content management systems

[customle]

How do pages that are created/managed by content management systems fare with SE's?

I have never used one before because I am aware that any "?" or other symbol in the URL of pages would hurt ranking; however, I have over 100 sites that are related and it is becoming a management nightmare.

I have a sign business and I am writing separate sites for each product and each keyword/phrase of each product. It is working well as far as ranking very well with search engines; however, I am very tempted to try a content management system to keep them organized.

Is this a route you would persue and is there a content management system you would recommend or would you stay with the separate HTML pages/sites?

[TheWebDoctor(tm)]

True enough, there are many search engines that ignore anything beyond the question mark (?). The best way to remedy the situation is to use Apache mod_rewrite which will allow for creating search engine friendly URLs.

Apache mod_rewrite can be used on any server operating system to which Apache has been ported.

For instructions on Apache mod_rewrite please refer to http://httpd.apache.org/docs/mod/mod_rewrite.html.

Any dynamic site can be set up to provide search engine friendly URLs. Unfortunately, I haven't worked on Microsoft servers in over 4 years so I'm no help there unless you use Apache as the web server versus IIS.

[customle]

Thank you so much for the information. Unfortunately, I am moving to a Windows server running IIS.

Does anyone out there know how to set up dynamic sites to provide search engine friendly URLs?

Can anyone recommend search engine-friendly content management system for Windows server?

Also, on another subject, do you roughly know how many people browsing the internet that use pop-up blockers, percentage wise? I have a critical part of a program that gives information via a pop-up and it was brought to my attention that many of my visitors might be unable to use it if they use pop-up blockers.

[RDD]

I too am needing a URL friendly option for IIS & CMS drvien web sites.

As for the pop-up blockers: I disable pop ups in my IE options plus I use the Alexa toolbar that gives me the option to block pop-ups.

I make it a strict policy NOT to use pop ups in any of my web designs.

I also help all my family & friends learn how to block as many pop ups as possible.

I personally hope & feel that pop ups will go away one day. I see them as the graffiti of the Internet.

I don't think they will be going away any time soon though. :(

[StepForth_Jim]

I have recently run across major problems with three clients using Content Management Systems for their websites.

In all three cases, the code output by the CMS is just fine and should not prevent SE spiders from tranisiting through the sites but working within the often limited confines of the CMS drives me batty. Each CMS is somewhat different from the others, thus adding the time it takes to learn the system to the job.

I am having problems applying unique titles for each page in the 3 sites I am working on. For some reason, the programmers who designed these sytems seem to have assumed that one title cascading across all pages in the site is OK. There are other massive limitations in moving text around and adding different elements to the site but with a bit of work, most of them can be overcome.

Does anyone have any secrets to working with a CMS that I have missed? I am at the point of wanting to return my SEO work in NOTEPAD files to the my client's webmasters and instructing them to deal with it. (I don't think that would be very polite though)

thanks

[DanThies]

Jim:

If I had the choice, I'd rather work with a properly set up CMS for optimization, because you can structure documents so that they're optimized by the template.

For example, if you can translate a page topic into the <TITLE> and initial heading, force users to input section headings, etc. you end up with a lot of optimized content without having to tweak each page.

The biggest mistake I've seen is redirecting from www.domain.### to something like www.domain.###/default.asp, and that kind of thing that can prevent the site from getting crawled.

[jholcomb]

Most CMS systems that are widely distributed by sources like PHPNuke, PostNuke, YaWPS, XOOPS, are PHP based. I have personally worked with all of them and had no trouble getting an "index.php" page optimized and in top position. However Apache mod rewrite can be used ofr all of the above. As well most of the above systems support Microsoft servers. re-write is another issue but a good windows based web host should have a solution for you.

See a full list of CMS systems here:
http://www.opensourcecms.com/

The trouble comes in when you want other pages listed. Google (in my opinion) still does have MAJOR issues with indexing dynamic URL's. I have seen Googlebot hit dynamic pages repeatedly in attempts to read past the "?" and not be able to make it through. This is NOT a selective problem or a problem generated by webmasters who accidently "restrict" access to forums for example.

On good option is a PERL based CMS. Engines should have no problem reading these. See Slash Code:
http://slashcode.com/

Hope this helps!

[DanThies]

It's not the scripting language behind the system that's the problem, it's those darned question marks. Perl and PHP don't look any different to a search engine.

Using mod_rewrite or similar to put the slashes back in the URLs helps, but more for other crawlers than for Googlebot. Even then, you still can't use session variables in the URL, which may be the most common mistake/problem with CMS systems and forum apps.

Google has pretty much our entire "Inside Out Marketing" site, it's got question marks and ampersands in the URLs. It also runs on PHP, but the scripting language used doesn't have any impact on this at all.

[jholcomb]

The scripting language is the reason those additional characters are in the URL's in the first place.

And I'm glad to hear Google has managed to spider your site with dynamic URL's. Still I'm sticking to my guns saying that Google does not have this problem solved. Until my clients sites start being indexed properly by Googlebot then I'll have to maintain that stance.

[TheWebDoctor(tm)]

I've seen many attempts to create a viable CMS that will be indexed by the search engines. Many times the programmers make mistakes in their attempt to be fancy.

These are the mistakes I have seen. Some may not consider them mistakes while others may agree with me.

1. JavaScript links
2. urlencoded session variables
3. urlencoded links using ? and &
4. no site map
5. lack of adequate text links
6. Server load, response, incomplete page presentation
7. Database not repsonding correctly

That is not in any particular order of importance. But they are things I have seen that can stop search engines in their tracks.

If a search engine attempts to call the same page more than once during the same session/visit then there is definitely something wrong with the presentation/server response. It may also be due to the 'bot' finding the same link with "hash" (#) attributes in the links on a page.

Again, I hope this helps everyone.

[jholcomb]

WebDoctor you are absolutely right in what you said. However in most cases I have checked the servers for their configuration or to see if a hash might be present in URL's. In all cases (except for 2 where there were hashes in the URL) I have found it to be a matter of Googlebot simply not being able to "read" some URL's properly. The disappointing fact is that it seems to be selective with no real pattern to it. While some pages are indexed fine others are left out.

I have even tested the usage of sitemaps and although it does indeed help to some extent it still is not enough.

I think the solution for CMS systems is to actually take into account mod-rewrite when programming the CMS itself. The disappointing factor is that nearly all CMS solutions do not. As many have grown well past beta stages they do not want to go back and take the time to address the issue.

[raygonzalez]

Today more and more sites are utilizing a CMS.

Historically it has caused nightmares for the SEO.

Most CMS do not take SEO in to consideration, so when it comes time for SEO a ton of changes need to be, other solutions including static landing pages or mirror the site for the search engines.

At first we thought it was simply the ? mark, however many of our clients sites have awesome rankings inside of Google even with the ?.
We have been able to get Google to index sites with the ? mark.

A big part of SEO is naming the pages, unique titles and descriptions and naming the images and using the ALT tags.

All of these things can be taken into consideration when building a CMS.

Over the past year we have actually built our own CMS, the entire program was built around SEO. The first few versions were really only for our internal staff and to speed things up.

Soon after that we landed a large ecomerce site that was built using a CMS.
When we first optimized the site we were contracted to optimize 150 pages of the 500+ pages. The system made things really difficult, when the client would create a new page the optimization was not included in the new page.

Our client actually makes changes 3 or 4 times a day to the site.
Depending on what is in stock pages are taken offline and then put back online at a later date. But when new pages were created the optimization again was not included. WIth in 2 months there were only 15 pages with the optimization code left on the live site.

We took the CMS a step further, considering how often the client is changing the site, we intergrated his current CMS with our CMS, this gave the client control over adding the optimization code to new pages.

We went from having to twak 150 pages down to tweaking 20 templates.
Saved us a ton of time and in turn a huge savings for the client.

We built our CMS around the SEO, but sicne we are also a complete web development compnay, many of our clients don't go with the optimizaiton.
Instead of building two CMS we just built the one, however we are able to turn off the optimization part.

Our designers and programmers are a trained it what it takes to create a search engine freindly site. Every site we build is built with SEO in mind whether or not the client takes the SEO.

An added benifit to do things this way is 6 months down the road when they decide they need the optimization a good portion of the set up is already finished, thus creating yet another savings for the client.

The new version which we are thinking of selling on the open market will be ready in about a month. Once it is we will be releasing the case studies and the do's and don'ts.

[ibuiltDOTnet]

Check out the

iBuilt.net Sitemanager

This content management utility will allow you to create complete sites that can be crawled by search engines.

[DanThies]

Originally Posted by jholcomb

The scripting language is the reason those additional characters are in the URL's in the first place.

I'm a little confused. Are you saying that PHP forces you to use them, and Perl somehow doesn't? That's simply not true.

None of the search engines have the problem solved of indexing dynamic sites. In my experience Google is better at it, but that may be more a function of where our sites sit in terms of PageRank, which perhaps causes them to do a complete crawl. Having other sites that link to those internal (dynamic) URLs is no doubt a big help as well.

RDD: if you're still around, you should look at IIS Rewrite (http://www.iisrewrite.com) - it works pretty much the same as Apache's mod_rewrite.

[jholcomb]

Dan I'll take a look at "iis". Someone else had mentioned this to me before and I never got a chance to have a look.

Still, I disagree. I know that pagerank may have something to do with indexing. However I still see Googlebot having problems. When it goes to a page and hits it 30 to 40 times and get only halfway through the URL and stops at a "?" sign then you have to believe there is a problem.

When writing PHP in order to call certain dynamically generated pages raw PHP will force you to use additional characters. CGI is easily changed and additional characters are usually not necessary. I never said CGI had a problem with this.

Rewrite (to my knowledge) is the best way to force PHP to behave like you want it to. If there is a better method then I'd love to get my hands on it.

[TheWebDoctor(tm)]

Apache mod_rewrite doesn't care if the scripting language is Perl, PHP, Python, C variants, or any other language. It is a way the server converts and handles exchanges from files. You could modify your URL responses to be any extension you wish to use.

When outputing a different extension, the server needs to know how to handle that extension. Therefore, in Linux-variants you need to update your httpd.conf file so that the extension is understood. Much like this [ htm, html => pnd, tsf ] or whatever you choose. The point is any file extension can be used. One trick we use is to hide the server type for the less astute by changing the file extension to .exe.

If Apache is running on Windows or Mac then it will have the modifications set up somewhat differently.

[quot]I have a sign business and I am writing separate sites for each product and each keyword/phrase of each product. It is working well as far as ranking very well with search engines; however, I am very tempted to try a content management system to keep them organized. [/quot]

You are using a method the search engines provide advise upon. They suggest you do exactly as you are doing. Then you link all the alternate sites back to the main site and to each other. That helps you in more ways than one.

I would advise against moving your situation to a content management system. That action would hurt more than help. You are much better off doing what you do now. I actually employ that tactic and it works fine, but also gives people other ways to my main site without using "doorway" pages as so many think they need.

[DanThies]

Originally Posted by jholcomb

Still, I disagree. I know that pagerank may have something to do with indexing. However I still see Googlebot having problems. When it goes to a page and hits it 30 to 40 times and get only halfway through the URL and stops at a "?" sign then you have to believe there is a problem.

No question that they still have a problem. Nobody's figured it out yet. The above sounds like they're hitting a bad link or something - they should be fetching the full URL of any link they try to follow. It would be great to see an example, though, where they just can't follow a perfectly good link, so post it if you can.

Originally Posted by jholcomb

When writing PHP in order to call certain dynamically generated pages raw PHP will force you to use additional characters. CGI is easily changed and additional characters are usually not necessary. I never said CGI had a problem with this.

PHP, on many servers, is implemented as CGI rather than as an Apache module, but I think you're misinformed. PHP and Perl/CGI will use the "GET" method exactly the same way. They work the same way, and the language used has nothing to do with what's in the URL. What you're probably seeing are lots of really ugly implementations (like PHP-Nuke), or people using session variables and other unnecessary stuff.[/quote]

Originally Posted by jholcomb

Rewrite (to my knowledge) is the best way to force PHP to behave like you want it to. If there is a better method then I'd love to get my hands on it.

Rewriting does the job, and I don't know of a better way to make PHP, ASP, or Perl based dynamic sites "look right" so that they're easier for spiders to crawl.

[jholcomb]

Dan

Being that we handle many nuke based web sites you may be correct in your thinking that the CMS is forcing the usage. However, PHP/CGI are not the same language. They CAN BE implemented to work together but that is on the server level and most web servers are not set up to do this in virtual hosting environments. They are different modules in Apache. You need a module to make them work together, I forget the name of it but I am sure I could dig it up if someone needs it. Raw PHP will force the use of extra characters in the URL.

As far as hitting a URL goes it usually goes something like this for example:

http://www.domain.com/anypage.php?blahblah=xyz123

Google will hit this:
http://www.domain.com/anypage.php
repeatedly and never make it past the "?". Funny thing is that it is hit-or-miss with no definite pattern. Some pages make it, while others do not. I will point out that this is for pages without session variables.

It seems pretty definitive to me that there is still a problem with Googlebot. Maybe there is something I am doing wrong, but I sure can't see it.