Forum Moderators: Robert Charlton & goodroi
Anyone else seen this?
Could this just be a bug in the way cookies are set and read?
Cheers
Sid
We know Google constantly plays with its search algorithm, so i'd assume this is something they are in the process of trialing.
Distinguishing between Brand identity, Corporate identity and just Information or a question would add to Googles useability which is what they are trying to do.
Will be interesting if this is something for the summer.
Can you tell i was booking my holiday at the same time as writing this post :()
i have a client whos page renders exactly the same with capitalised and non capitalised of a website tier 2 page, one has a pagerank 3 one has a pagerank 0
If you're talking about using different cases in the URL - yes, a different case in the filepath does technically designate a different url. Filepaths have always been case-sensitive - that's the W3C recommendation, not just Google's idea. If differently cased versions of a url get into the index, one version is often splitting off PageRank that could be accruing to the other version.
But case-sensitivity for keywords, not urls, is a relatively new phenomenon.
It was a liberation of sorts when engines ceased to be case sensitive, because common usage, not just in search, but in email text and things like usernames, was dropping capitalization.
I think it would be a very strange (and unwise) attempt to turn back the clock by purposely re-introducing case sensitivity for search. That horse was out of the barn many years ago, and I've got to assume that Google knows it. I also can't imagine what they might be gaining, except for some test of human behavior, to make capitalization a factor and not tell anybody.
[edited by: Robert_Charlton at 7:18 am (utc) on May 31, 2008]
SERP output isn't necessarily solely dependent on the user input format (e.g. capitalization), but potentially on user behavior too. The "test-searching" that we do to visually validate listing placements is atypical in that we don't necessarily click-through and that in itself could IMO be the driver for different results when rapid-firing queries. This is Google's opportunity to display different sets of Big 10 results. Why show the same results over and over if no one is clicking?
Results are very similar.
I noted just one or two sites having a different position in the first page result.
Probably Capital letter is just one of the thousands algo criteria, having a very light weight..
I have the same capitalisation thing happening with my site, logged out of my G account I get the following SERP results.
b*** t*** - top of page two
B*** t*** - top of page five (searches changed to protect the innocent:)
I think that it is linking text. I have the lower case keywords in my URL so naturally I have more links with that combination which accounts for the better SERP! I think this is a step BACKWARDS for search not forwards. If anyone from the Big G is following this thread.
I've found no hint of logic in the re-ordering process for this. I accept that it may be some machination beyond my understanding :)
Regardless: I don't think there's a wide-reaching effect, since the vast majority of searches seem to be lower-case anyway, and the re-ordering is minor.
[edited by: Receptional_Andy at 9:10 pm (utc) on April 28, 2009]
Filepaths have always been case-sensitive - that's the W3C recommendation, not just Google's idea.
But domain names themselves are not.
But domain names themselves are not.
I'm not sure I understand, dragonthoughts - it doesn't matter which way you case ExAMple.com - there's only one domain name.
Any takers?
I think that, if you have the domain name www.pinkwidgets.com then you probably have more links with the linking text 'pink widgets' and hence you will rank higher than for 'pink widgets' than you will for 'Pink Widgets'
I don't bite, Andreas8 ;)
It was a while ago, but I tested rankings for capitalisation against on-page usage and incoming links, and found little correlation.
I think the "capitalisation effect" is a close cousin of the "mis-spelling effect". Itself, surely a relation of the "wrong country effect" ;)
I think it suggests something about the properties of a URL. But for an exact cause - I'm stumped.
Isn't a user more likely to Capitalise a 'Brand Keyword' than an ordinary 'keyword' even if it is a Place or a Product?
Capitalisation of search terms could be another way for Google to discern user intent, which is now a major focus, or simply identify brands for future treatment?
I'm wondering whether there's some group within Google that's testing something, without having communicated to the broader search quality team.
Or maybe it's the Search Quality team testing, without having communicated to the Search Engineers. ;)
Or, maybe it's a bug.
Note:
So the SERP's changing due to captials in my key words has stopped. Yes I no longer go back to page 5 for one of my keywords I stay steady at the top of page 2! Lets hope they don't switch the captial algo effect back on again ever!
That would be a captial offence!
Interestingly in Adwords I had an ad group that targeted both Widgets and widgets and Google treated them separately (gave them different click history and served up different destination URLS).
This was a US based search.
