UNICEF appears to be in the market for its own TLD.  The UN organization dedicated to providing humanitarian and development aid to children and mothers in developing countries recently announced a Request for Information from potential vendors to help acquire and operate .UNICEF.

Back in October 2009, the Universal Postal Union (UPU) entered into an agreement with ICANN to operate .POST, making UNICEF the second UN organization to pursue its own TLD.  I’m curious to see what will happen going forward – will even more UN organizations follow suit and apply for eponymous TLDs?  Moreover, I wonder why the UN is so interested in owning TLDs and why these organizations feel they need these kinds of labels when they have .INT all to themselves and there is no fee for registering an .INT domain.

I’m also wondering who will be next – maybe the World Health Organization with .WHO?

I’m up in Eagle River, Wisconsin on vacation with my sons this week.  We visited last year around this time, and I blogged about that trip as well. 

My birthday was on Saturday, and in addition to my two boys Charlie and Oliver, my sister, my parents and my grandmother were all here to celebrate with me.  We spent all day out on the lakes, fishing, swimming and boating.  For my birthday dinner, I chose bratwurst – which you can see pictured here – and my mom made her homemade coleslaw. 

As the brats sizzled away on the grill, I got curious about who owns bratwurst.com.  Turns out it’s Johnsonville Sausage, which got its start here in Wisconsin.  Bratwurst.com is obviously a great name for the company to own, so way to go, Johnsonville.  Now, if only they would build a separate site – even a very basic one – at bratwurst.com, they should get it to rank separately in organic search results.  Their competitor, The Sheboygan Bratwurst Company, did and they fairly easily grabbed a spot on page 1 with bratwurst.net.  Johnsonville.com ranks on the first page of a search for “bratwurst,” but if the company could also get bratwurst.com to rank, they could potentially push another site off page 1 and collect even more traffic.

The clever marketer behind the Charlie Murphy ads should leverage its digital assets better and dominate online just like it does in the grocery store.

FairWinds recently released a study that calculated the cost of typosquatting for the brands behind the 250 most highly trafficked websites.  The total included costs from visitor acquisitions, lost sales and impressions and also recovery of typosquatted domain names, and came out to $327 million per year.

A major source of losses for brand owners is pay-per-click (PPC) sites; when a company invests in paid search, its ads appear on PPC and other websites in addition to the search engine’s results pages.  When an Internet user clicks on the sponsored link, either on the search engine site or another site, the company pays a click fee.  When an Internet user makes a typo or keystroke error while trying to reach a brand’s website and lands on a typosquatted domain site that is leveraging PPC to monetize traffic, we have found there is an 18% chance he or she will click on the targeted brand’s link (meaning the brand suffers a loss in the form of a click fee for gaining the visitor) and a 7% chance he or she will click on a competitor’s link (meaning the brand suffers a loss in the form of diversion while another advertiser pays the cost-per-click (CPC) and picks up the visitor).  The formulas that we used to calculate those losses are as follows:

For users who click on the targeted brand’s link:
18% x (Annual traffic per domain) x (Average CPC) = Advertising costs for the target brand

For users who click on a competitor’s link:
7% x (Annual traffic per domain) x (Average CPC) = Advertising costs for the target brand’s competitor

Originally in our calculations, we used an average CPC of $2.74 for each formula.  Given that we could calculate the CPC for the specific keywords in this study of 250 sites, we leveraged today’s Google estimates for each keyword CPC rather than looking at an historical average for a broad range of keywords.  Because there was a range of CPC and traffic values that could have created bias, we then extracted the weighted average CPC across the dataset.  We determined that $2.03 was the weighted average CPC for all the sites included in our study and therefore, $2.03 is the actual cost the site owners incur when a visitor lands on a typosquatted parked page and clicks on their link. Ultimately, this is the amount of money paid by the advertiser and shared by the contextual Internet advertising provider and the owner of the domain (and sometimes an intermediate party such as a domain parking company).

Obviously, this reconsideration alters our final calculation somewhat.  With the new average CPC, the total cost from ad clicks on typosquatted PPC sites in our study is $187,288,458 per year.  When added to the costs of lost sales and impressions, plus domain recovery costs, the re-calculated total cost of typosquatting is $364,276,874 per year.

If it’s not already apparent in the scope of this study, this figure applies to the 250 sites that we studied, and is not a comprehensive measure of all typosquatting (that figure would obviously be much greater).  To read the updated paper with an in-depth discussion of the damage that typosquatting inflicts on the 250 most trafficked websites, visit the FairWinds site for the entire paper, or the revised section for a discussion of how we calculated the sample set CPC.

I was driving with my family to Nationals Park to see the Nats play the San Francisco Giants on Sunday, when I noticed a Windows Catering truck drive by. As it passed, (because I can’t help myself) I noticed the company’s domain name displayed on the side of truck. Again, because I can’t help myself, I asked my wife to slow down so I could take a picture of the truck for this post. I knew that a local business using a name like CATERING.COM would make a dynamite blog post. Alas, the truck pulled off at Navy Yard before I could call up the camera function on my BlackBerry. (I wish I had gotten the photo; if we can get a picture from the catering company, I will update this post later.)

It’s likely that the domain name windows.com was not available when the catering company first looked to “get online,” for obvious reasons (think Microsoft). So Windows Catering had to come up with an alternative domain name.

I wish I knew the story of how they came to acquire this domain name because it’s extremely valuable. I routinely run valuations for my clients and research how much domains are trading for in the domain aftermarket. I think a domain like catering.com could easily fetch between $1 million and $3 million today.

Why? Because catering.com likely garners a fair amount of type-in traffic, but it also has a unique ability to rank well in organic search results. Google and Bing (especially) assign page rank to older domains with keyword-to-domain root parity and sites linking in, so this domain has special attributes that earn it “search love.” This means that with even a modest SEO investment in the website itself, search engines are likely to assign it a high placement in organic search results. This gives the domain/website maximum visibility and increases the opportunity for it to earn organic click-throughs, without the incremental costs associated with paid search. Internet users broadly search the term “catering” over 16,000,000 times per month, and each time, they see catering.com among the top results.

This high rank not only drives traffic to Windows Catering from local D.C. searches, but it also introduces the brand to the wider U.S. search audience by delivering their impression to searchers throughout the country. For another example of how this works, go to google.co.uk and search for “catering.” You will see a UK company that owns a different domain with “search love,” caterer.com – this site’s owner has earned the top position in organic search in the UK because the domain has a Google page rank score of “5.” (In comparison, Guinness.com and Cadbury.com each have a Google page rank of “6,” despite being much better known brands.)

Smartly, Windows Catering also owns WindowsCatering.com, which it redirects to catering.com, for those customers who are familiar with the brand and search for it directly.

I think that this is a great example of the positive effects that a well-chosen domain name can have on a company. Sometimes unbranded domains can enhance a brand strategy, and not just for smaller players like Windows Catering. Just look at what Toys.com and Mortgage.com have done for Toys “R” Us owner Geoffrey, LLC and Citigroup, respectively.

UPDATE: Windows Catering got in touch after reading this post and sent over a picture of the catering truck that a fan had posted on Twitter:

We’ve written before about start-up companies that use alternative spellings of common words as their brand names.  I came across another example recently: Loopt, a social-mapping service that offers a variety of mobile apps.  Loopt uses loopt.com as its domain name.  It also owns the more intuitive and conventional spelling of the domain, looped.com, which it points to the Loopt homepage. 
 
It’s always nice to see companies that get it.  Kudos to Bidz (bids.com), Loopt (looped.com) and Flickr (flicker.com).

The other night, I was at dinner with my wife and she pointed out that, in addition to the usual sweeteners (sugar, Equal, Splenda, etc), the restaurant also had a stevia sweetener.  I had not heard of stevia before, and was unfamiliar with this particular brand, Stevia Extract in the Raw (SEITR), so I picked up the packet to take a look. 

Stevia is a plant with sweet leaves from which a low-calorie sweetener can be derived.  It’s regarded as a more natural alternative to artificial sugar substitutes like Equal and Sweet-n-Low.  SEITR’s parent company, Cumberland Packing Corp, also owns Sugar in the Raw, a popular brand of unrefined sugar.  Presumably, the brand name was chosen based on the “Sugar in the Raw” convention.

On the back of the packet, I quickly noticed the domain name: SteviaExtractInTheRaw.com.  Even though this is simply the brand name appended with .COM, it’s a terrible domain name, mostly because it is so long.  Also, it’s very easy to overlook the word “Extract,” because it is in such a smaller font, tucked under “Stevia,” meaning consumers will be likely to forget to include it when typing the domain name and they might assume, like I did, that “Stevia” is the brand name.

Unfortunately, Stevia.com has already been taken by another company, Healthworld Online, Inc., an online resource for health and wellness information. The site heavily promotes another brand of stevia sweetener: SweetLeaf. 

Here’s a screenshot:

Luckily though, Cumberland Packing Corp also owns SteviaInTheRaw.com and points it to the main SEITR site.  This is a better domain name than SteviaExtractInTheRaw.com because it is shorter and because, based on the logo, it is more intuitive for consumers.  Additionally, it reinforces the ties to the sister brand, Sugar in the Raw.  I think brands should be conscious of these issues and focus on making it easier for consumers to access their content online.  One way SEITR could do this would be to promote SteviaInTheRaw.com as its primary domain name.  Another way the company could gain considerable competitive advantages is to acquire stevia.com and own the category online.

I just finished a conversation with a contact who is still at the ICANN meeting in Brussels.  He attended the ICANN Public Forum and let me know that in his opinion it looks like the .XXX gTLD, an extension dedicated to hosting adult content, should be approved soon.  Overall, he said, there was very little objection from the forum attendees.  In fact, the discussion lasted only about 20 minutes.  I can’t say I expected much more – many people who attend ICANN meetings and participate in the ICANN process are, generally speaking, either interested in selling more and more domain names (like registrars), or are totally oblivious to the ICANN community’s bias toward selling more names.

On the other hand, I heard the discussion over whether or not ICANN’s Chairman of the Board should earn a $75,000 salary lasted a full 45 minutes.

George Kirikos wasn’t the only one who raised really interesting and challenging points at the ICANN meeting.  There are two more people who made comments and asked questions that I think are important and that I’d like to talk about here.

The first is Zahid Jamil, who hails form Karachi, Pakistan.  The port city in the past has reclaimed land from the sea, and the idea of a mass reclamation project was proposed not long ago.  While the initiative had many potential benefits (increasing the amount of land would lower housing prices for citizens and generate revenue for the government), it was widely agreed that the economic impact of such a project needed to be assessed.  Eventually, the geological and environmental analysts determined that there was not enough information and hard data to thoroughly measure the environmental impact of the project.

Zahid compared this process to ICANN’s new gTLD rollout, pointing out how ICANN is pressing forward “without enough hard statistics, feasibility studies, and analysis, etc.”  He went on to say, “Here I would caution against a wholesale launch of new gTLDs in the context, similarly, of the root scaling study and economic studies, and the cautions that have come out in the recommendations, such as recommendations for surveys, etc., and the fragility and the risks to the Internet we heard all about at Nairobi.”

Another great contribution came from Bertrand de la Chapelle, the Representative from France to ICANN’s Government Advisory Committee (GAC).  Betrand made a spot-on analogy that compared the Draft Applicant Guidebook (DAG) to the process of drafting a parliamentary law.  When drafting a new law, a huge amount of preparatory work is involved that results in, as Bertrand quaintly put it, a pile of documents “as high as the Eiffel Tower.”  The resulting law, however, must be understandable and concise.  In this case, the DAG is equal to the preparatory work – long, dense and unrefined.  Bertrand went on to suggest that the DAG, along with all the comments and work that have gone into it, should be reworked into a much shorter and more user-friendly document of around 20 pages or so.

Both Zahid and Bertrand’s comments get at a common sentiment that I’ve been hearing, both at the meeting and in other conversations: that ICANN is nowhere near ready to launch new gTLDs.  In fact, it has only really begun to do its homework on the matter.  For example, aside from releasing multiple versions of the DAG, ICANN only just released an analysis of the economic impact of new gTLDs last week.  That kind of study should have been conducted years ago, when new gTLDs were first being discussed.

Unfortunately (and unsurprisingly), ICANN doesn’t seem to want to hear from anyone warning them to slow down.  In response to Zahid’s comments, Kurt Pritz, the moderator of the New gTLD Update session, basically said that ICANN was moving forward with new gTLDs because that has been the direction of ICANN’s policy – without any greater justification.
 

While in Brussels, I had the chance to participate first-hand in one of ICANN’s international meetings. One of the sessions I attended yesterday was the “New gTLD Program Update.” George Kirikos, who many in the ICANN community know and appreciate for his knowledge and commitment to improving ICANN, submitted a question via remote participation about an important paper Tim Berners-Lee wrote back in 2004 about the detrimental costs of new gTLDs (at the time, .MOBI and .XXX were among the group of proposed extensions). Berners-Lee is an engineer and computer scientist who is credited with inventing the World Wide Web (the system that links hyertext documents over the Internet, not the whole Internet itself – Al Gore would be very offended if I didn’t make that distinction). The paper is very insightful, and especially applicable today, as ICANN is on the verge of opening up the domain name system (DNS) to an unlimited amount of new gTLDs. The paper is not very long and completely worth reading, but one of the most salient points I took out of it is the following:

“Our first instincts, then should be not to change the system with anything but incremental and carefully thought-out changes. The addition of new top-levels domains is a very disturbing influence. It carries great cost. It should only be undertaken when there is a very clear benefit to the new domain. In the case of the proposed .mobi domain, the change is actually detrimental.”

A big thanks goes out to George for showing me this paper, and also for pointing out how ICANN has ignored its existence. To my knowledge, George has not heard from anyone at ICANN after he sent a message about the Berners-Lee paper and linked it to the current environment. They also dodged his question during the session, so I wanted to shed light on this subject:

According to Kirikos, the DNS is a tree structure with a single root, and it was widely agreed that such a structure was an improvement over the previous “flat space” of host names. Infinite new gTLDs would be a movement backwards into that flat space. ICANN is basically on its way toward institutionalizing this backward movement, and in turn, cause widespread instability and harm.

In my view, ICANN’s proposal for how to roll out new gTLDs will create needless chaos and instability in the DNS. New gTLDs could potentially be interesting to brands and to the greater Internet community, but ICANN is not creating an attractive environment for businesses to invest in. Rather, brand owners will spend money in order to protect themselves from infringement, but that spending does not equate to an “investment” in new gTLDs. It is likely that many brand owners will view new gTLDs as a financial burden, instead of an economic opportunity.

What could the domain names coconutloving.com, desperaterussianhousewives.co.uk, and jellyparties.co.uk have in common? The answer is surprising - all three sites are registered and run by the BBC.
 
Following a Freedom of Information request, the BBC recently released a list of 154 of the domain names that it owns. The sites range from the predictable, bbcfrench.com, to the peculiar, watchagrownmanrot.co.uk. While seemingly random, they are all relevant in some way to BBC content. Some of the stranger addresses are derived from the plots of BBC TV shows. The site bestmurders.co.uk, for example, was registered in connection to the dark comedy Psychoville. The domains and the sites they host are creative ways to give the shows more exposure.
 
The BBC was careful to ward off cybersquatters, though, and did not disclose the complete list of the domain names it owns. A number of the domains it omitted were related to forthcoming TV programs and storylines and had been preemptively registered. The BBC did so to avoid giving away future plot details, but also to prevent revealing its domain name strategy to cybersquatters who could then exploit that information to predict and register new names.
 
It’s really interesting to see the BBC display so much humor and creativity with its domain portfolio. The corporation has been able to generate a significant amount of buzz with its unexpectedly odd domain names and has demonstrated the value of owning them even if it never plans to use them for the creation of real sites (just try typing in desperaterussianhousewives.co.uk).