Eric Enge noted that at the Searchology event that Google's Udi Manber stated that 20 to 25% of the queries that Google sees in any given day are queries that they have never seen before.

eBay recently banned search arbitrage, and Google is also trying to clean up their network. eWhisper posted that he thought the recent clean up may be an attempt to make the network look more appealing before showing advertisers what URLs their ads appear on. A neverending Webmasterworld thread highlights that many thin AdSense publishers got emails from Google stating that "our specialists found that your business model is not a good fit for the AdSense program".

The hard part is both judging intent and forcing quality. On one front Google offers a heat map showing you that your ads should go in the middle of your content, then they tell you to use AdWords to drive traffic to AdSense sites, and in their guidelines they state:

No Google ad may be placed on pages published specifically for the purpose of showing ads, whether or not the page content is relevant

In a market where virtually every business arbitrages other markets what is an improper business model? SEO is a form of arbitrage. I even had one engine call me up and ask to pay me to spam arbitraging Google's search results. That engine may even be one of Google's largest syndication partners! As ShoeMoney pointed out, many large arbitragers will continue arbitraging. Companies like Business.com or the 100s of shopping search engines are not going to get booted out of syndicating ads.

I think the big long-term distinction between what Google considers low quality and of quality is going to be brand equity. Do people visit your site from channels other than Google? Large brands get more return out of buzz marketing, while smaller businesses lean hard on search. inbound made a great post in that WMW thread about the arbitrage and MFA changes, which notes that Google is getting better at coming up with proxies for visitor value.

Based on Google's authority-centric relevancy algorithms and this move it is clear that Google wants to trust the larger businesses so they have less work to do policing the web. The way around getting forked by Google is to create something that does not need Google to exist. Create the type of site that people would link at if Google did not exist, and the type of site that they would want to advertise on directly. I have a large AdSense site that needs a re-brand and some serious work on content quality if it is going to stay viable in years to come.

For those who just got the death letter here are a few options:

• Buy old trusted sites with a clean core and real value add. do arbitrage stuff off of them

• Look to other related programs like YPN, Searchfeed, and Ask.com's feed.
• If your site is of low value consider showing fewer ads short term...to make up for non arbitrage ads and to lower your risk profile.
• Create a new corporation, open up a new AdSense account and figure out ways to get your sites approved.
• Publish fewer and higher quality sites with fewer ad units on each page.

From a risk management perspective, I think every web publisher should own at least one real branded site, even if it offers low returns for the amount of work required to maintain it.

If the people you are talking to gain by sharing your story.

For example, right now Google is desperate to show the value of video so they can turn that huge copyright nightmare into a revenue stream.

Nick Carr recently published an article about how page views are consolidating:

He found that between the end of 2001 and the end of last year, the number of Internet domains expanded by more than 75%, from 2.9m to 5.1m. At the same time, however, the dominance of the most popular domains grew substantially. At the end of 2001, the top 10 websites accounted for 31% of all the pages viewed on the net. By the end of last year, the top 10 accounted for fully 40% of page views. There are more destinations online, but we eem to be visiting fewer of them.

Now that you can play music in Google's search results you have to think that consolidation is only going to get worse.

The cited research also notes that social sites are playing a role in the consolidation of traffic. Andrew Goodman recently published an article about why people would want to contribute to online communities. He noted that much of the community giving spirit is based on a common thread of a shared sense of justice in society. It then makes sense that political blogs, religious sites, deep niche sites, and things built on heavily biased/warped worldviews are typically stronger communities that are likely easier to maintain than those based on broad interests.

Add to that the following, and a clear trend emerges:

• people pay more attention to and believe what resonates with them

• people are more likely to share with those that resonate with them
• fringe biased content is typically more remarkable than vanilla unbiased content
• controversies spread far

As the big players keep taking verticals away from smaller players, back-filling their databases with partner licensed content, and personalizing results to match our worldviews, the profitable (and thus sustainable) communities will move more toward the fringe edges.

Unless there is some other change to modern relevancy algorithms, the top ranked results will be watered down vanilla content from editorially trusted sources or heavily biased content from the fringe. A mixture of various biases will also allow people to believe in whatever they thought was right before the consulted the oracle. There will be a reasonable accessible source for every lie. Thus we will think the information is of greater value, while it turns out to be low value entertainment leading to self fulfilling prophecies...but at least the ads will be targeted. :)

Having coined the term domain investor, Frank Schilling is a recognized leader in the domaining field. He talks about domaining on his blog at Seven Mile.com

I recently asked Frank Schilling if he would be up for an interview and he said sure.

What makes domain names so powerful from an investment standpoint?

Several things.. When you build on a domain name you are the master of your destiny because you are not beholden to anyone else's platform. It's the Internet comparison to owning the building vs. leasing from a landlord. Internet law is new and undefined. Google and Ebay can be brutal landlords changing rules or algorithms that put you back out on the street without notice. If you own a powerful generic name or name phrase; you will get internet traffic independent of what the search engines and auction marketplaces try to do to you.

Why are [secondary market] .com's typically so much more expensive than other extensions?

Dot com's are the most readily understood domain extension. They are so powerful even my daughter knows what they are (she's 3) .. Many people will simply append the subject matter they seek with '.com' in their address bar, expecting to find products and services that match the generic keywords they entered. That world-wide mindshare has been reinforced since the very beginnings of the Internet via trillions of dollars in collective global marketing and serve to create a premium value for the extension. The marketing that you, I and others have done, has served to make .com the first extension most of us try in the browser's address bar.

Have you ever sold a domain name for a loss?

No. Even names where folks scoffed saying I overpaid at the time, look cheap in hindsight.

What is the best domain you regret not purchasing?

Cameras.com sold for 1.5 million.. That was a really tasty one.. I chickened out over a million. Wish I could get in the Delorean and go back in time on that one. Also Food.com sold to the food network in a San Francisco bankruptcy court in 2003. it went for $300,000 ish .. I should have bid 500k back then.

Is it too late to get into domaining? If you were starting today which model would you go after? Would you try to buy a few strong domains or try to own a much larger portfolio of weaker ones?

I think there are so many untapped opportunities here.. Within a few years, hundreds of thousands globally are going to be directly employed in this industry. It is early not late. This is like California in the 1960's. -- yes, it's not the 1920's anymore, but there are still mountains of untapped opportunity. I would probably focus on buying and selling, flipping up and bootstrapping profits back into the business if I had to start today. Also SEO and PPC keyword arbitrage.

Are new heavily marketed extensions like .tv or .mobi a good opportunity?

Only to flip.. If you can get something good cheaply and sell it to somebody else then do it, but I am avoiding those extensions completely. I like .com, .net, .org and the CCtld of the Country you are in because the name spaces are established, the renewal fees are certain (consistent/predictable) and because that's where the organic, generic intent type-in traffic is.

What are your favorite cities to visit? How does real estate there compare with domain name prices?

I like Los Angles and Las Vegas a lot. It's funny because those cities real estate histories have parallels to the domain industry. In Southern California you have Irvine where one man basically acquired millions of acres through the early 1900's and then sold to a large corporation in the later 1900's. Today the seller looks like a fool because he sold so cheap when viewed against the development which has occurred in the surrounding area. Yet had he not sold, none of the roads, utilities, infrastructure would be there, so the area would not really be as valuable. So if there is a comparison between domains and real estate, I think "development naturally follows acquiring the land" and "prices increase as the people come in" are the two over-riding factors.

If I am planning on developing a site, and am working with a small budget, do you think it is better to buy the .com, or to buy a .net or .org and spend the difference on more marketing and development? What about country level domains?

I like Country Code domains in Countries that have even-handed registration rules which allow all all sorts registrants to invest and develop there. Country codes I like include .CA (Canada), .co.uk (UK), .de (Germany), .br (Brazil), .cn (China), .in (India). In many circumstances Country Codes are stronger than .com. There are a host of reasons for this including currency issues, language, nationalism. I would always try to get the .com because it helps you to build traffic outside of the search-engines (everybody winds up at the .com eventually). I do like .nets and .orgs when they are priced low enough.

Years down the road when all the best names are gone and many of them are beyond the budgets of individuals and small businesses do you see outlier names like .info, .biz, or .cc getting any traction?

I do not.. I think it will be .com, .net, .org and the CC TLD of the Country Code you live in, ten years from now. If the Web extension gets approved, it could eventually unseat .net but it would take time to catch on. That forecast is assuming names get rolled out in their current way. The only other thing that could change destiny is the wholesale addition of hundreds of new extensions such as .GOOG, .MSN, .IBM, .YHOO, where every company got its own extension. It's problematic because corporate jealousy precludes adding just one or two.. and companies can barely manage their names, let alone an entire GTLD. It would take a generation to roll out and would ultimately strengthen .com relating to generic words such as Maps, Books, Shopping etc.. So while I find that kind of wholesale change revolutionary (ICANN and Verisign would resist it), it still 'could' happen.. and that would change to weaken .net and .org.

If I'm planning on developing a site when is it best to buy the core related keyword domain? When does it make more sense to create a unique word or add a common word like "hub" or "community" to the name to get an $8 domain name instead of spending thousands more for the exact match domain?

You can focus on building a great company without a great name. I like generic word + 'hub' or 'web' or 'world' style domain names. But if you build the world's biggest ceiling fan company at fanhub.com and then you want to acquire ceilingfans.com .. it is going to get much more expensive as time goes by. Names like those are going to be worth millions one day, so the time to acquire them is when they seem cheapest and unimportant to you. That's always the best time to acquire great names btw.

Are there any good $8 domain names available right now?

Yes.. but mining for them is getting harder. I don't buy anything in the available space anymore.. haven't in a long time. Too time intensive and "domain tasting" has creamed off most of the generic defensible undiscovered names.

How many ways do you categorize domains? What types of domains are the best from an investment perspective?

We have 60 main categories such as 'cars' and then 1600 subcategories including 'car accessories', 'towing', 'insurance'. etc. The best domain-names are generic defensible keyword-style (one two and three word) phrases which get some trickle of organic generic-intent type-in traffic; for nothing more than the keyword-weight, gravity and resonance of the generic words that make up the domain name.

What are your favorite spots for buying domains right now? Which auction do you like the best? What changes would you like to see to how domain names are auctioned off?

We are going through a seasonal dry-spell at the moment.. There is still an annual echo effect of expiring domain names which results from the dot-com bust where millions of folks in 2001,2,3 let their domain names expire.. the expiring name echo-period runs from November through April, so we are in the quiet season at the moment. I like Snapnames, Enom, BuyDomains, Godaddy. The auctions are presently run by for-profit clearing houses which inject themselves between expiring names and bidder registrants. One day ICANN will probably get involved auctioning new names like the FCC does with reissued airwave licenses.

What percent of domain sales do you estimate are publicly known?

About 5%.. maybe less. I was at my daughter's friend's birthday party recently and the father of one of the children confided that he sold a terrible made up brand-like sounding name for $23,000 (it was either 23 or 33k.. can't quite recall) This guy was not a domainer.. you would never hear about the sale. I get folks from regular walks of life coming up to me all the time confiding that they are part time domainers.. These are folks who have never visited a domain chat room, have never visited another domain site. Their only connection to the industry is through their registrar. It is a billion dollar business.

Many domains tend to sell for a multiple of PPC earnings. In 10 years time do you think the baseline will move to some other metric?

It already has.. No good domain portfolio has changed hands since BuyDomains and that business would have sold for considerably more than the rumored amount had the company's former owner been engaged in selling advertising alone, vs selling his names. Prior to that there was Name Development's sale to Marchex (Yun Ye transaction). No large, high-quality portfolios have changed hands since. Other sales have been smaller or split-portfolios consisting of good names interspersed with trademark issues. Individual names often sell for 100 years PPC.. No high quality domainer would dream of selling a portfolio worth potential billions to a third party for 10-12X PPC revenues. PPC is a flawed multiple because it works off a rev-share. If you buy a portfolio for 10X and it is on a 50/50 rev share (after 'cost of services' through Google Adsense) that means you sold your portfolio for 5X what Google could make with it. Maybe 3X if you exclude the amount Google shaves for smart pricing. That is so insanely cheap. Only a fool would give names away like that. If I were selling I would pick a walk away number (the youtube style multiple) or I would sell names individually. The breakup value of large portfolios will be in the billions if they aren't already.

You have mentioned that you thought search was promoting too many anchor stores vs smaller boutique websites. Do you think this creates an opportunity for other search players or adds value to topical community resources? Do you see it becoming more or less profitable to make niche websites and domain names?

I think of every domain name as an alternative search engine under the keyword embodied within the name. Search-engines by their nature can only display 5 to 10 results above the fold. As markets get more competitive that 11th result will become a different website, with a greater frequency. It's always getting harder for search engines to rank the top results.. as that competition intensifies as domain implementations get better, the domain name becomes a more viable alternative browsing experience. It's going to become much more profitable to run boutiques that sell things in the years ahead because software, fulfillment, products are all becoming cheaper as marketing costs go up. A good domain name reduces your lifetime marketing costs and increases marketing opportunities.

Are you concerned with large web companies claiming certain websites and types of websites as being unsafe to end users?

It depends how far they take it. Clearly I'd be concerned if browsers incorrectly claimed that advertising was bad and tried to block all sites with ppc ads.. While that wouldn't impact my lifestyle it would stymie newcomers and limit folks abilities to browse the web. I think the browsers are already on thin ice with a lot of the error search stealing to the right of the dot.. type your favorite website with .xom or .con watch where you go, think about where you were intending to go. That's stealing in the browser. It's unseemly to see that kind of conduct coming from a major US Corporation. I think in time there will be more freedom of navigation not less. Users (by their nature) want to be free. So to answer your question, there are probably too many sites with advertising on them for the browsers to do something draconian or to limit browsing freedoms. People would type in a website and say the internet's broken, my browser won't let me go anywhere .. too many sites would be impacted.

As the web gets more competitive, I believe any single sign of quality will likely have less of an overall effect on a website's position on the web. Do you believe that is true for domain names as well, or will everything being so gamed only increase the value of domain names?

I think probably the later.. Mark Twain said: "History doesn't repeat but it rhymes" .. The past may not be a true indication of the future, but domain names 'are the Internet'. You need a domain for email, in fact the only constant since the dawn of the commercial internet in 1993 (Netscape 1) has been the domain name. If you feel comfortable investing in anything related to the Internet it should be a generic domain name.

What makes Riesling so good?

I was picking up dinner at www.pappagallo.ky they have this new house wine from Germany.. very light refreshing.. good lunch wine. Darn, can't remember the name - Maybe if it ended in .com! :)

----

Thanks Frank. Check out Frank's blog to read his latest insights on the domain market.