Where to find statistics proving shorter, easier-to-remember URLs get more visitors?

December 1, 2005 10:15 AM

I am trying to make the case for an organization that it is important for them to have a shorter, easier-to-remember URL for their new website. It seems obvious to me that a website like this would get more visitors, more word-of-mouth spread, and more repeat visitors as well--but I can't find any statistics on this in my searching. Can anyone point me to a study or report which may be useful? Thanks!
posted by spinpapi to Computers & Internet (13 answers total)

I don't know of any 'stats' but you shouldn't have much trouble finding reports to back you up. Here's one.
posted by justgary at 10:33 AM on December 1, 2005

In all seriousness, have you considered making up a study? It seems to me that anyone who doesn't immediately see the value of an easy-to-remember URL isn't the sort of person who would go digging too hard if you told them that a study claimed whatever you wanted it to.
posted by Faint of Butt at 10:40 AM on December 1, 2005

FoB's right. Get 4-6 of the site's users to come in. Bribe them with $20, or a couple slices of pizza for that matter, to give you an hour of their time. Turn on video camera, ask questions. If you plan it right, you'll not only have ample evidence to settle this issue, but a bunch of other good leads on how to improve the site.
posted by nakedcodemonkey at 10:50 AM on December 1, 2005

I think what FoB meant is to merely pretend that a study had been done, not to actually conduct one.

That said, I once had a client (GlaxoSmithKline) who was unconvinced that just because text is blue and underlined, people will understand that it's a link. This was in 2001, mind you, not 1993. A study was necessary to prove to them that, really, it's OK to just write "Prescribing Information," and not "Click here for prescribing information," &c.
posted by waldo at 11:13 AM on December 1, 2005

Jakob Nielsen recommends short URLs. You'd probably have to pay to get access to the usability studies this is based on, but his report is probably convincing enough. Among the many arguments: a URL longer than 78 characters will wrap in email, which may break the link for some people.
posted by teg at 11:44 AM on December 1, 2005

Do you really have to convince these people? Can't you just shorten the URL and get on with it?

I don't mean to be blase, but I can think of lots of ways to make a URL shorter, without even nessecarily rendering the old URL inactive. And for a lot of clients I've helped with IT issues, just having it done already can take the wind out of the "should we do it?" conversations.
posted by chudmonkey at 11:50 AM on December 1, 2005

I've got to say that most studies are unconvincing. Most of them are done by setting up memory tasks for the subjects, when the best studies should be done by processing site visitor logs. If I am remembering the last stats I ran on this three or so years ago for an advertising client, we determined that what with bookmarks, autocomplete, and the tendency of people to find new web sites via other sites or via email, far fewer than one percent of visitors to a web site averaging more than 100,000 visitors a day hand-typed the URL to get there. I don't remember the exact percent but it was a decimal, which, when factored with return rate and the buy-through ratio, affected the bottom line not at all.

However, a short URL is good for other reasons:

--Easier to design with.
--Easier to rattle off on the phone.
--Easier to "own" a single word, meaning the URL is an extension of brand identity, whether or not people ever type it in.

Also, I am a HUGE proponent of buying all possible URLs and then redirecting them to the main URL.
posted by Mo Nickels at 2:22 PM on December 1, 2005

There are definitely a lot of benefits of short URLs but remember that larger and larger numbers of visitors will be coming to your site via search engines.
posted by clgregor at 2:31 PM on December 1, 2005

I would agree with clgregor. Your PageRank is more important than your URL, so time spent debating URLs instead of performing SEO is generally wasted. The length of the URL doesn't effect PageRank or its equivalents.

I have a couple sites with both URLs like "hostname.subnet.domainname.tld" and URLs like "www.sitename.org". For promotional purposes we use the short URL but for search-engine purposes we use the long URL (because the TLD is .edu for one thing). Folks that hear about the site through a conference or seeing something in print generally get to the site by searching for the site "name", not typing the URL directly. Despite using the short URL for external promotion, we get 10 times the number of hits on the "hostname.subnet.domainname.tld" address that search engines use.
posted by McGuillicuddy at 3:10 PM on December 1, 2005

Despite using the short URL for external promotion, we get 10 times the number of hits on the "hostname.subnet.domainname.tld" address that search engines use.

Part of that is from lack of computer experience. Tell my grandmother to go to yahoo.com and she'll open up google and search for it.

That certainly doesn't mean a shorter url isn't better (for about a thousand reasons), some already listed. Also, some sites by their own nature or more difficult to search for.

Go for the shortest, catchiest url you can find. Suggesting otherwise is nonsense. You don't need a 'study' to prove it.
posted by justgary at 4:01 PM on December 1, 2005

Part of that is from lack of computer experience.

I'd argue the exact opposite. Searching is generally easier for everyone. My home page is Google and I use Firefox and its little Google search box. If I want to go to Yahoo, I can click in the Google search box and type "yahoo" and get there just as fast with less chance of making a typo (obviously bad example, news.bbc.co.uk is a better example). Especially in terms of New Product XYZ, if I'm interested I'lll remember the product name or idea but not the URL.

Not that I'd say you shouldn't get a memorable URL, but I agree with Mo Nickels. The difference between www.productxyz.com and www.companyx.com/productxyz will be hard to notice in most cases.
posted by McGuillicuddy at 6:35 PM on December 1, 2005

It seems obvious, yes. But:

> Another surprise from the contest was the long domain names of some of the most popular entries. Contestants were allowed to register any domain name available, as long as they mapped it to the ContagiousMedia.org server for traffic-tracking purposes. But the winning entry was at the domain forgetmenotpanties.contagiousmedia.org until recently, as was ringtonedancer.contagiousmedia.org. It shows the power of hyperlinks on sites, and links spread by e-mail and instant messaging.

"We all had different hypotheses about what would do well," Peretti said. "A lot of people thought that Forget-Me-Not Panties wouldn't spread, and sites that had the contagiousmedia.org domain wouldn't spread. But Forget-Me-Not Panties and Ringtone Dancer both spread without their own domains. There's a lot of conventional wisdom about the Web that's wrong, like people think you have to spend hundreds of thousands of dollars for a short catchy domain name, and people have to remember that. There is some value to that, but even with a long name like forgetmenotpanties.contagiousmedia.org, it can be the #1 site in a contest."

I think a short URL is important if your main promotion is print. It's good for billboards or buses. But if your promotion takes place on the web or in e-mail, there's little advantage. Ultimately the shortness of the URL is irrelevant if the site itself is of no value to visitors and has low stickiness.
posted by dhartung at 10:12 PM on December 1, 2005

Thanks for the link, dhartung. I knew about the contest, but hadn't checked back for the winners or analysis.
posted by McGuillicuddy at 7:00 AM on December 2, 2005