Welcome to the Q&A Forum

Hey,

The same basics apply but I've found that links to the domain and authority of the domain matter more than page authority. In terms of fast indexing, make sure you have an RSS feed. Also, if you are trying to get accepted in to Google News, do make sure you have met those requirements: http://support.google.com/news/publisher/bin/answer.py?hl=en&answer=40787. Having a news specific XML sitemap is also helpful: http://support.google.com/webmasters/bin/answer.py?hl=en&answer=74288

Why would you remove the content a day or two after publication? If you have an online newspaper, that content may still be relevant for some time after the original publication date. I've tended to keep news content around years after original publication and still gotten some traffic to those articles.

Given that, I've not has experience removing content that quickly but when I have removed content from news websites, I try to redirect the old url to the most relevant new article (if possible) or news category page (usually easier to do). This preserves link value, which is obviously important. For really old stories that don't have links or traffic, I set the article to return a 410 status on that URL to signal that page is removed permanently. Again though, I'd default to keeping the article around as long as possible.

Hope that helps.

Hi Johan,

A 410 response code is perfectly acceptable for expired pages. With a 410, you are communicating that page is "gone" and expired content usually is "gone", so it fits. However, with 410 you are going to see that page fall out of the index and that page will lose traffic (assuming it would get any given that some expired content likely won't get any traffic since it is no longer timely) and, more importantly, lose link value (if you had any links to those pages).

As for 301 redirects, I'd start tracking visits to the 410 expired page and links to the 410 expired pages. How much traffic are you getting? How engaged is that traffic? How many links are there and are they good quality? Links are easy enough to track in OSE and for tracking traffic you can use Google events (http://antezeta.com/news/404-errors-google-analytics).

When I see a lot of links or a lot of traffic (especially traffic that leaves), I've converted a 410 page into a 301 redirect that goes to our best (programmatic) guess. For instance, 301 redirect the user to a search for the properties in a similar location or similar price range (or etc.). What I've often found is that when I get the the user redirected to the best page, I'm more likely to see them engage and use the site. Along with the user benefits, I've also see that help with overall organic performance when there are a lot of links back to these pages.

Hope that helps. Thanks,
Matthew

Hey, You are solving multiple problems. The code looks properly written to solve one of those problems--the naked domain to www domain redirect. So long as going to http://mydomain.com 301 redirects to www.mydomain.com, then you know that piece is working.

The second "problem" you have is that you can reach your home page with /index.html and without /index.html in the URL. So long as only one is indexed by Google, this isn't that big of a problem. You should however put in a canonical on your home page to make it clear which version you do want indexed. Then make sure all internal links go to that URL.

Alternatively, you can 301 redirect /index.html to the root via the htacess file. That code would go something like this:

RewriteCond %{THE_REQUEST} ^[A-Z]{3,9}\ /.index.html\ HTTP/ RewriteRule ^(.)index.html$ /$1 [R=301,L]

The third "problem" is that your home page can be accessed with or without a trailing slash. Again, I wouldn't say this is a problem unless both versions are indexed. And, like the home page, you can fix this by adding a canonical element to the home page and link to that canonical consistently within your site.

However, this too can be fixed via htaccess. Here is an example of the htaccess code:

RewriteCond %{HTTP_HOST} !^.mydomain.com$ [NC] RewriteRule ^(.+)/$ http://%{HTTP_HOST}/$1 [R=301,L]

Thanks,
Matthew

You can do this via RedirectMatch statements. There is an example of different match types you could use here. Basically, you create a pattern and then the code is smart enough to redirect based on that pattern.

http://www.askapache.com/htaccess/301-redirect-with-mod_rewrite-or-redirectmatch.html#301_Redirects_RedirectMatch

Your redirect might look something like this though you might want to adjust the pattern match the directory in between /reviews/ and /item/. Mine is just a basic alphanumeric catch all.

RedirectMatch 301 ^/articles/reviews/([-A-z0-9]+)/item/(.*)$ http://clients.qwconsulting.com/reviews/$1

I hope that helps.

I would say it depends on what the website is and the nature of the links.

Are the multiple links all on the same page or are they on several different pages of the site? Getting links on many different pages of a website would be more beneficial than getting multiple links on the same page.

What about the placement of the link to your website? Is it just a link in the directory or is the link part of a highly relevant blog post? Multiple links inside a relevant blog post can continue to add value if those links go to different pages of your website.

Also, it is worth considering if it is a high authority or low authority website. If it is a high authority site, then multiple links would be worth pursuing. If it is a low authority website, there wouldn't be much value to multiple links from that kind of website.

Hey,

I don't think you are safe from an issue with cannibalization just because the one tool says you don't have that. From the situation described, it sounds like you do have issues on your site with overlapping content and that could be hurting your ranking (as could other factors too, obviously). The best way to find out is to remove the content that you think is overlapping or self canabalizing and then see if you regain rankings.

I'm not certain how the cannibalization check happens in SEOmoz's on page tool, but I have seen instances where cannibalization is missed if those pages are not linked together. As I understand it, that on page tool checks only the one page as opposed to an in depth site crawl.

Hope that helps.

Matthew

Hey Marie,

If it wasn't for the Panda situation, I would say you should 301 redirect the subdomain back to the main site. I've had to do that more than once (unfortunately) and a redirect has worked every time to remove the development subdomain from the search results.

However, because of the Panda flag, that may result in both the main site and the subdomain site getting flagged by Panda and thus both sites falling out of the index (which might make it even harder to regain).

Here is one thought. Is it possible to redirect a few pages on the subdomain and put a canonical on a few other pages on the subdomain? That way you could see what happens with those pages. If I were you, I would test out both options before applying to the whole site and I'd test on the most insignificant pages there are on the site (privacy policy, terms and conditions, etc.)

This is a really interesting question and I'll be anxious to hear what others have to say. Please let me know what ends up happening and how you go about fixing this.

Matthew

That is a good point!

On that though, what would the more preferable placements be for a multiple link situation? I would say it would be better to have multiple links in the content area as opposed to multiple links in the footer. Or, do you think a distribution of links would be better...for example, one link in the content and another in the footer?

Hey,

You can do cross domain canonicals. Google does support that (http://www.youtube.com/watch?v=zI6L2N4A0hA) and the one time I had to use that, it did seem to help. That being said, I'm not sure if Bing supports that or not.

Hope that helps.

Hi Joel,

Yes, you can use multiple h1 tags across a page. From what I've seen, it doesn't have much, if any, impact on your search performance. As for HTML5, it is completely valid to have multiple h1 tags or to structure it with an h1 per section.

All that said, I still prefer to have only one h1 tag per page. Realistically, each page can only have one heading that is the most important. However, this isn't about search or usability or programming requirements, it is just my preference because it helps me prioritize my content.

Thanks.

Sadly, I think that is part of what you have to do unless you want to risk the paid link route.

Usually in these situations, I've tried to find alternative ways to rank in areas that the competitor with paid links is ignoring. For example, targeting the long tail searches is usually a good option as many paid link setups narrow targeting only a few keywords. Another route is to build up up social proof (including reviews on local sites if it is a local business).

While there is still a lot of work to do in those areas, that work can at least see returns as compared to competing against somebody with thousands of paid links.

I guess it depends on what is the intention behind these pages? Are you attempting to get more traffic for search? Or, are these pages simply for people visiting the site? If it is more about people and not search engines, I wouldn't really worry about the duplicate content issues (provided the pages make sense to users). Adding a nofollow to the links referencing these pages (and a noindex,follow on these pages) will prevent Google from crawling unnecessary pages and indexing unnecessary pages.

I've had similar situations come up (it always seems to on large sites). Typically what I've tried to do is identify which pages have the best shot at getting activity within the search results (based on keyword research, link research, etc.). On those pages, I do everything I can to have really unique content that sets those pages apart (I'm talking bigger scale changes, as opposed to geo-targeting changes). For the other pages where there isn't such a great chance with search engines, I use the nofollow tag to prevent Google from accessing those pages. That seems to help keep the focus for SEO on the right pages.

Hi Richard,

Updating htaccess files can be tricky, especially with that risk of wrecking your site if you do something wrong. So, not a dumb question at all.

WordPress is a great way to add redirects, but if you are dealing with hundreds that can equal a lot of time.

So, if you are looking for a way to quickly build the htaccess file vs. doing everything manually in WordPress, my trick is use to a concatenation function in Excel. That way I can write hundreds of redirects at once and then copy the resulting text into the htaccess file.

The general idea is you have your old URL path in col A, the new URL (full URL) in column B, and then in column C, you'd have a formula that looks something like this (obviously line numbers might change):

="redirect 301 " & A2 & " " & B2

You could then copy column C and add that to your .htaccess file. Of course, back up the htacecss file first. The only word of caution would be to make sure you have valid paths in column A (no special characters, no full URLs, etc.).

Hope that helps.

Yes...but, that rests on a couple big assumptions. First, that all 100 pages are equally important. Second, that all these websites you are blogging on are all equal (same authority, same relevance, etc.)

However, it is likelier that some of those pages are more important than other pages. It is also likely that some sites are more authoritative than others. Given that, you'd want to promote the more important pages more often. And you'd want to promote the more important pages on the more authoritative sites.

Yep, on two different sites we did thousands of redirects at a time with no issues. In one case it was annual and the other it was quarterly but I don't see any reason monthly would be any different.

Definitely post your findings after implementation or maybe even write a YouMoz post about what you find out!

Hey Richard,

You are really close! In column A, do a find for "http://www.global-lingo.com" and replace with nothing. That way you are left with:

A = /news/2009/12/news-internet-translations/

B = http://www.global-lingo.com/blog/

C = redirect 301 /news/2009/12/news-internet-translations/ http://www.global-lingo.com/blog/

Thanks,
Matthew

Are you referring to the ratio followed vs. no followed links? There are some good answers about that here.

http://www.seomoz.org/q/does-the-ratio-of-external-nofollow-links-to-external-do-follow-links-matter-in-terms-of-serps-ranking

Hi,

Here is an article explaining how to do this in robots.txt:
http://sanzon.wordpress.com/2008/04/29/advanced-usage-of-robotstxt-w-querystrings/

Depending on what you are trying to do, it might also be worth investigating parameter handling in Google Webmaster Tools:
http://support.google.com/webmasters/bin/answer.py?hl=en&answer=1235687

Thanks,
Matthew

Hey Kat. I'd tend to agree with Andy's response that moving this into one site makes sense. You could then redirect the local pages into the appropriate page on the main site, creating a single authoritative domain. You'd want to make sure as you do this that each local page is truly unique...if the Chicago and Detroit local sites both contain similar pages about a dog safety campaign, you'd want to consolidate those (probably) into a single dog safety campaign page that could work for both locations (or find a way to distinguish the content for each location).

Now, having gone through a few projects like these, I know full well that what is best for SEO or UX isn't always the most popular solution. Internal politics play a role, and I'm guessing you might be in a spot where some of the local chapters don't want to relinquish control over their site. So, as an alternative solution you could look for some other ways to link these sites together. For instance, you could have a dog safety page on the main site that all the local chapters can link to and possibly, let the local chapters adjust the content slightly via the link. As in, if the url includes a query sting referencing Chicago (maincompanysite.net/dog-safety.php?location=Chicago), the content on that dog safety page could be programmed to show Chicago's phone number and address. That way you have a definitive page that can rank, but also a way for local chapters to share that content through their domain.

A lot of that alternative comes down to content governance rules, communication about who creates what page, a clear understanding of how people link to sites, and probably some clear design/brand standards. It is a bit of a mess, but not an unrealistic reality if you can't convince all the stakeholders in the value of consolidating to a single domain.

Hope that helps!

I've had issues getting images indexed as well. A few thoughts:

--> It could just take time. It has been a week since you posted, so if you are still having this problem, then clearly something isn't right.

--> Along with checking GWT, have you been searching for the image in a Google search? Sometimes GWT doesn't show the most accurate number.

--> Have you verified that your image sitemap is valid? Sometimes a simple mistake can prevent indexation. If you want to send me the link via private message, I'd be happy to take a look.

--> Obvious question, but have you checked things like robots.txt to ensure images can be accessed by Google's bots?