 |
A project to make businesses more aware of their customer experience, and how to fix it. By Mark Hurst. |
| About Mark Hurst | Mark's Gel Conference | New York Times Story on This Is Broken | Newsletter: Subscribe | RSS Feed |
Search this site:
Categories:
- Advertising
- Current Affairs
- Customer Service
- Fixed
- Food and Drink
- Just for Fun
- Misc
- Not broken
- Place
- Product Design
- Signs
- Travel
- Web/Tech
Previous: Discovery Channel book spelling | Main | Next: Broken: coffee machine
June 13, 2005 11:00 AM
Broken: Google URL change
When people ask me for more resources, I often cut-and-paste links to my past newsletter columns. The fastest way has been to Google the name of the column and cut and paste the URL from the results page (as shown in the screenshot at left).
The resulting URL used to look something like this:
http://www.goodexperience.com/blog/archives/000021.php
Recently, though, Google changed the results page so that now the links start and end with some Google tracking data, nearly doubling the length of the address. It now looks like this - too long to paste into an e-mail unless I manually edit the URL:
I imagine that this change helps Google "enhance the service" by tracking where people click on the results page. However, I'm not yet clear on how this is an improvement for the user, the person doing the search. The exclusive commitment to the user experience is what made Google the success it is today; does this change (however small) show a lessening of that commitment?
I'm a fan of Google and would like to be educated why I'm wrong... please discuss!
-mh
I could be *completely* wrong, but I think that you are right in thinking that it helps with determining how many users click on that link. This helps them decide how to rank the pages on their results pages. Users might not care very much, but the site owners do.
(BTW, it still looks the same to me. I did the 'four words' search and it uses the first URL you showed.)
Google has been doing this for a long time. It's only occasional. I think this is probably just the first time you noticed it happening. It's definitely not on for everybody or all the time.
I believe they do this randomly for some tracking purpose, but I don't know what.
Sending a URL through a redirector is a common technique for tracking the clicks to external sites. Google is not the only search engine that does this. The reason Google got to be popular is that it is fast, and is very good at pulling up relevant results. I would expect that the tracking of the search results are clicked corellated to which search terms being used would help them tune their algorithm for ranking. As such, it may impair a secondary use of Google (your copying of links in search results), but improve the primary purpose.
I've noticed this sometimes too. My workaround was to just (manually) copy the green URL text shown with the search result rather than copying the URL. It's not quite as elegant but it works every time.
If you get rid of the top Google frame then the extra URL goes away. (that's for the image search, anyway. if that isn't it then i don't know what you're talking about)
Anyway, MH, shouldn't you just go to the newsletter and copy it from there? It is your site, I'm sure you know your way around. I agree with Mr. Schend.
I'm pretty sure Google does this because it links to it's cache of the page, instead of the actuall page. They probably do that for speed and/or seeing where you click.
That last post is totally wrong. They are not linking you to their cached copy. They're simply recording the fact that you clicked on a search link and then redirecting you to the real website.
As noted above, it doesn't happen that often, and it's standard practice. It's simply not broken.
Side note/helpful tip: When I get long urls, I copy them and travel on over to www.tinyurl.com, paste in the url, and end up with a, well, tiny url to send.
Among the many advantages of a 'tiny url' is that wrapping, which can sometimes mess up a url in an email, doesn't occur, so urls are typically pretty reliable.
The disadvantage, of course, is making the journey to tinyurl.com. (However, I believe OSX Tiger has a widget made by tinyurl that will allow this to be done much easier.)
If you like what they do, support them with a donation.
I've noticed that a couple of times. Why don't you just use firefox/ie's search functionality to point to the search functions of your own website instead? If I type in "abso whatever" into my address bar it automatically searches my website for that hit.
You can also get bookmarklets for tinyurl so you don't need to visit tinyurl before getting the shortened url.
I know what Mark's doing. He searches for his article in Google, and then **right-clicks** (or holds the mouse button down in Apple world) to copy the link **directly from Google's search results**.
I'd argue that he's getting exactly what he asked for, which is **Google's link** to a particular page. I do this all the time myself, for example, here's a book you should buy:
http://www.amazon.com/exec/obidos/ISBN=0465051359/ref=nosim/robbyslaughter/
Note that this is NOT just a link to Amazon. It's MY link to Amazon. I've set up the URL in such a way that I get a kickback from the sales of that book.
The standard for URLs inadvertantly allows you to encode information about the SOURCE of an address as well as it's destination. Neat! Feature, not a bug.
Incidentally, Mark should really click on those links from Google to make sure they actually work and are sending users to the right place, and then copy from his browser's address bar. It's possible that typepad or someone might change the URLs behind the scenes. (Which is a bad thing to do)
I have to disagree with the general consensus here.
Yahoo! and a few others have used these for a while, but I really really don't like them.
I really really don't like them because enough times for it to matter the referrer process will die or be full or 500 erroring or SOMETHING... and then you get a net result of a search result you can't get to.
Wait a minute, Cheryl. There is a difference between *poor design* and *poor execution of good design.* Are you complaining about the design of using this kind of URL, or just the fact that the reliability of the referrer is not high enough?
It's a shame that they're doing this, since it could be avoided and still provide tracking data.
However, I can't post an example of how, because typepad seems to be rediculously restrictive when HTML comments are disallowed.
I agree. I copy from the address bar after reaching the site. And as for the question of who uses Google to search for their own articles, I do. I like to see whether or not the link actually works. I know people who search in Google in order to spellcheck! To each his own.
In Google's defense.
If they have done usability studies or tried to understand user goals in anyway, 80-90% of users probably "search and click" not "search and copy link into email"
So, on a site that has a monstrously diverse user base, should you spend enormous amounts of time designing for 10% of your population?
Well I am sure I am missing something here, but how would the tracking data help Google? The USP of Google was that it based relevance on the number of links that a site had. If Google are tracking how many people click on the results that by their own definition are the most relevant, what does this add?
I've noticed this change about a month ago. Like others have said, it's probably used for tracking purposes, which may or may not be in the best interest of the user. Whenever I see a web site tracking user clicks, I assume it's for some type of marketing purposes, advertising, etc. I hope that's not what Google has in mind here, because it would have a major effect on the quality (or perceived quality) of Google's search results. However, according to Google's privacy policy:
Google may present links in a format that enables us to understand whether they have been followed. We use this information to understand and improve the quality of Google's search technology. For instance, this data helps us determine how often users are satisfied with the first result of a query and how often they proceed to later results.
I think that you're taking the long way around. Thy not just CLICK on the link and then copy the address in the address bar?
Also, I have not noticed google doing this (I have a windows; when I click on the link and find the target there is usally no google prefix)
I'm curious as to how Josh plans to get this functionality without a redirect link. I can't think of any way it would be done (and yes I do know how HTTP and web browsers work). Perhaps it could be done with javascript or something, but that is worse many times over.
My suggestion matches the one above: click the link and visit the page to make sure it's exactly the one you are looking for. Sometimes pages redirect you after you've come in so that you get to the same page but in a new, updated URL scheme. No point in propogating old URLs.
i beleive that link is so if you use personalized home page, it will know how many times you've visitd if you hav search history enabled.
Comments on this entry are closed
Previous: Discovery Channel book spelling | Main | Next: Broken: coffee machine
Not to say you're wrong, but I'd definately like to say that:
1) The longer version of the link *does* work, it's just a little bit longer than the link they used before, and
2) I'm sure very few people run searches JUST to copy the URLs into an email... since its your own newsletter anyway, it seems to me that you'd have a easy way of tracking those on your own computer. Kind of goofy to have to use Google to search a site you write...
Posted by: James Schend at June 13, 2005 11:47 AM