Web. They want to be the “must buy” for advertisers. And Mr. Yang said he would assume an “aggressive investment posture.”

The only thing missing from that is the substance. Why would users start at Yahoo? How are advertisers going to find Yahoo superior? And what will the company invest in?

The fascinating part is that Yahoo absolutely *is* the starting point for many users on the web, particularly when it comes to e-mail, IM, and news.

The problem is bridging the gap between two "modes" of user behavior:

Communication:

• e-mailing
• IM
• news
• entertainment
• etc.

to:

• search
• shopping
• reference

… the latter which are better for monetization, because ultimately it combines motivation and intent with sending traffic off of the site.

Sound familiar?
Of course, this problem exists not just within Yahoo, but for any company that focuses on a high-traffic, high-growth area which has poor monetization.

This includes companies like:

• Hotmail
• Geocities
• MySpace
• Skype
• Facebook
• etc.

All of these companies ended up with big valuations, but never made the jump to monetizing the way Google’s able to. In Yahoo’s case, this is particularly depressing because the core of the company was always a place to jump off to other sites via a directory – but now it’s a communications hub.

I think the woes of Yahoo are one in a long string of companies that have had huge traffic, but the wrong KIND of traffic. Question is, who will figure out how to cross this gap? This is probably the single most important monetization issue facing the social networking and UGC industry.

History of MySpace’s relationships with widget companies
Let’s review the history so far:

• Dec 23rd, 2005: MySpace blocks YouTube
• Oct 2nd, 2006: MySpace blocks Stickam
• Feb 26th, 2007: MySpace blocks Imeem
• April 10th, 2007: MySpace blocks Photobucket videos

MySpace was both hostile and inconsistent towards outside widget companies, and seemed to mostly treat the outside companies like they were stealing users without creating value.

Along the way, the company also did two interesting things:

• Rewrote all links as msplinks.com
• Removed outgoing links inside widgets

… both of which seemed like hostile actions designed to thwart the viral growth of widgets.

Then the world changes!
Then of course:

• May 24th, 2007: Facebook announces Facebook platform

… and then, as a response, finally MySpace announces their own platform – with an amusing quote from their new COO Amit Kapur:

Yeah, I think ultimately that may be an area where we are able to
differentiate. If you look at the past, companies like Photobucket and
YouTube did contribute to the success of MySpace. They were continuing
to build on the user experience in ways we were not focused on.
Philosophically, we want to make that easier for companies.

It really tells you how fast the tech world can change.

In 6 months, the largest and meanest gorilla on the block goes from actively trying to kill developers building on its platform to embracing and encouraging them to make money. Hilarious!

Of course, it took a worth competitor to emerge that showed that you can drive a ton of value (and valuation!) from opening things up and thinking strategically and long-term about your business.

Who else can learn from this?
It strikes me that there are many situations in which companies are being treated like platforms, but don’t realize that’s what they are. And as a result, they end up being very aggressive against developers when in fact they should probably be embracing them and trying to figure out the next generation of their business.

These include activities like:

• Startups scraping e-mail addressbooks to find peoples’ friends and the larger companies getting annoyed
• Companies WANTING to syndicate ads (or near-ads) like music videos, movie trailers, etc but unable to do so because the content isn’t embeddable
• People crawling flash games online (aka stealing flash games?) and posting them somewhere else
• Companies "spamming" the Google index using SEO techniques

All of these different scenarios are ones in which large companies are being leveraged by lots of smaller companies. And of course BigCo typically doesn’t respond well.

Question is – what’s the best way for them to start taking advantage of this developer base that wants to engage their distribution platforms, their content, and their audiences?

The sad answer is – we’ll probably never know until these companies are weakened and are significantly challenged enough by competitors to reverse course as MySpace did.

Interesting analysis and speculation: F8ce the music. The articles includes this article, which shows that a number of top Facebook apps have now fallen from their peak:

It strikes me that a lot of these apps have now gone through most of the "carrying capacity" of the Facebook audience. Everyone’s seen them now, and the novelty has worn off a bit.

Definitely worth thinking about more – how do you create apps that are both viral and are good at driving engaging experiences that last a long time?

My sister Ada sent me this interesting article: Is the Tipping Point Toast?. Thanks sis!

Marketers spend a billion dollars a year targeting influentials. Duncan Watts says they’re wasting their money.

Worth a read.

(Stolen from Uncov)

After the Techcrunch Bump
Josh Kopelman writes a great post titled "After the Techcrunch Bump" where he talks about some key quantitative measures of how your site is doing:

• usage growth
• viral coefficient
• engagement level
• cohort analysis

Absolutely worth reading in more detail, particularly for folks who primarily focus on blog "buzz" or PR in order to generate traffic…

My position on this topic
I’ve written on some similar issues, taking a more extreme viewpoint here, where the point is to say, forget Techcrunch, just focus on metrics and building your business.

My take on it is that if you can quantitatively measure user acquisition, retention, and monetization to your site, and optimize towards that, then the entire decision whether or not you want to talk to press and bloggers is an independent decision. What are the advantages and disadvantages of making what you’re doing public? And for the most part, until you’ve perfected your traction, why alert your potential competitors on what you’re up to?

Instead, it strikes me that one of the smartest things to do is to keep a low-profile, gather millions of customers, and go from there. The caveat to this being that you can often find interesting partners, investors, and employees by going public.

How quantitative are consumer startups?
The heart of the issue in Josh’s post, however, is that consumer internet startups are very very quantitative. Another smart VC, Vineet Buch from Bluerun, recently remarked to me over coffee:

It’s a myth that enterprise companies are more metrics-driven than consumer companies – in fact it’s the other way around.

I totally agree with this, because when you have millions of users driving billions of pageviews, that’s actually trillions of datapoints to track, analyze, and act upon.

Companies like Amazon and Google, with real revenue streams attached to their businesses, track a ton fo data and optimize. But for a small startup, it’s often overlooked even though the process of validating user-traction is often the #1 job of a seed-stage company.

An alternative approach
I’d like to see more startups actually start with a wonderful, artistic idea for a consumer product, but then spend the first 4-6 weeks working on an analytics architecture around that idea. It’s something that very few companies are willing to do – stats often ain’t fun – but is obviously hugely useful.

Know of anyone doing this??

PS. Cutest thing ever!

Max Levchin recently launched his blog – which I’m sure will be excellent – with an article called, "How to successfully launch a social networking development platform." It’s absolutely worth reading, so go click here and read it.

If you’re lazy, here’s the quick summary:

1. Create a feeling of technological openness.
2. Treat developers equally, but leverage the best ones by letting them closer in.
3. Plan and manage a community, and introduce a community manager early
4. Shift the support/documentation load onto the early developers.
5. Respond very quickly to platform issues, and take the early scaling problems seriously.
6. Emphasize the money-making nature of the platform.
7. Make your campus a place that developers very much want to visit.
8. Pre- and over-communicate policy changes and make major changes with at least the perception of open debate.
9. Make the #1 measurable goal of your PR team the amount of coverage that successful (or just interesting) developers get.
10. Hold frequent developer events and invite leading developers to speak at those.

Looking at this list, it strikes me that there are really two different types of developers, the for-profit type and the for-attention type. Different tools (listed above) appeal to the different groups.

For-profit developers
These days, there are many companies in the Bay Area that are building app companies, completely without destination sites. Many of these have gotten angel funding – I personally know of at least a dozen and a half. And a few have even gotten venture funding.

For these companies, particularly the ones with venture funding, there are significant issues to overcome:

• How do you generate significant revenues? ($50MM+/year)
• Is Facebook ultimately friend or foe?
• How can I beat my other for-profit competitors?
• Does more PR in this space generate higher valuations or advertiser interest for me?
• etc.

So as an archetype, you can think of these teams as highly experienced serial entrepreneurs with lots of money, offices in SF, staff of dozens, millions of app installs, etc. Kinda like Max ;-)

Thus, the list that Max writes include some of these issues – openness to make dependent companies feel good, money-making nature to appeal to revenues, etc.

For-attention developers
But compare this to the other group, who are the for-attention developers – or as you might call them, enthusiasts. These folks have vastly different concerns:

• Do people like my app?
• Am I meeting interesting people via my app?
• Do I have enough money to pay the server bills?
• etc.

This group is interesting because I think they are the ones actually hanging out on the Facebook Developer Forums ;-)

I was also able to meet a bunch of them at the recent CommunityNext conference on platforms, where we had dozens of Facebook teams present. They seem to be very small teams – like 1 or 2 – who end up working on a cool Facebook project because they think it’s interesting. In these cases, they’re almost as interested in the technology as an art, rather than being successful. Thus, the process of app creation is almost as fulfilling as being successful itself – that’s why I’m finding many hackers trying out new apps all the time.

For this group, I’d argue that fame/status/respect are more important than money. While it’s fun to think they can make a buck, the more important part is that they have X million users using their Pokewallgift application, and it makes them feel good. Building a $50MM revenue stream or the strategic implications of Facebook’s moves are less important than to the folks who are really trying to build businesses.

Viral channels and scoreboards?
One thing I found curiously missing from Max’s list was viral channels – obviously it’s very easy to build social websites without easy ways for apps to spread themselves. In fact, the OpenSocial spec has a newsfeed, but no invitation and messaging system, both of which can be critical for growth. After all, what’s the point of building on someone else’s platform if you don’t also have access to their audience for growth?

Also, as I’ve commented to friend before, being able to generate a "score" of millions of users is critical, even if those millions of users don’t actually equate to huge revenue. IMHO, one of the biggest mistakes MySpace has made in this has been not publishing a scoreboard with how many millions of users have whatever layout site, music widget, virtual pet, slideshow, etc. I know for a fact there are internet companies with millions of widgets on MySpace, but the press can’t report on them, and people don’t talk about them.

Here’s my Feedburner graph of # subscribers over time:

As you can see, in the first 12 months, this blog slowly built up (linearly, pretty much), an audience of about 1000 readers. The big spike that you see is when Scoble linked to this blog – since then, in about 2-3 weeks, the blog picked up its 1500 subscribers.

The traffic was actually more than I’ve ever gotten being on Techmeme or Hacker News, which tells you how much traffic he gets. Pretty amazing!

I love this headline from Valleywag: Quantcast gets $20 million in funding — and more attention than it deserves. Haha!

What is absolutely true is that the analytics space has been changing very rapidly. Traditionally, there’s been one main axis for analytics, which depended on whether or not you were publisher-leaning or if you were advertiser-leaning:

publisher leaning <—-> advertiser-leaning
omniture/google <——–> comscore/nielsen
(pardon the ascii art!)

Publisher-leaning analytics
Under publisher-leaning bucket, you have companies like Omniture, Coremetrics, Urchin (acquired by Google), etc. These are companies that primarily focus on the usual things, like:

• pageviews
• conversions
• unique users
• etc.

And as to be expected, these analytics providers charge on a monthly basis to publishers, who take the data and publish it in media kits as well as for internal purposes.

Advertiser-leaning analytics
On the other side of the coin, you have companies like comScore and Nielsen, which traditionally help advertisers figure out CROSS-site metrics, so that they can figure out where to buy advertising. As a result, many of the stats they offer include things like:

• demographics (age/gender/etc.)
• audience composition
• cross-site overlap
• etc.

Their primary goal is to answer things like, "I’m trying to reach teenagers – which sites should I be talking to, for buying advertising?" This cross-site process has necessarily ended up with the panel-based measurement that everyone freaks out about ;-)

Other interesting divergences
And of course, advertisers and publishers have a little bit of an adversarial relationship. After all, if advertisers can "find the needle in the haystack" before the publisher does, they can buy crappy advertising inventory that’s in fact, really targeted inventory. And similarly, the publisher would love to obscure some of their ad inventory as to force the advertisers to buy both good and bad advertising at a premium.

So across a variety of tools beyond analytics, you have mirror-versions of each tool. A good example of this: A lot of publishers have their own internal ad servers like Doubleclick’s DART, but a lot of advertisers use so-called "third party ad servers" like aQuantive’s.

Analytics by channel
Similarly, I’ll expect that any new analytics product – be it in video, social network analysis, widgets, etc. – to always have an advertiser-side version of the company as well as a publisher side.

This is where things are very interesting for Quantcast. Seems as though the highest margins for them are going to be facilitating the ad-buying process, yet they are also having the publishers do a pixel implementation to track more specific information. Question is, long-run are they more of an advertiser-leaning company? Or a publisher-leaning company? Only time can tell…

Useful article from Advertising Age called "Recession Hits: What It Means for the Ad Biz":

Merrill Lynch last week became the first major U.S. investment bank to
declare the U.S. economy in recession. Two days later, Goldman Sachs
jumped in, predicting that the economy would enter recession during
2008, if it wasn’t there already. [snip]

So in order to find out what that means for the marketing world, Ad Age
asked leading industry executives what they think a recession means for
the business in 2008.

Their reply: It might not hurt in the short-term, but if things don’t tick up, gird yourself for a tough 2009.

Worth reading.

Trends within Second Life
The full article is on Massively here, which is a great news site on MMOGs. (I originally saw this article in Charles Hudson‘s Google shares, thanks Charles!)

The article goes on to summarize a number of trends within Second Life:

Total signups increased by 529,224 (4.73%) in December compared to a
growth of 605,095 (5.72%) in November. New-user retention to 90 days is still about 10% according to Linden Lab, having apparently remained more or less unchanged over the last 12 months.

Overall age demographics continue to trend towards us crusty oldies.
74.17% of the population is 25 years and older – a slight gain from
November, and the older users spend far more time in Second Life
than younger users with those over 44 years old continuing to be the
heaviest users on average, and teens and under-25’s spending the least
time.

Definitely worth reading.

What’s the right activity metric?
Over time, the # of registered users has become a pretty meaningless metric to me – I mean, big is good, but given that some sites can generate users but churn them right away, it’s not so useful. Similarly, the proliferation of Facebook apps and MySpace sites makes it so that large numbers of "uniques" can be generated without effort also, but these users are not worth the same as the others.

Peak concurrency is something I read a lot about in the MMOG world, yet it seems very much rooted on capacity planning :) It is certainly a different view though, and one that has its usefulness.

Probably worth writing another blog post one day, about different user activity metrics…

Dear blog readers,

If you have recommendations for designers (particularly in SF), please e-mail me at: voodoo [at] gmail.

And of course, if you want to introduce yourself for any reason, shoot me an e-mail anytime ;-) I try to respond to every one…

Thank you!

Amusing observation overheard at dinner last night:

"PlentyOfFish? If I hear about another ugly website that hits millions of users, I’m going to lose hope for good design"

:)

Another point that will stay with me is one of the folks there saying that he’ll avoid designing consumer websites with glitter, hearts, etc., by never making websites for the mainstream audience. Very perceptive and consistent, imho – it’s the folks that want to design for the masses, but force an unappealing top-down aesthetic that I get annoyed by.

Will people pay for anything on the internet?
A recent article on GigaOm:  O2 Offers Napster For Free! Big Question – What will people pay for in the future? with the quote:

It is our contention that in the future, we will only pay for broadband access. Voice, Video or whatever is going to become part of the “access” offering.

There’s been a similar discussion revolving around PlentyOfFish.com, which is turning online dating from a paid subscription model, to a free ads-supported model.

However, a colleague of mine Basem Nayfeh once remarked, "It doesn’t make sense for everything on the internet to be free and ad-supported – SOMEONE has to pay the bills!" (And note that Basem is the CTO of one of the largest and most well-funded ad startups in the world right now)

Let’s dive into why he says that…

Are you a free site? Or not?
One of the key drivers of Web 2.0 is that the sites you are building are likely free, and ad-supported. This is great in theory, since customers like things that are free. However, imagine a world where EVERY internet site is free and has advertising revenues. In that world, every site would be buying traffic from every other site, and because it’s not a frictionless exchange, money would be lost from the system and revenues would eventually go to zero.

As a result, the only way to prop up a system where it’s just ad-supported companies buying from ad-supported companies, you need a bunch of external money like venture capital. But then you’d just end up with the Facebook installs ecosystem (haha! Just kidding Mark!)

So it’s pretty clear that you need some amount of external money flowing into the ad buying, and thus, let’s look at what an old guy thinks: Newton’s Third Law, which relates to online advertising (I swear!):

To every action force there is an equal, but opposite, reaction force

I cite this because the same holds true in online advertising – Newton did not say this by the way:

For every online advertisement, there is a publisher and there’s a buyer

and thus its corollary:

For every free ad-supported site, there’s some number of transactional sites that support it

And in fact, this statement probably goes even deeper down into the "type" of advertising inventory that’s growing on a site. For example, even if there’s a lot of advertisers willing to buy car/truck/bus ads, if there’s not enough ad inventory being produced by online publishers that fits into that, then that creates an imbalance as well.

The big question
So as a web entrepreneur, you have to ask yourself a Big Question. In the long run, are you a:

• Free, ad-supported site (an "Ad publisher")
• Or a paid site that buys to get clicks (an "Ad buyer")

Do you know the answer to that? Sometimes it’s not immediately obvious.

Here are some signs you might be better off as an ad buyer:

• It’s hard to get traffic :)
• People don’t use your site every day
• People use it only when they are "in-market" for some number of weeks
  • (Dating, shopping, researching homes, researching schools, etc.)
• It’s hard to get people to come back for multiple sessions
• People spend a lot of money in your category
• It’s unlikely people want to tell their friends about your site, or that their friends are unlikely to be "in-market" at the same time

Similarly, these are signs that you are better off as an Ad publisher:

• It’s easy to get lots of viral traffic
• People use your site all the time, all year around
• People don’t spend much money in your category, or there are lots of competitors
• There’s no "point" in using the site – sometimes it’s time-wasting activity like entertainment

Where’s the gray area?
Obviously there’s a lot of gray area for this – in many ways, I’m sure that PlentyOfFish is as much a communications site as a dating site. Or the same as MySpace. Similarly, for shopping or music or photos, a lot of these sites are as much social entertainment experiences as much as they are transactional experiences. That’s why you have transactional sites like Shutterfly, but also ad-supported sites like Flickr.

So for each entrepreneur’s business, it’s worthwhile to evaluate it in different ways – but please keep in mind, just because your competitors charge money, it doesn’t mean you can go free and succeed. Because not everything on the internet can be free :)

Are you a product fanatic?

If you are, you might recognize yourself in the following:

• 10. You think MySpace is soooo ugly
• 9. You only use the best products:
  • Macbook Pro
  • Mac OS X
  • Firefox
  • Gmail
  • Google Maps
  • Facebook
• 8. You never use the popular products:
  • Dell
  • Windows
  • IE
  • Yahoo Mail
  • Mapquest
  • MySpace
• 7. You try out new sites and judge them based on their features and functionality
• 6. You assess a site’s quality based on if it’s written in Ruby versus Java
• 5. You like things simple, functional, and uncluttered – cuz what else would you want?
• 4. You think your startup’s “secret sauce” is in the technology, or platform, or programming language (LISP!)
• 3. You don’t require users to give you their emails, because it’s not what you’d want as a user
• 2. You say things like, “Just build something people want”
• 1. You secretly worship Steve Jobs :-)

Do any of those sound like you? Maybe it’s not even that secret that you worship Steve Jobs?

 I confess that I personally fulfill many of the characteristics above, for better or worse. And I see significant dangers in becoming too fanatical about product, for the following reasons:

Product fanaticism != User fanaticism
An overfocus on product – meaning the features and functionality, or the underlying technology – can cause friction with motivations from the user. The reason is that product people often feel a great urge to be “product artists” and experiment with new features and interactions, when it’s not what users want.

In many cases, users want to do things that don’t mess with your view of the world. They want to make things look like Geocities, and muck everything up. They want you to take a bunch of existing products and mush them together, not make a single high-quality product. And the visual aesthetic they respond to changes whether they are teens or moms in the midwest, or seniors, and sometimes the Googley thing to do isn’t the right thing to do.

Product fanaticism overlooks innovations in other places
People who are overly focused on the product may overlook opportunities in innovate in other places than the product. This might be a create new way to make money (subscriptions versus virtual goods versus whatever), or maybe a novel method of distribution. Or you might focus completely on features, but ignore the metrics and analytics you need to drive them.

Would product fanatics have invented Wal-Mart, or Microsoft? Some of you wouldn’t WANT to invent these businesses, even if you could. The reason is that both of these businesses are big, ugly, non-product-driven businesses, and that’s why everyone makes fun of them. But they work because they’ve completely locked the distribution channels, can essentially charge whatever they want, and are great businesses.

If your goal as an entrepreneur is to be successful, it pays to be open to such innovations. What is the Wal-Mart version of your business? What’s something that’s boring, execution-oriented, but could be an industry changer? Think about it.

Product fanaticism can lead to beautiful, empty websites
And the final danger of all of this: It takes a lot of effort and resources to get a website off the ground – you need people, financing, and distribution strategy, technology, hackers, etc. If the focus becomes to ONLY focus on “build something people want” then that attention is taken from the other, critical areas. Now in many cases, it doesn’t matter, because you don’t need to figure out your financing structure right away. But you probably DO need to put appropriate thought into the business model, your distribution strategy, and all that other good stuff.

Otherwise, you’ll end up like a lot of the websites that show up on Techcrunch – cool ideas, maybe even good implementations, but ultimately, you get the traffic spike followed by nothing.

So given the choice between the extremes, which one would you pick?

1. Beautiful, innovative website that no one uses
2. Ugly, copy-cat website that lots of people use

If you’d rather have #2, then you might want to think about taking some known user mechanics that people already love – thus reducing your risk there – and instead innovating in something other than product.

Have fun!

What kind of web analytics package are you running?
Chances are, if you run a site, the only analytics you’re running is Google Analytics, or something equivalent. For most high-level traffic analysis, it’s Good Enough, but for anything else, it Sucks Bad.

The reason is that for most entrepreneurs, they have a small # of issues they care about, when it comes to their traffic:

• Is their site growing?
• Are users sticking to the site?

Problem is, beyond this high-level view, the underlying DRIVERS of these statistics aren’t available. The goal is not just to passively observe what’s going on in your site, but to understand the underlying variables enough to make changes and see what happens.

For example, let’s take a look at viral marketing metrics…

Are female users more viral?
Jeremy Liew recently wrote about the observed phenomenon that female users are more viral:

I recently spoke to the people who run a popular social network and they shared some of their stats with me:

1) In 2007, 56% of total signups were female.
2) Females are 33% more likely to invite friends than are males.
3) Females are 10% more likely to respond to an invite from a female vs. a male.
4) Males are 50% more likely to respond to an invite from a female vs. a male.

Definitely interesting stuff.

And it reminds you that if your business is based on viral marketing, and you don’t know who in your system is likely to be viral, it’s impossible to optimize for increasing your distribution. And that sucks.

Build custom integration to dig one level deeper
What you really want to do is to dig one level deeper in your traffic analytics so that you can start to make changes to the underlying levers. After all, you’re tracking this information not as a passive-activity, but as a way to adjust and optimize your business, right?

If you’re a video site, it’s not enough to just track pageviews – instead, you want to know how many comments are being left for each video. Or how many videos are being watched in a session. Or the difference in these stats between users that churn out, versus users and stick around. Or how many friends an average user has. These are all super-important metrics.

This is all the more important when you are dealing with ads. After all, a pageview is not a pageview, depending on whether it’s US versus international, whether it’s the 50th ad a user has seen or if it’s the 1st, and so on. The most specific your goal is (monetization versus growth versus retention), the more you want to tie your statistics to user-centric stats versus the generic stuff that’s available in any analytics report.

Continuing from my previous post "Do you ever say, ‘MySpace is soooo ugly’" that touched on scrapbooking, I was amused to read about the following scrapbooking scandal that’s been compared to the baseball steriods scandal:

Scrapbooking bloggers called it "Hall of Fame-Gate," naming it the top
scrapbooking scandal of 2007. They compared it to the
performance-enhancing-drug controversies involving major league
baseball player Barry Bonds and Olympic track star Marion Jones.
Someone wrote that Contes was as polarizing a figure as Martha Stewart.

The rest of the article has some interesting tidbits about the motivations behind scrapbooking, which, as I argued before, has a lot of the same hallmarks of the "self-expression" driven MySpace profiles:

The edgier scrapbookers thought of it as an outlet — much like keeping
a diary — in which they expressed political views, decorated pages of
their poetry or paid tribute to television shows like "Project Runway,"
using torn and faded materials not guaranteed to last long enough for
their grandchildren to see.

… and, another data point on the market size of all of this:

As popularity soared, scrapbooking — in all its forms — exploded into
a $2.6-billion industry where enthusiasts young and old, conservative
and radical, grudgingly put aside differences to compete in national
contests, attend global conventions, build blogs, join chat rooms,
create online portfolios, and view YouTube and other online
instructional videos.

[snip]

In 1987, Rhonda Anderson of St. Cloud, Minn., co-founded Creative
Memories, a company that aimed to take scrapbooking to people of all
backgrounds. Creative Memories now has 90,000 consultants who sell the
company’s products in stores, online and in classes they teach in 12
countries. The company earned $300 million in 2005, and slightly less
in 2006 because many people shifted to computer programs to create
digital albums, a niche the company is now expanding.

Definitely worth reading the article if you’re interested in the industry.

My friend Jared Kim recently launched his company: WeGame Launches As YouTube For Gamers.

I’m a huge fan of Jared and what he’s doing – definitely worth checking out. It’s a great exercise in taking a vertical stripe of content/audience which has tons of monetization potential, and building deep hooks into it to generate something high-value and proprietary. Classic "crossing the chasm" material for WeGame to then enter into many other strategic areas.

Congrats to Jared, Naval, and the rest of the team!

As far as overall advertising goes, Internet at 8.7% of ad dollars, whereas TV
(local+network+cable) is almost 75%.

There are two ways to look at this:

• Wow, look how far online advertising can grow!

Or:

• Look how insignificant online ads are compared to other media!

It’s really a combination of the two – it’s true that both online ads has a lot of headroom, but when it comes to ad agencies, the total amount of spend going towards this medium is still small compared to everything else.

This is part of the reason why even if you have an awesome website with lots of branding opportunities, it can be hard to move the needle and get agencies interested unless you have massive scale. It’s definitely something to keep in mind if you are building a social media site with no direct commercial intent in order to attract direct response advertisers.

PS. MarketingCharts.com is a pretty awesome site to grab numbers like these!

Fakesters versus Friends on social networks
On the social networks Facebook/MySpace/Friendster, one topic that people often discuss is "fake friends" or Fakesters. Another interesting, and more nuanced discussion is, "How well do you have to know someone to be Facebook friends with them?" And yet another is people complaining that you’re not *really* friends with people on social networks, you’re just "MySpace friends."

Furthermore, there’s recently been a lot of discussion of e-mail, and how you might be able to leverage your inbox as the foundation of a social networking application. Here’s a blog post by Don Dodge from Microsoft, a NY Times article on "inbox 2.0," GigaOm, and also Mashable. There’s absolutely some really interesting data contained inside your e-mail inbox, especially for the older folks that prefer e-mail over other means (texting, IM, social networks, etc.), and as a result, the data will absolutely be an asset.

Really, this discussion is about a bunch of very fundamental issues:

• What does it really mean to be "friends" with someone?
• What are the dimensions for measuring and representing that relationship?
• How do you represent more complex dynamics, like grouping, subgroups, personal identities, etc?

These are very hard problems, which I will not attempt to solve in
this blog post ;-) Needless to say, these are very complex and subtle
issues, which are not likely to resolved perfectly, ever.

But either way, the $64,000 question is:

Who has the best access to the "right" kind of data to solve this problem?

Is it e-mail? Is it Facebook? Is it the wireless carriers? Who is it?

To give some perspective to this problem – let me take you through the life of a person’s social relationships

(more post below)

The life and times of social relationships
As I mentioned in a previous post, I have been interviewing random people at different life stages (high school, college, and work) in order to better understand social networks and in particular Facebook. If you haven’t read the post, I’d encourage you to: Why your friends list gets polluted over time.

The core of the interviews has been a process in which I ask them the following:

1. Draw me a diagram showing everyone in your life
2. Group them based on whatever makes sense to you
3. Describe how you interact with each group

I’ve talked to about a dozen+half different people so far, and the data has been very interesting – definitely relevant to the e-mail vs Facebook social network debate. In general, there are some very strong patterns, which I’ll cover below, but of course it’s always the exceptions that are the most interesting.

OK, let’s start aaaaalll the way from the beginning. Well, your social relationships when you are first born are very simple:

That’s right, all you have is your mom ;-)

But then, as you grow up, you start to get to know the people around you, like your immediate family:

And for years, that’s all you have. Then one day, you’ll start going to school, and then your social network gets a whole lot bigger, really fast:

All of a sudden, you have a bunch of other kids around you – you make friends with a lot of them, and see a pretty consistent group every day, so you all bond pretty well. Of course, this doesn’t last, because you grow up, and eventually you’re in high school, which has more kids, more complex subgroups, and it may be that only a certain group of your friends from grade/middle school come with you into high school:

In the picture above, the blue squares continue to be people who you interact with frequently, and the gray ones are acquaintances – people you saw every day before, but who you’ve lost touch with. At this point, your entire group of friends/acquaintances/etc may go from a nice big number (like 30) down to a couple besties aka BFF aka best friends forever that you still talk to.

I adjusted the parents down to small squares to reflect the teenage angst that drives kids away from their loving moms and dads ;-)

In high school, you inherit a new, large social network, who are now the most important people in the world to you…

… until college:

Then yet again, you inherit a large group of college friends, including people that were potentially in your high school (but really, it’s quite unlikely). In addition, your group of friends from middle school and high school shrink down, so that you really have a couple.

Because you moved away for college, you never play fetch with Fido, your family dog, except around the holidays :-(

Then, once you graduate from college, you go through the same transitions – first with work, and then a new location:

In the above diagram, you’re moving to a new city after you graduate, and you get a bunch of friends from your new job (yay!) and from the new city you’re living in. Also, one of your old high school friends joins you in the new city.

Then a couple years later, maybe you switch jobs, your dog dies, but your sister moves into your new city:

… and the social network grows on.

Growth and staleness in social relationships
From the diagrams above, you’ve probably already drawn some conclusions, including the idea that:

• Each time you go to a new school, or geography, or workplace, you quickly inherit a large disjoint set of relationships that causes your social network to grow very fast
• However, each new "move" makes your previous relationships become stale quickly as well
• The absolute # of relationships grows quickly, but the actual # of people you might consider friends either grows quickly (at the beginning of your life), and eventually reaches a cap (maybe the Dunbar number) later in your life
• Between the different social networks you move through, there’s always a little bit of intermixing – that is, middle school friends that join you in high school, or high school friends that go to the same college, or any of the aforementioned groups that end up in the same city as you
• As a result of the intermixing, a person’s social network is actually complex – they might be BOTH local friends and a middle school friend, or a work friend and a college buddy. Or maybe it was someone who you dropped off with after high school, but ended up getting reacquainted with in college
• The # of "stale" relationships increases very quickly relative to the actual # of friendships in your network

This discussion ought to also include the ramifications of dating someone seriously and inheriting their social network as well, but I didn’t pry too much into the lives of my interviewees ;-)

The point is, most social networks and your e-mail addressbook allow you to list out your friends, but really you need some more dimensions than that to capture the reality of that relationship, including:

• A notion of an "active" friend versus a "stale" friend, which is a time dimension
• Another might be the actual strength of those relationships, because you might be active with someone at work, but not be very good friends with them
• Similarly, friendships need to have some grouping to them, but they can’t be exclusive groups – instead, what group someone is might be fluid or encompass multiple definitions
• Yet another subtlety is the professional versus personal distinction, and the social protocols that come into those interactions based on the groupings

Once you add in attractive people you’re stalking, or your-significant-others’-friends-who-would-leave-you-if-you-broke-up, and this thing becomes complicated very quickly. Furthermore, users aren’t going to want to fill in all this information – they are going to do something coarser and simpler, like creating multiple accounts. That’s why people have multiple e-mail addresses, to segment their personal lives.

As a result, it’s probably also important to collect all these nuances as passively as possible – if you don’t, then you’re going to make the process of managing these relationships very tedious. Your social network then becomes as fun as a CRM system.

How does this pertain to email versus Facebook?
To return to the original question, does e-mail have the best representation of your true relationships? Or does Facebook? Or does someone else?

My personal opinion from this discussion is that actually *no one* has all the right data yet. The reason is that people choose to communicate to through many different means – some through phone, some through e-mail, and some through social networks. Each communication method means a slightly different thing.

The group of people that I’d call is much smaller than the one that I’d e-mail. The more ways I’ll try to get in touch with someone, the closer I am to them. Trying to say that the most active friends can be measured by e-mail or Facebook ignores the fact that maybe I see someone every day at school, and thus prefer in-person with them.

Until wireless data, e-mail, social networks, VOIP, and all the other data lives in one place, it’ll be hard for any one provider to figure out who’s closer to whom in as nuanced or holistic of a way as people will want. As a result, it’s my opinion that most users will select "best of breed" options at every level of communication, and use them to segment off their complex relationships.

(That is, until Google buys a wireless carrier, Facebook, and everything else on the planet)

;-)

Blogging is considered harmful?
Just saw this in the NYT: Some Brand-Name Bloggers Say Stress of Posting Is a Hazard to Their Health.

Obviously, blogging can be a high-stress thing if you are trying to do it on a regular basis, as a job, rather than I do, which is write whenever it strikes my fancy.

The stats
That said, I sometimes wonder if blogging is worth it from a purely ROI basis. After all, my stats are pretty measly, like so:

VISITS
	Total	49,639
	Average Per Day	443
	Average Visit Length	2:00
	Last Hour	8
	Today	221
	This Week	3,100
PAGE VIEWS
	Total	72,501
	Average Per Day	612
	Average Per Visit	1.4
	Last Hour	9
	Today	342
	This Week	4,285

As you might guess, some of my blog posts take upwards of an hour to write, and might have been percolating in my brain for days. (Compare this to the far more efficient "publisher" model, where you selectively post links throughout the day.)

But after week in and week out of this writing, the fact is, the average # of pageviews on a visit is 1.4. Which means the vast majority of folks come to the blog, read a page, and leave. Some might click once, then leave, but it all averages out.

Is there an opportunity cost?
Compare that to almost any website I might write using the same amount of time – it’d be a path to get to millions of pageviews, rather than a blog which seems like it’d be amazing to have more than a couple thousand per month.

Furthermore, the stuff I write on this blog is pretty inside-baseball. I’m actually amazed that according to the Feedburner stats I have, I actually have over 2000 readers. Strange.

The blogging ROI
These days, I mostly think of having a blog as the same as having a
giant mailing list or CC: line on an e-mail that I might send out. In
that way, 2000 people is actually a really big number. In addition, it’s useful for a bunch of other random things, like:

• Helping me organize my own thoughts and ideas
• Keeping in touch with a large group of professional acquaintances and friends
• Meeting a small # of high-quality people

All of these things, especially when really interesting folks e-mail me, make blogging worthwhile. But sometimes, I still wonder ;-)

Dear lazyweb,

What was the last 9-figure ($100MM+) exit for a New York based company specializing in consumer internet?

For some reason, I’m having trouble thinking of one…

Thanks!
Andrew

Greg Linden writes in a recent blog about his one prediction for 2008: The coming 2008 dot-com crash.

Early January is the time we see many predictions for 2008. I have not played this game since 2006, but I want to chime in this year.

I am only going to make one prediction, but one with broad impact. We will see a dot-com crash in 2008. It will be more prolonged and deeper than the crash of 2000.

I love it ;-) Yet, I’m much more optimistic about the coming year.

So I think things will continue at the pace they are, and even pick up, at least until late 2009 to 2010. Here’s my reasoning:

• The "broader" economy and investors have not been investing as bubbliciously as in the late 90s
• As long as there are big exits from tech startups, investor interest will only grow – this will cause the late majority and laggards (as far as investors, press, etc.) to get excited about the tech economy, and then throw money at the situation
• As there are good companies out there with real revenues, when these companies exit, it will continue an upwards trend in interest. For example, if the tech industry started a slump but then Facebook exited, wouldn’t it only cause renewed interest?
• HOWEVER, as these good companies exit, then all there’ll be left is bad startups (at least percentage-wise) – thus causing more bubble-like trends to occur

The question then, is how many good companies are left in the ecosystem, and how long will it take before they exit? As long as there’s a stream of exits, I think everyone will be happy.

Based on what I’ve seen out in the SF tech scene, there are at least a dozen or more startups with revenues beyond $10MM, growing at a significant rate. And IMHO, their revenues are coming from sources that are not likely to crash in 2008. A large percentage of these will take 2-3 years to exit, and beyond that, bubble conditions might be ripe – but not before then ;-)

What do you guys think??

Great discussion on the overall trend for advertising CPMs: JP Morgan Predicts Display Advertising CPMs Will Rise, But Will They?.

The quick summary in terms of the pros and cons for CPMs increasing:

• Pro: Technology and optimization is improving
• Pro: More ad dollars are moving online
• Con: The amount of available inventory is increasing

My quick take: the problem is not just that the amount of available inventory is increasing – more importantly, this inventory is concentrated in social media, UGC, and all that fun stuff.

Unfortunately, marketers and agencies don’t seem to have figured out a way they can throw a large % of their budgets against this type of inventory. In fact, there are significant cons for this specific set of UGC inventory:

• Editorial adjacency issues: They want to be next to brand-friendly content where they can control – 50% of the pics on MySpace are probably not brand- or family-friendly
• Lack of qualified interest: For properties focused on communication like social networks, there’s very little context from pageview-to-pageview. You don’t want to put up expensive personal finance ads hoping for a transaction, when in fact the user just wants to check out hot girls
• Non-standard ad units: Given the very low clickthrough rates of banner ads on these sites – MySpace and Facebook reportedly have ~0.05-0.2% CTRs, people are looking at video ad units, widget, and other alternatives. However, these ad units are all very new, and marketers that want to put significant budget towards these ads still need to work through a lot of inefficiencies
• Opacity and weakness in metrics: And of course, the lack of metrics that adequately reflect the strengths of in-social-network advertising makes it so that it’s unclear when you are running a successful campaign versus an unsuccessful one. Traditional metrics like pageviews, CTRs, and such probably don’t prescribe appropriate value to folks that really know how to dominate on social networking sites.

In short, I think the proliferation of this type of inventory will continue – Silicon Valley is great at creating this stuff – yet at the same time, Madison Avenue will still take years to figure out how they want to play out this round.

So I’d definitely vote for a decrease in CPMs, maybe even a big one, over the next year, even while overall online ad spending increases dramatically. I just think the $0.10 CPM social network remnant inventory will dilute out any gains elsewhere, but that’s IMHO :)

The story so far

As of the new year, I am finishing up my quick tour of the venture capital industry.

I spent most of 2007 affiliated with Mohr Davidow Ventures, where I started out as an Entrepreneur-in-Residence, and then transitioned into a consulting role about mid-way through the year. It was a great experience – more on that below.

As I mentioned in a previous post, What is an Entrepreneur-in-Residence, the key elements of a program like that are that:

• EIRs come from industry with background in a specific area, in my case advertising
• They are sponsored by a partner to start a company after some time period (let’s say 3 months to a year)
• EIRs may come in with a specific idea, or not – in my case, not :)
• They find team members, incubate the idea, and find funding
• The EIRs will also often work with the investment team to do due diligence on new companies coming in
• There is no quid-pro-quo that you have to get funding from the
  specific firm the EIR is at, and no guarantee the VC will fund the
  business

Of course, I wanted to get into the thick of it right away.

I had a preconceived notion that I’d get an office, start
working on a project, and 6 months into it, be pitching for capital. In fact, in anticipation for the EIR program, I spent 2 months up in Seattle meeting with all the folks I could from the games industry. I was in a big hurry to reach an idea and start working on it ASAP.

What I really did
The wise advice of several of the partners (Thanks Jon & Dave!) was NOT to jump right in, but instead focus on being a sponge and absorbing what the Bay Area had to offer. After all, the entire point of why entrepreneurs thrive has to do with the environment, people, and resources. Moving here from Seattle and not immersing myself would be a big shame.

As a result, I turned my attention to going to dozens of events including:

• EconSM
• Game Developers Conference
• Web 2.0 Expo
• Mobile Persuasion
• CommunityNext
• CHI
• Future of Entertainment
• Virtual Goods Summit
• … and many other random ones

The focus here was to meet as many folks as I could from the San Francisco area, from all walks of life, and pick their brains about everything. I remember that in my first 4-5 months at MDV, I almost never ate at the Sundeck, which is pretty much the only choice for food on Sand Hill Road but instead spent it driving all over the place. I met 2 or 3 new people every day, which was great.

4 months on the conference circuit was enough for me – it started to get boring, particularly the ones that evolved purely around Internet media. The best conferences were the ones on games, mobile, entertainment, etc., which were peripherally related to online media, but only somewhat. From there, you could get a different angle on the world we live in.

My goal shifted into finding a co-founder, and moving on to prototyping some new ideas.

The venture capital business
Along the way, it was a very compelling opportunity to get to see entrepreneurs and deals coming through the door. With my background in online advertising, these were the types of Internet businesses I looked at:

1. Businesses that either were advertising plays in themselves (infrastructure, ad networks, etc.)
2. Or, they depended on advertising to monetize
3. Or, they depended on advertising to acquire users

Of course, in 2007, that pretty much constituted 99% of the Internet companies that came through the door ;-)

Overall, I was very impressed by the venture process, and the honesty and integrity of the people involved. The highest consideration was given to reaching decisions quickly, giving insightful analysis, and providing useful relationships. I saw this across the board, not just at MDV but also with other firms I interacted with.

The biggest lesson I learned was about the way VCs qualify deals, and that in many cases, getting venture money is not the right thing to do for the business – and that’s OK. I saw many companies which I thought would be strong, profitable companies, but didn’t have billion dollar potential. In many of these cases, the companies are likely to do quite well, but things might actually be a less optimal outcome for them if they took venture money.

Another important thing I learned was that, for all the ways that the VC industry is opaque to outsiders, the folks are actually very accessible. It was hard not to run into people I knew from the venture industry at any and every conference, regardless of what city it was hosted in. I also continue to see VCs blogging, which is great, because you can always send them an e-mail and start a conversation about anything.

Also, one nice outcome of immersing myself in the conference circuit was that I was able to bring a couple interesting companies to the partnership. One company that was funded, as a result, was Hi5, and the other is still unannounced.

Conclusion and next steps
Overall, the experience at MDV was great, and I would do it again. For entrepreneurs, I think EIR programs are a great way to transition into a new geography, or market, or otherwise learn a lot more about the industry.

Later in the year, I ended up transitioning into a consulting role as a motivation developed to live in San Francisco rather than down in the peninsula. Also, I had ended up finding a co-founder for working on prototypes, and we wanted to go and learn from the best internet folks – which today, are mostly located up in the city particularly around the SoMa area. We’re still continuing to tinker away on a couple projects.

And I won’t leave it unsaid – Happy 2008 everyone! ;-)

(For a quick summary of Search engine optimization, check out the Wikipedia entry)

SEO is a black art – with an emphasis on Art not Science
If you’ve ever wanted to increase website traffic, you’ve likely stumbled on a dozen or more guides on SEO. In fact, if you google "SEO tips" and look specifically for PDFs, you end up with this results page. Click through a couple and check them out.

In general, you’ll find that most of the SEO articles seem to consist of a long list of rules for what’s "good SEO" like:

• Put the right keywords into your URL
• Change your META keyword list to include important phrases
• H1 is so much better than H2
• Use sitemaps, because that helps your indexing
• … blah blah

OK, I’m sure that most of these work, at least to some degree. And at the very least, large groups of consultants and agencies benefit from this confusion, because it ends up looking like a tax code – if you don’t know the ins and outs of hundreds of possible rules, then it’s hard to think about tackling this in-house, or even evaluating the performance.

Like viral marketing, the vast majority of the knowledge out there consists of little lists of "rules of thumb" ideas to implement. And also like viral marketing, there’s a lot of tribal knowledge and hucksters who make a living writing about cute ideas, without putting it all together into a coherent theory.

If SEO were a science…
Let’s think about how the medical sciences has evolved over time:

• 15,000BC to 1700s AD:
  • Hmm, if you’re sick and you eat god-knows-what, you seem to get better!
  • Also, you might decide to make up some rules for WHY it is that something happened
  • This could be, of course confusing correlation with causation
• 1746 AD:
  • British guy does the first ever A/B test to figure out that citrus cures scurvy
  • Of course, you still don’t know why – but it works
• 1853 AD:
  • Aspirin is finally extracted from its herbal form, which is tree bark
  • Yay for drugs ;-)

And of course, it was only recently that we started to understand HOW and WHY certain drugs work, through a deeper understanding of the human body and the chemistry involved.

So back to SEO, if the accumulated knowledge essentially consists as a bunch of "best practices," then what you have is a very nonscientific approach to the problem. It’s basically at the stage of eating various things, seeing if they kill you, and collecting a list of plants that may or may not help you. And as a result, you’d have a list of "Eat this, and not this"-type rules, rather than a unified way of approaching the problem.

What are the right SEO metrics?
First off, if anyone has found some interesting data or articles in this direction – please send them to me – I’d love to read more about the topic.

In the meantime, I’ll talk about what I’ve found so far: After searching around, I definitely found some interesting articles for how to incorporate these ideas into a broader framework – here are two:

• A Year in Adsense Stats – Some statistics from one batch
• SEO: Metrics that matter

Both are worth a quick read.

The idea is to separate out different parts of SEO. You have 3 components:

• Methods (do X not Y, do A not B)
• Performance dashboard (our # of pages crawled is X, let’s get it to Y)
• A/B testing

So generically, you basically have the actual experiment you are running – like whether or not a change you make to the site helps – and then an experimental framework that helps validate what is happening.

As a result, you end up breaking down a performance dashboard that has variables like:

• Number of pages crawled
• Number of pages indexed (3 Main Engines)
• Number of days X % of the pages survive in Google from first full crawl to deindex (What I call ‘Burn Rate’)
• Indexing methods used
• Uniques
• Page views
• Clicks
• Revenue

(quoted from SEOIdiot)

Or, metrics like:

• Brand-to-nonbrand ratio
• Unique pages
• Page yield
• Keyword yield
• Visitors per keyword
• Index-to-crawl ratio
• Engine yield

(quoted from Stephan Spencer)

But having all of this stuff together is pretty confusing, isn’t it? Especially when you can’t figure out what question you are really trying to answer?

Putting these metrics into a dashboard
Tracking a large collection of metrics is not interesting in itself. What you are essentially looking for is a series of steps – I’ll frame this a community site that generates SEO pages, since I’m sure that’s what a lot of people are working on:

1. User comes to the site
2. User writes an article
3. Article is then published to a unique URL
4. URL is then crawled
5. URL is then indexed
6. Keywords are then searched
7. URL then shows up in the search result page
8. URL is then clicked on
9. Then a user ends up back on the site

What you basically want to do is to step through and track variables against each step of this process – the more granularity you can get, the better.

In fact, let me restate the really interesting question you’re trying to answer from these stats:

For each page created by a user, how many additional
users its likely to attract over its lifetime?

Integrating this into a
holistic dashboard, you’re then able to see how the changes you make
impact the variables throughout the site.

However, that’s not what most people do – they might track only steps #1,#9, through aggregate pageview numbers. Or if they are going to track steps #1,2,3,9, since those are the only things happening directly on their site.

But of course, in the case where you can’t measure directly, you can measure indirectly via using comScore or Compete data, or doing external searches yourself, or whatever. Or, you start using measures like in-bound links, or some other metrics, which lets you at least roughly estimate how a change you make alters the numbers for what Google will give you.

If someone has developed or used a dashboard like this, I’d love to talk about it ;-)

This still leaves A/B testing, which I’ll leave to a thought experiment for the reader, or a future blog.

UPDATE: My good friend and old college roommate, Eric Peters, gives his take here. Eric is a metrics guru and was previously at MSN working on monetization, so his opinion is absolutely worth reading.