OK, let's get this straight before I go any further - I am NOT saying that ALL academic papers are worthless - just some.

Neither am I saying that; [a] all academics are stupid, or [b] that academic research does not have a place. So having inserted my get-out-clauses for any complaints that come my why, read on to find out why I have written about the subject.

I suppose my outlook on this issue dates back to when I studied for my first degree [in Business Studies] as a mature student. Fresh from the 'real world' of business I was disappointed [an under-statement] to find out that most of the theories and concepts that I was introduced to were simply re-hashes of what folk have been practicing for years [I use the 4Ps all the time when teaching marketing, but hey - are you trying to tell me that since mankind first traded they didn't realise that you need the right product, in the right place, at the right place using the right promotion?].

As both student and lecturer [no, that does not mean I am an academic] I have come across numerous papers on my subject - marketing - that I considered to be of spurious quality. But hey, I'm a lecturer in a provincial university - what do I know? However, on my specialist subject - Internet marketing - by definition of having written a number of books, I am an 'expert' [see it's in a book, so it must be true]. I would argue, therefore, that on the subject of e-marketing my opinion carries some validity.

As some of my books are 'academic' texts [I know, I appreciate the irony / hypocrisy] they must have 'academic' under-pinnings [say the publishers] and so I have to keep up with the journals that publish papers in 'my' subject. And so I read some really interesting, well researched and relevant articles. I also read some that have my office-mates grimacing as I shout and wave my fist at my PC because I am reading rubbish [I'm at a university, I access journals using an online facility].

I'm going to use one as an example, but if you are the authors - I'm not picking on you, I just happen to be a 'real' expert on the subject you wrote about - domain names. Similarly, you should appreciate that when you put your thoughts into the public domain you open yourselves up to criticism from that public - as I do. Indeed, I welcome the views of folk who disagree with me, it's one way I can improve my understanding of a subject. I also appreciate that much of marketing is 'art' rather than 'science' [though some might disagree on even that!] - and so people can have different opinions with none being 'right' or 'wrong'.

It is also the case that I actually think this article carries academic merit - it is a good academic paper. Indeed, reading the biography of one of the author's would suggest that he actually knows more about the practical aspects of domain names than he is 'allowed' to include in an academic paper that requires evidence to support any comments and/or findings. In terms of academic vigour, the article is sound. What I am railing against is the value of such papers in the real world.

I should also add at this point that much of the following is my opinion - and as my friends will gladly endorse - I am often wrong.

The article in question is: An Investigation of Global versus Local Online Branding - you should read it before considering what I have to say.

Issues that I would raise against the 'real-life' validity of the findings include, in no particular order of relevance/importance:

• The research supposes that the organization - and as the subject is 'branding', ergo, the marketing department - gave the choice of the original domain name much thought. Back in the mid-1990s when these names were being registered, the decision was [and still is] often given to 'IT' - and as the USA was ahead of the game in development of the commercial Internet, it was American IT depts that first registered domain names, and they registered dot coms. There was little or no input from marketers [it is still the case that in 2008 few marketers understand the value of the organization's domain name. In 1996, they didn't know what they were]. Having developed websites on those names, and used them in promotions, company literature, email addresses, business cards, the sides of trucks etc etc etc any change of domain name is a serious - and costly - decision.


• Common with most [all?] academic research is the lapse in time between research and publication. In this case the research is stated as taking place in 2003 - the publication being in 2007. In Internet terms, 4 years is a long time [in 2003 few had heard of blogs and MySpace, for example - and the dot eu and dot asia domains did not exist].


• Some of the research on which this paper is built is flawed; [a] for the same reasons I argue this article is flawed, and [b] because of the 4-year time lapse that research is out of date [eg, out-dated by advances in technology].


• The dot com suffix is accepted as being the one to use if your organization trades outside its own country. This is still the case in most scenarios. It is the advice I - and others like me - have been giving since 1996. So [whether I am right or wrong] many companies registered the dot com because they were advised to do so.


• Hypothesis #1 says that having a dot com gives the site a higher PageRank. Two points on this. (1) A high score on PageRank is dependent on many variables not factored in this research, and (2) whilst considered as a useful guide by some, in SEO terms having a high PageRank is known by experts [ie those who practice it] as being of little - if any - value.


• Also on the subject of Pagerank, the article states "The Google PageRank metric for the MNCs ranged from one to ten, with a median rank of six. Fourteen companies had no PageRank, which suggests near invisibility with Google." Errr, no. PageRank is only one element of the Google algorithm, a website could be number one on a SERP for any given keyword while having a zero rating on PageRank [at the time of writing - June 08 - alancharlesworth.eu tops the Google SERP for a search on "alan charlesworth", yet the site has a PageRank of only 3].


• Dot coms were the first domain names to be commonly available, so they have the longest 'history'. This is important as one of the 200 or so variables used in the Google algorithm is length of registration of the host domain name. Ergo, all other things being equal, a website hosted on a domain name registered in 1994 will have a higher Google ranking than any other domain registered after that.


• There is some confusion between PageRank and rankings in Google's search returns. The two are not the same. There is also the inference that the dot com is a better suffix to have for search engine ranking. Whilst this might be true where a user has searched on Google's 'the web' search, the reverse is true if the searcher uses their own country option [eg 'pages from the UK']. Searchers not all being in the USA has also been ignored in this element of the research.


• Also on the subject of where in the world a user is, the article mentions users being 'redirected' to ccTLDs when they type the dot com into a browser - its all about IP recognition and sounds a reasonable idea to me.


• No other variables used in the Google algorithm are considered in this research - therefore any conclusions related to Google listings are flawed.


• Hypothesis #2 is little more than coincidence based on other points raised here and factors external to the Internet. For example, the article states "MNCs listing a dot com domain name have a higher Fortune ranking than MNCs with a local domain." Sorry folks, this is simply a statement of fact - or is there a suggestion that having a dot com actually helps you in the Fortune 500 ranking?


• Hypothesis #3 takes no account of the potential technical reasons for using a dot com or ccTLD. Though I would always argue that marketing considerations take precedence - issues such as hosting in local countries and using directories on a single domain giving greater control and security are often the deciding criteria.


• No consideration is given to the availability of ccTLDs in all the countries in which the organization trades. If you cannot register your domain name in the suffix of every country, then you may as well stick everything on the dot com. This might well be the case where the organization's name is common [eg a person's name] or a generic word - why do you think new or renamed global entities use a word that they have invented? Yep, 'cos if the word doesn't exist no one has registered it as a domain name.


• Hypothesis #3 uses Hofstede's dimensions as a criteria. For all of the above, I would discount it as any guide on why a domain name was registered. Even outside my opinion on the subject of domain names, Hofstede's original 'dimensions' are widely questioned when used in a contemporary environment.


• And finally - and this really is a personal viewpoint - is the use of 'logistic regression testing' in Hypothesis #3. The sight of any kind of mathmatical formula sends shivers up my spine [I think it was an Act of God when I got a maths 'O' level] - but to use them to investigate why any given company uses a dot com domain name is - I think - a nonsense.
  My experience is that some time back in the mid nineties a techie was bored one day and registered thecomapanyname.com because they thought it as cool. The techies then started using it for 'email' purposes or it may have even been an element of an EDI or data processing facility [you need an IP address for some things]. Then, when someone up in the ivory tower finally took notice of what the Internet could bring to business, they rang down to IT and said "we need one of these domain names thingummybobs for the informationsuperwebnet, get one." To which the head techie replied "already got one boss". It was only several years later [though it still hasn't happened in some organizations] when 'marketing' got involved that anyone asked any questions about the name by which the company is found on the Internet. And if you think I am joking, try researching how and why some organization's came to register their original name. If they can't tell you exactly, then my scenario is how it happened.

So there you have it. An academic paper that in an academic environment has merit - but as a document that might help an organization choose what domain name or names it should use in its global marketing strategy, it is worthless.

This 'musing' is about the real-life value of academic research, not domain names. if you want to read my views on choosing the right domain name see: domain names - a marketer's perspective