Welcome to the Q&A Forum

Official SEOMoz complaint: I never did receive my hug from Roger. Just saying...

Update: Hug received via Twitter!

Hi Mickel -

The answer is both. Keep your html sitemap, and keep the link in the footer as is. Crawlers will look at these but they are generally more for human visits.

Then create your xml sitemap (www.url.com/sitemap.xml) and verify it in Google Webmaster Tools.

Hope this helps!

It looks like these guys do a lot more than just concrete repair.  For that reason alone, I wouldnt try and fully optimize the home page for just concrete related keywords.

Since the site itself has very low metrics to begin with, you basically have a clean slate.  Id optimize all of the services landing pages for their respective keywords (i.e. concrete repair, concrete repairs, concrete repair contractors for the concrete page, though I would not triplicate any one keyword) making sure they had good titles, good amount of unique copy and so on.

Its a natural approach to the taxonomy which always works well for Google placement.  It wont get your to rank with on-page alone (except maybe some of the more obscure services) so some inbound links will probably be needed.

Good Luck!

Hi Brian -

Simply use the same analytics account - you can usually copy paste the code, but make sure if your tracking is different (maybe you are tracking a subdomain now, or maybe you are an ecomm site where variables need to be edited for ecomm tracking).

When you add it to the new site, make an annotation in Google Analytics showing that you made the domain change on that date (its just handy when youre looking back at October a year from now).

Dont forget to 301 your old pages to your new ones!

Hi Ilya, I replied to another recent question so Im pretty sure I know you're dilemma

From what we are seeing, averaging thresholds where your kw anchor is less than 20%, and then using Brand & URL, as well as others "click here" to fill in the rest. Brand is good I still wouldn't exceed a threshold of 30% on any one keyword. So for example, a company called Sparkling Flooring, we would use Sparkling Flooring no more than 30%, www.sparklinkflooring.com around 30% (but you can go higher with URLs, we just choose not to), and no more than 20% on 'hardwood flooring', etc.

Combined with excellent unique copy, a good information section like a blog or installation tips section, and other good Optimization techniques, and youre golden.

Hi Joshua,

There are a number of ways to stop Google from counting your dynamic urls as duplicates. Its unclear from your question why you can't use canonical tags for this.  If you went here:

http://luxuryhomehunt.com/homes-for-sale/lake-mary/hanover-woods.html

And add the canonical tag in the HEAD section:

It will solve your issue of duplication when people choose property variables like waterfront or bedroom #. I think you were trying to point out the reason this wont work at the end of your question but Im not exactly sure what you are eluding to there?

James can you elaborate a bit on your answer?

You do lose some juice on 301's so obviously the best course is to contact sites that house the link and ask them to change it to the new URL (once the new URL is live, and you would still 301 redirect).  This isnt always easy to do particularly when there are 1000's of backlinks, so it really depends on how feasible an option that is, and how many backlinks you have (are there only 3? then having an optimized URL will probably be more beneficial than leaving the URL unoptimized, even if you cant have the links changed).

As to 'exactly' how much do you lose? I dont think anyone has a definitive answer. But I have worked with websites that 301 redirect almost every page when they migrate to a new platform and the SEO impact is not severe if done properly.

I still recommend mining your backlinks and having their targets chaged (at least for the more authoratative ones).

Hi Dana - Let me see if I understand this correctly:

In question 1 you asked if this would be a duplicate content issue. The canonical tag retains the exact same URL regardless of the search parameter (and resulting search results). Therefore, regardless of the search being made, Google and other crawlers will not index page with a search parameter since the canonical references to the original url (http://www.ccisolutions.com/StoreFront/category/search-return). This means that when Google accidentally lands here http://www.ccisolutions.com/StoreFront/category/search-return?q=countryman it sees the canonical tag and understands that it should not index this page as it is only a variation of the core page.

This would of course be a problem if you actually wanted Google to index every query page. Alternate methods could be to disclude the query parameter in WMT or Robots.  But the canonical is built in for you so that you dont have to.

In situations like this I also like to add site search to analytics and block the query parameter so no query pages show up as landing pages.

Definitely to reiterate Dana, need more info about the specifics because yes there could be disastrous effects if not done properly.

Besides meta refresh, there are other things to consider. Probably most importantly is the positioning of the home page in search results for the 'product'. If you are ranking top 5 for your product keywords, and the ranking page is your home page, you may find that when you rework your home page to reflect your brand, you lose that positioning.

If you were moving from one landing page to another (i.e. /products to say /products/buy, it would be a bit different because you could 301 the /products page and officially tell Google and other SEs that you are simply moving the page. You cannot, obviously, do this with the root domain (301) or that would defeat the branding purpose.

I would definitely check positioning and revenue for your money keywords first. Also, once you move the product, i would at the very least have a link in the main navigation of the home page that links directly to the product with the appropriate anchor. If you only have one product, or one product set, i would also encourage you to optimized the URL (i.e. instead of /product it could be /

Still need more info to make solid recommendations.

You can always check by testing in your browser but the best way is to check the header response to make sure the server is sending the proper response (a 301) - your landing pages look good (see below). I use Live HTTP Headers which is a firefox plugin - hers what it tell you:

http://pharmacy777.com.au/our-pharmacies/applecross-village/

GET /our-pharmacies/applecross-village/ HTTP/1.1
Host: pharmacy777.com.au
User-Agent: Mozilla/5.0 (Windows NT 6.0; rv:15.0) Gecko/20100101 Firefox/15.0.1
HTTP/1.1 301 Moved Permanently
Date: Thu, 04 Oct 2012 03:23:17 GMT
Server: Apache/2.2.22 (Ubuntu)
Location: http://www.pharmacy777.com.au/our-pharmacies/applecross-village/

So the redirect is working. The only thing i noticed was that the home page instantly switched to www and didnt even return a 301 so it appears you may have implemented a redirect there outside of htaccess.

If your report is still showing duplicates make sure that its not the trailing slash. Your URLs can be loaded as such:

http://www.pharmacy777.com.au/our-pharmacies/applecross/

http://www.pharmacy777.com.au/our-pharmacies/applecross

The best way to find out if the SEOMoz report is counting these as dupes is to Export the crawl report to CSV (top right of crawl report). Then go all the way to the far right column called 'duplicate pages' and sort it alphabetically. This column will show you all of the duplicate urls for each particular URL row. Lots of times you can find little 'surprises' here - that csv report is priceless!

I agree with Dana - the only thing you would need to worry about is if the link was a direct text link using a keyword anchor that is used abundantly throughout your link profile. Even then, if its an authority site that is relevant this is not something to worry about at all. In fact its probably a great link (regardless of the sitewide instances).

Yes this is exactly how you do it:

https://getsatisfaction.com/mindbody/topics/chalk_talk_how_to_setup_google_analytics

Basically all you are doing here is creating an analytics profile, and then entering your user account ID from analytics into the Mind & Body back-end (and Mind body does the rest.

Once that is setup properly and working, you can go to analytics, open the profile you created and then choose 'Ecommerce' to see your sales stats.

You should also create another profile, on the same analytics account, for the WP site. Then you can see how many visits you have there, or how many visits to the plan and pricing page, and can compare it (though not in the same window because they are two different profiles) to the sales data you see in your Mind Body profile.

You can't create a goal to see, for instance, how many people landed on page x, clicked button x, and then converted into a sale, because the site is spread across two domains, but you can still track sales. The Ecommerce section in Analytics will also show you conversion rate, abandonment etc once your visitors are referred to your Mind Body page.

You're opening a can with this question!

The efficacy of meta tags is much debated. Most people believe that the keyword meta tag has no effect whatsoever on SEO, and some believe the same to be true for the meta description.

The original purpose of the meta keywords tag was to help Search Engines understand what your page was about. After years of unabashed over-optimization, the tag slowly became less and less of a signal.

The meta description tag is a brief description of the page, and is sometimes used as the description in the SERPs. There are varying arguments on the efficacy of this tag as well, although it can be useful from a clickthrough conversion standpoint.

I'm sure youll get a lot of varying opinions on this one!

Its definitely hard to say with that many URLs - I would definitely point a few at the sub's home page however. It could be that those sub-domains were cached at such long intervals, that Google simply hasn't checked the site again.

Sometimes, adding the sub to WMT, then submitting an xml sitemap, waiting until Google acknowledges it (and tells you how many are indexed) then removing the sitemap can help.

If and when the subdomains are de-indexed (and theres no reason to believe they wouldn't be), then watch your positioning for a week or two after - if it doesnt change, you have to consider that the drop in positioning may be from another cause. For example, the way that each sorting variable for the products lands on its own static page can be viewed as good for SEO but slightly risky since so many pages are so close to duplicated.

This is a tough one. The worst thing to do is to start with citations and then find out they arent working, and then fix the issue and start all over again (now having two NAPs floating around.

Some businesses use virtual offices in an attempt to rank in different cities or areas. If this is the case with your client, it never hurts to contact that service and explain that you must have a unique suite number. Ive found in some cases they will be quite accommodating.

As for the effects - Ive performed local optimization for clients in this scenario and it still worked fine (and the other businesses using the same virtual office also were in the maps for their keywords), but with constant changes in Local, its risky (in my opinion) to continue without getting a unique address first.

Just my 2 cents!

Right - analytics doesnt actually use the domain in the tracking code, unless you are doing special types of tracking, so as long as you are doing regular Analytics tracking, you can simply port over the code.

Go take a look at your code, and see if there are any setDomain parameters or anything else where your domain is listed right in the tracking code. If not, you're OK.

Also, I would go to the analytics profile and then the ADMIN > PROPERTY SETTINGS and change the default URL (though this doesn't affect tracking code, its good for continuity.)

View the page source (right click + view page source), look in the section.  Youll see it -

<meta name="keywords" content="keyword1, keyword2" />

Tip: if theres a lot of stuff crammed into the HEAD section, just CTRL-F and search for ```
(meta name="keywords") without the parenthesis

Oh - if the subdomains showed no pages indexed, and then all of a sudden at the exact time you dropped, the subdomains showed thousands of indexed pages, then you can definitely assume they are related.

I didnt realize there was such a clear correlation. The suggestions above still stand - you might want to go one further and simply add a noindex right in the robots.txt on those subdomains (make sure its on the subdomains and not the money site!).

Dont forget in WMT you can also do a change of address under Configuration. Youve already completed the first two steps, so you can simply tell Google exactly where the subs have moved.

There's no reason at all why these steps will not prompt google to de-index the subs. The links by the way are simply a 'nudge' to get Google to look at the subdomains again and 'discover' the changes.

Marie is exactly correct - the unnatural links penalty and Penguin are two separate entities.

The best way to combat a penguin related traffic drop is to use your SEOMoz tools to mine your backlinks and then sort by anchor. Traditionally link builders would buy a majority of their links with the same anchor text (or close variations). Website that have a majority of their links as an anchor can get hit by Penguin (there are some instances where this isnt the case when brand and exact anchor domains are concerned, but generally this is the case).

If your links were natural, Google would see links of all types in your profile including "click here" and "www.url.com" etc. You should try and decrease your keyword link anchor threshold to below 30% by either getting rid of spammy keyword anchor links, or building more generic anchor ones.

"So I'm just wondering, since CMS pages are pretty much created on spot and not retrieved from a library, how this affects backlinks and anchor text? How exactly does the external website point to yours if the URL is dynamically generated?"

Firstly, different CMS's create pages differently. CMS just means content management, which means the platform just provides a gui for you to add content or make changes. If you are using WP and creating pages, then these pages wil be indexed as any other page, and links pointing to it would simply target the page's URL.

Wordpress uses permalinks and Drupal uses pathauto to redirect platform generated links into SEO friendly one. They use an internal redirect and the resulting URL is indexed in Google. Therefore, you simply treat the resulting URL as the "real" url, and external links to it work fine.

For whatever reason it seems like that node is not redirecting to the proper clean URL. It might be because the submission was improperly placed, or because the video doesnt exist. All other user content appears to be redirecting fine from NODE > Clean URL.

I think its safe to just delete the user content on this one.

Pay close attention to what Ben mentions about Local SEO. If you poke around long enough with your keywords you'll notice obvious trends in your area (that may not equate to other regions) particularly with the words "in" or "near".

You'll start to notice how certain prepositions trigger the map packs. It's just something to pay attention to depending on your keyword + locale.

Hi Tung - welcome.

A lot of times when you search for a keyword that exactly matches your domain, Google will show sitelinks, as long as the keyword isnt generic, even if the site is relatively new. Looks like you just got cached today.

The link you referenced has 'www' in it, is that how the link is targeted on your website? If so, its probably the double redirect that is causing the issue. Since WP is set to 'non-www' - every time there is a call for the www version of a url, WP automatically 301 redirects it to the non-www version. There is nothing wrong with this.

Its when there is a call for a 'www' version of a URL that has also been redirected, as the one you cited has, where a double redirect now takes place:

http://www.luxuryhome..../sunnyilses.html

to the 'non-ww' version:

http://luxuryhome.../sunnyisles.html

then from there to the new html file version:

http://luxuryhome.../sunny-isles.html

The header check shows a normal www to non-www redirect first (WP is doing this), and then the 301 redirect that changes the sunnyisles to sunny-isles.  Both server responses seem OK so the redirects themselves seem to be working. What you want to make sure of is:

Any internal links linking to the old sunnyisles.html page do not contain 'www'. (And in any event, these links should be changed to point to the new page anyway).

Any inbound links from external sources do not reference the 'www' version.

It would be helpful if we cound see the htaccess file as well.

Great points here by both Matt and Nakul. Of particular importance is understanding who will have access to edit, and site duplication. If you want each of your franchise owners to have editing capabilities for the own store, then it may be easier to use subdomains from a permissions perspective.

As for duplicate content, you'll need to worry about that regardless of whether you use a sub or a folder.

As a case study, Ive worked two very large projects like this, and in both cases I used subfolders for the locations. The folders were the city name in both cases. Each location is positioned position 1 or 2 across the board, and each location shows well in the map packs.

Doug is spot on - its just that easy. Those csv exports literally tell you everything you need to know.

Sameer is also recommending a good tool, Xenu, if you don't want to use your crawl report. However, Xenu can cause issues (rare) with sites with dynamic content, or it can just keep crawling forever with newer dynamic sites.

Don't look for a question mark, go to the site's profile and look for an email - it will tell you that Google has found unnatural links. Keep in mind that sometimes these emails come 4 or 5 days after the [penalty has been applied.You may have to wait a few days (unless this happened some time ago).

If it's Penguin you wont get a notification. Others may disagree but the Penguin recoveries Ive performed were almost exclusively due to a higher than "natural" appearance of keyword anchor text links in the link profile. If you check your backlinks in OSE, and find that you have a high percentage of links that all use your money kw, then yo will need to have those backlinks diversified.

Hi Dana -

I think in the case of Google Custom Search, there is no need to worry about duplication. The reason is that although the rel="prev" etc tags are not being used, a blanket solution already exists: the canonical tag. As you mentioned, the canonical tag never changes, regardless of the search - therefore the crawlers only ever see the Custom Search page as a single page regardless of the queries being made. Thus there is no duplicate issue.

Nathan - the truth is that simply having more social media signals, more links and more indexed pages doesnt mean you will outrank a competitor. Although theses statistics may give you a better DA than your competitors, that metric is used more as a general reference. What you should be doing is breaking down the three signals above as follows:

1. Social Media: OK so you have a lot of 'likes' - are people tweeting, posting on your fb page? Bookmarking? Social activity is the big signal.

2. You have more links - are they relevant? Are they from separate relevant domains? I've had instances where clients come on board and say "I have 5000 links and my competitor has 200 and they outrank me". Further investigation shows that the '5000' links were from only 58 domains, most of which were spammy blog links that were not directly relevant in content (paid blog networks). The competitor however, had 200 links from about 150 domains, and many of those domains were specifically about the vertical. In other words, they were natural relevant links.

3. A lot more indexed pages. Again this means nothing - what matters is that each page has value, is unique and informative according to Google. In fact, if the pages aren't deemed informative and unique, having more indexed pages can be harmful. Just because Google hasn't dropped duplicate/spammy pages from their index doesn't mean that they are 'helping' you rank. In this case you are blogging so as long as the blog posts are unique then you should be ok, but always remember quality not quantity.

Hi tact - what were your subdomains?

You mentioned that you sent in a Recon. Request - did you receive an unnatural links penalty in WMT?

If you have properly 301'd your subs so that NO subdomain page can be accessed, then by simply pointing a few links at the redirect like Ben said should help it de-index faster. Make sure though that the 301's are properly set up (do a header check) and also make sure that no content from the sub is available unless you are certain the the redirect is applied properly (clear the subdomain of files).

CMC is correct - thats how I do it for larger sites.

1. delete all columns except the URL column (col A) and the duplicate pages column (now Col B)
2. in cell C2, enter this formula:  =len(b2)  it will calculate the characters in dupe pages cell
3. drag that cell down to last row
4. select all three columns and sort col c by largest to smallest

Obviously this isn't going to give you an exact number of dupe pages since URL text strings can vary in length, but it does give you a pretty good idea of the worst offenders....

From a strict A/B standpoint where the two variables are:

www.domain.com/california

www.domain.com/state/california

You will see no discernible difference in SEO results. That being said, if you plan to expand the URLs later, i.e.

www.domain.com/state/california/santa-monica/our-service-keywords.* then you should probably consider what length they will be and factor that into your decision.