Microsoft has just released a paper (.pdf) on the value of good domains calling it “Domain Bias” which looks like one of the best papers ever written on the subject of domains that prove out the value of a good domain.
This paper may become a domainers best friend in helping to sell domains to end users.
The authors call this as Domain Bias according to the author is defined as “a user’s propensity to believe that a page is more relevant just because it comes from a particular domain.””We provide evidence of the existence of domain bias in click activity as well as in human judgments via a comprehensive collection of experiments. ”
“We begin by studying the difference between domains that a search engine surfaces and that users click. Surprisingly, we find that despite changes in the overall distribution of surfaced domains, there has not been a comparable shift in the distribution of clicked domains.”
“”Users seem to have learned the landscape of the internet and their click behavior has thus become more predictable over time.””
“We find that domains can actually flip a user’s preference about 25% of the time.””
“The existence of domain bias has numerous consequences including, for example, the importance of discounting click activity from reputable domains.”
“”Our goal is to provide incontrovertible proof of the existence of domain bias.”
“We do so via a series of carefully designed experiments.”
“We ask if a search engine drastically changes the surfaced domains, do domain clicks also change accordingly? Amazingly, the answer turns out to be no.”
“Instead, we find that users click on the same domains despite changes in surfaced content. In a similar vein, if we take two search engines of wildly different relevance, we ask if domain clicks also swing wildly. Again, to our surprise, the answer is no.”
“We observe that the top domains garner a larger and larger fraction of the clicks and it is not because search engines are surfacing a smaller number of domains. On the contrary, search engines are changing the domains they show. It is users who have decided to visit a smaller number of domains.”
“It should not be surprising that users have learned to trust some domains over others. ”
“”What is surprising is that users click on results from rep-utable domains even when more relevant search results are available.”
“Our experiments are geared towards proving that domains can so drastically influence perceived relevance that users will favor some domains, regardless of content. Viewing content on the Internet as products, domains have emerged as brands. And users have developed such fierce brand loyalty that their clicks are tainted by domains.”
“Our experiments also reveal that search results concentrate over time on fewer domains with increasingly larger share of results pointing to the top domains.”
“This trend is accompanied by an increase in click-through rates (even after factoring out query distribution changes) and is in contrast to the growing size of the web content and the number of registered domains.”
these are the types of whitepapers that add creditability to the domain industry and are worth their weight in gold. mckinseley utilizes this approach and you never see an advertisement for mckinseley anywhere… a third party unbiased endorsement…and a person like frank or maybe even mike mann
could use this type of validity on their domain base..this is how an emerging industry gets to be crediable in the eyes of marketers and madision avenue..
I wonder how much longer it will still take the average person to realize this. Or even the average person looking to buy a domain for their business/website.
Thanks for the article, Mike, and thanks to Microsoft for this piece of gold!
The study seemed to be more about “web sites you already know” than the quality of a domain name.
Maybe click bias will emerge in geo/city/topical/community New Top Level domains.
tours.london would seem highly relevant to a search for ‘Tours in London’ in addition to being the shortest exact match possible.
We’ll see, I guess.
Interesting reading, given Microsoft has been a quiet buyer of some of the best domains in the world (e.g. Office.com).
After reading the study, it seems to me that it would be more accurate to say there is a ‘brand bias’ rather than ‘domain bias’.
They compared click percentages for queries between webmd.com and genetichealth.com and found that regardless of positioning, more people clicked on webmd.com.
Well, duh!
That could be explained simply because webmd.com is a well known brand, not because webmd.com is a better domain name.
of course, people are going to click on the name they have grown to trust as a source for information.
The conclusion should be ‘Branded Domains Matter’ not ‘Domains Matter”.
This study doesn’t really say anything about how users would click any random domain over another random domain, but that they would rather click webmd.com or amazon.com over any other random domain.
Information like this needs to be put out in front of eyes that don’t already have skin in the race. Most of us here already had a sneaking suspicion, if not outright knew the information released in the PDF.
I dont think many people here will be surprised by the findings but if you broke it down into an easily digested bit of information and put it in front of domain virgin eyes… then I think this would really be useful.
Just my thought.
Cheers
In this case, you have to get to page 1 of the search results for this to be relevant. And at which point you are going to run into the amazons and webmds of the world anyway.
Vanilla ice cream is the most successful ice cream, and is crucial to the ice cream industry. Without vanilla there would be 50% less ice cream sales, and many people would never experience the taste of sweet cold ice cream.
This “research” cost about $1,000,000 and is worth about it’s weight in human seman.
People like familiar things, and are generally dumb. Can I get a research grant to prove this? Or can we all get a degree in COMMON SENSE.
I agree with Tom G. I doubt the rest of the posters even took a look at the paper. It’s about brand bias and not really domain bias by comparing well known domains to unkown domains. This doesn’t reinforce the notion that strong generic or even brandable domains are valuable only that already branded ones are.
“In this paper, we uncover a new phenomenon in click activity that we call domain bias—a user’s propensity to click
on a search result because it comes from a reputable domain,
as well as their disinclination to click on a result from a domain of unknown or distrustful reputation. The propensity
constitutes a bias as it cannot be explained by relevance or
positioning of search results”
Savvy.Me anyone?
Well, the paper said:
(1) the domain name matters (i.e. whether you call it “trust” or “brand”, etc.), regardless of the content on the matching search engine result.
The “leap” (which should be tested empirically) is whether “exact match domains” or “category killers” or whatever are *correlated* with the factors that are behind making the domain name matter.
So, if “better domain” equates to “higher trust”, then that should result in higher clickthrough rates, etc. Since Microsoft didn’t study that question directly, it’s open to interpretation.
But, folks know this intuitively. If I email someone saying “I own Moose Pasture Education dot biz, and I’d like to talk to you about doing business together, I’m pretty sure that I’d get ignored. But, when I substitute “School.com”, the response rate is a wee bit higher (or even better, people contact me, instead of the other way around). What causes that higher response rate? Probably all the same factors in Microsoft’s study.
Domainbias.com anyone
LOL, freshly re-registered, check domainbias.org 😉
‘We find that domains can actually flip a user’s preference about 25% of the time’
I think the important takeaway is that url is a significant determinant for click behavior.
It shows that people definitely look to the url before making a click decision.
Google adwords have already determined that exact match, or keyword domains perform better in click thrus
Go to brief.ly and see what catch short urls they provide – it is all about first impression first.
This reinforces what everyone knew intuitively for a long time.
Dirk Lemmons
Dirkster Productions
I think we all knew already, but it’s about getting those who aren’t familiar with this industry to know, and that’s what Mike addressed in his initial statement about this being a domainers best friend when trying to sell to end-users. The interesting thing to me is, the “crticial mass” is just now stepping into the web, going behind the scenes and recognizing the commercial use of domains at a very fundamental level.
Kind of jacks up the whole brand domain vs. keyword domain argument in the domaining world. There is no doubt though that some brands in some areas of focus will never be beat by any keyword domain. Sure, with time and SEO, SMO, the new keyword domain will grow in ranking, but it takes time. However, a brand takes much longer in many niches. So, KW probably still best for smaller endeavors.
Similar examples to Edwin Haywards past research papers
poka3ownage: you made me laugh out loud.
microsoft is a huge company.
there are many, many people there being paid generous salaries to work day after day on things that contribute zero to the bottom line. though they might believe otherwise.
thanks to gates’ deal with ibm, microsoft has never needed to innovate.
however they do need to block (acquire) competition to maintain their monopoly.
do they need 20,000 employees to do that? no.
within the domain name context, the story of how they financed and then killed realnames is perhaps a reasonable example of how microsoft works.
whether one hand knows what the other is doing at microsoft, or whether any one project is worthwhile, probably does not even matter if the context is outside the sphere of protecting the monopoly by acquiring or blocking competition.
it’s neither here nor there.
just my opinion.
What does this study mean to you as a clever marketer creating a new catchy domain name?
Thunder and using exact match domains (EMD,s) is still effective. Thunder is impressive; but it is lightning, not lighning bugs that get users attention most.
Rank doesn’t matter as much as having a quality, trusted domain name.
“Let me just give you a little color on this”:
Strong brands are not at the mercy of Google’s organic search, which is forever evolving.
If Google ever strengthens their search results, then a majority of generic sounding exact match domain sites will be left without the ability to market their content and credibility.
See also:
“Google’s Exact Match Domain Name Patent (Detecting Commercial Queries) By Bill Slawski, on October 25, 2011″
How a company is known to it’s customers is its most valuable asset.
“Domains can so drastically influence perceived relevance that users will favor some domains, regardless of content. Users have developed such fierce brand loyalty that their clicks are tainted by domains.”
Leverage domain name branding to your greatest advantage, because human minds are essentially associating machines.
Your brand, or reputation, is one of the most critical factors of your business success.
It cannot be left to chance. Inspect what you expect.
The personality of your brand determines how people react and listen to you…it even determines how much you sell and how expensive your product is.
Customers don’t buy products, they buy certainty. They buy trust and likeability. they buy perceptions and reputations. “If it isn’t in Voque, it isn’t in Vogue.”
“Users click on results from reputable domains even when more relevant search results are available.”
Your company’s brand or even your own personal brand is what separates you from your competition.
And it’s what will get your customers to buy from you over and over again.
“Now if you are still on the fence, let me just give you a bit of color. that we have looked at the rankings and the weights that we give to keyword domains, & some people have complained that we are giving a little too much weight for keywords in domains. So we have been thinking about at adjusting that mix a bit and sort of turning the knob down within the algorithm, so that given 2 different domains it wouldn’t necessarily help you as much to have a domain name with a bunch of keywords in it. “- Matt Cutts, Google Search Quality Group
As “filkertus” at the SEOBOOK blog commented:
“A strong brand transcends the search engines, has more staying power, is more flexible, and affords the business more control over its destiny.”