Forum Moderators: open
I am currently getting around 1000 - 1500 visitors/day to our ecommerce site. Because our shopping cart is very basic and sales have greatly increased we have been forced to use XCart which we are very happy with and have almost finished the design of the store to replace our current store on Sunday afternoon/ evening.
What really does concern me is that by moving to XCart i really cannot afford to damage my google PR 6/10 and strong presence from the majority of my static pages. What i really would like to know is if anybody has any tips i could use to maintain my listings. What i am thinking of doing is just replacing all the pages on my site listed in the Google index with a redirect to the appropriate category or product in the hope google will reindex the page from the redirect without affecting the position. Basically i really cannot afford to upload the new site this weekend and then a month later see Google has dropped all my old static pages and just left the home page in the index. Does Google place lower priority over dynamic content?
Any help on this will be GREATLY appreciated, thanks in advance!
Welcome to WebmasterWorld. I think that instead of a redirect with a META refresh, a 301 (permanent redirect) would do better. Unless there is some problem with extreme long URLs (many parameters, session IDs, etc), the ranking should not be a big problem.
I expect that most of the links to your site are to the home page (I don't know your site). If you have a lot of links to sub pages, make sure these pages have a redirect/301, and try to get these inbound links updated to the new page.
We run multiple x-cart sites. Google will index your dynamic pages but you need to upgrade to the latest version of x-cart that has slightly better search engine mngt.
You can possibly optimize your homepage as a static.html.
The current x-cart homepage is written in Php, with the smartie template system, it doesnt work well at all with google.
I would also link off to some static pages with content as you will not get any good points in google for the pages x-cart currently creates.
A good way is to also name each page after your catagory (new version x-cart).
Hope this helps
you could use mod_rewrite to transparently translate all the existing URLs to the new ones, that would be ideal.
Yep, thats right!
A good way is to also name each page after your catagory
One extra remark; if the name consists of 2 words, use a '-' instead of '_' to separete the words. So if you think about making a page called
www.mydomain.com/blue_widgets.php
rename it to
www.mydomain.com/blue-widgets.php
You are now redesigning the site structure, so it is better to do that at the same time. The reason that blue-widgets is better than blue_widgets is that the first one is seen as 2 words ('blue' and 'widgets') whereas the last one is just one 12-letter word no visitor will ever search for.
How do i go about macking a perminent re-direct to every single page?
Again, thanks for the help.
it's seeps that big sites have done their homeword and they separate words with _ rather than - for example check yahoo direcoty: For "Society and Culture" they will have /Society_and_Culture/ and not /Society-and-Culture/
is there anyone else with little homework on how to separate words in static urls? thanks.
/Homeowners/Homeowner_Associations/
Big sites do plenty of things wrong. When you're a PR10 you can get away with it.
That's true. But do realize that for a big site, that after they started this structure, there is no way back. Lots of internal and inbound links, bookmarks, etc. And at that time Google was not that important.
if _ represents one word then why bother to separate them at all?
what is the assumption behind the theory that an inteligent tool like google would understand - to be a word separator but would not understand _?
An underscore is interpreted as a letter. A long time ago I started learning C programming language made by Kernighan and Ritchie. This is what they wrote about a variable name back in 1978
Names are made up of letters and digits; the first character must be a letter. The underscore "_" counts as a letter; it is sometimes useful for improving the readability of long variable names.
Most of the source at Google is written in C++ (which is based on C). So for at least 25 years the underscore counts as a letter.
the problem is i just switched form dynamic to _ and not from _ to - i hope ms. google does not get upset. i will send nice flowers to googleplex.
takagi, I don't think that C++ is the reason for "_" counting as "_" and "-" counting as " ".
If Google WANTED to treat - diff than _, then they can and will, but there is no reason I can think of that would FORCE or require them into that situation.
Personally, both _ and - are used as separators, but I use _ exclusively when needing to separate words in variables. So if given a choice as to the logical separator between the two, I'd say _.
Where is this information coming from that is telling us that - is required to separate words for google?
see [webmasterworld.com...]
takagi, I don't think that C++ is the reason for "_" counting as "_" and "-" counting as " ".
If the software was written in a different language, they might have made the same decission. But the concept that an underscore counts as a letter is older than search engines. When I use Google as a user, I sometimes search for parameter names with an underscore. Usually Google will show me more relevant pages than other search engines because Googlee treads the underscore as a letter. But let's face it, most visitors will never enter an underscore in a search query.
Anyway, I don't pretend to have seen google's source code or have spoken to any of the programmers, so this is all speculation.
I still have no reason to think that google is forced into considering _ diff than -
I think they are not really forced to do so, and might quite easily change it. But at this moment it could improve your ranking on Google a little bit to rename a file or directory with an underscore.
With an old printer or teletype terminal, you could only get a bold character, if you would sent the character, followed by a backspace, followed by the same character. You could only underline a character by sending the character, followed by a backspace, followed by an underscore. That was the original reason to have an underscore; to underline/underscore some text. Therefore an underscore in a text didn't indicate a word had ended.
My reasoning for the C++ comment is related to this thread [webmasterworld.com], which wasn't posted at the time. I think that Google choose these things on purpose.
This query:
allinurl: search_engine
returns as though the underscore is a letter and it is all one word
This query:
allinurl: search-engine
not only treats the words separately, but Google shows sites in which neither word is in the URL (like google.com)
When you put it in quotes, this shows the hyphen is treated as a space:
allinurl: "search-engine"
OR
allinurl: "search engine"
both return the same results. (and now the words show up in the URLs)
Along similar lines: I run some dynamic sites and have found that Google does not mind the? in the URL but does not like to see a & which usually marks an action or parameter (to display a product in a shopping cart, or an article in a newsletter dbase, for example). Has anyone else found that the ampersand in a URL is Google poison?
AlltheWeb is not having a problem with it. All my dynamic pages are getting in there. Google take note!
Again, you can try an allinurl search for
allinurl: "?action"
which ignores the question mark as though it were a space
allinurl: "&action"
which returns no results
