Welcome to the Q&A Forum

linking is a valid concept when used sparingly and when it leads to good information right? I mean, that's what the web is at its foundation.  Since wikipedia is an information portal, I think it makes sense for the site.

I definitely link once to any service page I mention on other pages on my own sites.  You never know what page someone will come into your site on, and in addition to main navigation, linking internally to another page is good for user experience, as you've described.

I do try to mix it up though - not always the first mention, but that's not as critical an issue.  It reinforces relationships as well.

Ambiguity aside, SEO scales when implemented properly.  I have clients with hundreds of thousands of pages and clients with millions of pages.  Many aspects of SEO can be automated.  Of course, someone still needs to provide unique content, but they'd have to do that even without SEO.

Some important factors though include:

• Well structured data architecture (get those table relationships figured out for scalability from day 1)
• Content organization rules (category,sub-category and deeper relationship rules)
• Site navigation & related UX rules
• Standard Operating Procedure for social engagement and link building

Then we can even get into the debate about how to automate at least some of that social / link building effort as well.

Essentially though, with a proper plan and the right resources, maintaining the SEO of a million page site should not be any more difficult than a 500 page site.

An exact match or near-exact match domain name can be beneficial however it can also be detrimental.  It all comes down to how much effort you put into creating a high quality site with strong topical focus, topical depth and reinforcing content relationships, then follow that up with a strong mix of off-site trust and authority signals.  If you don't put in the effort in those best-practices ways, your site will be vulnerable to Google Exact Match negatives.

The best practice here is to create a sitemap index - a single sitemap XML file that lists each of the individual sitemap files.

That then becomes the sitemap you reference in your robots.txt file and to help speed up the process of crawling, submit that through Google and Bing webmaster tools. (You only need to submit that index file - their systems will then "discover" all the individual files listed in it).

See Google's sitemap index how-to.

Fantomaster   http://twitter.com/#!/fantomaster    http://fantomaster.com/

This is most likely a job for a WordPress programmer because it sounds like there's a major flaw in the site architecture.  Andrew Norcross @Norcross on Twitter, and Joe Hall @JoeHall would be the two people I'd reach out to for guidance if it were me.

David,

EGOL has already offered some good guidance. Beyond that though...

Since you've openly posted this request I'll respond accordingly (I do not usually even think about direct pitching in Q&A).

I've been doing internet marketing 20 years, SEO 15. My primary business the past few years is site audits. If you would like to learn more, you can visit my site, where I go into depth about my background and what you can expect from an audit.  Pricing is also communicated there.

And so much for BlackHats having an upper hand in the game

http://trends.google.com/websites?q=blackhatworld.com&geo=all&date=2011

Or those who believe it's all about backlinks

http://trends.google.com/websites?q=backlinksforum.com&geo=all&date=2011

Brutal honesty time.

You're in a very difficult position on this one.  From one perspective, the best practice would be to tell you all of those domains should be shut down, not even 301 redirects.  It's the only "ideal" clean-up of what could very well be deemed mass-scale spam.

However, doing so could literally wipe out a significant volume of existing SEO strength that comes from the links because there's just no way to know how those sites are or are not impacting the main site at this point.

Alternate option:

If there are pages on the rogue domains not on the main site, are those pages that "should" be on the main site when taking the "single site" approach?  If so, those should be moved and redirects set up.  And if that content doesn't really belong on the main site, those pages should be redirected to their parent home page or wiped out entirely.

In this scenario, you're communicating "there's only one site - it's the site we want ranking, and thus we've taken action to kill all the renegade content".

This is the option I would recommend if it were my client.

Except how can you determine all the other rogue domains out there? That would be important to the process.

I would personally also need to evaluate current organic search traffic to help decide what to do.  Is the total organic based visit count significant?  Or is it almost non-existent?  Because if it's almost non-existent, as painful as it might be to do so, I would opt for the first action - killing off all the rogue domains.

As drastic as it is, and as much as some SEOs would say "get what value you can from redirects", it really is the safest way, long-term to ensuring the site doesn't get slammed by Google. And if short-term or even mid-term rankings drop in that process, my position is still "I care most about long-term goals with sustainability and stabilization critical long-term elements.

Whatever you consider doing, I would also explain all of this to the site owner, and have them participate in the decision making process in a way you can be sure they first understand the issues and then that they have no ability afterward to blame you for anything that takes place due to past methods that were obviously ugly at best and reckless at worst.

I'd also love to hear what others here have to say about this and what they would recommend.

There is no such thing as a legitimate link wheel.  I don't care how high the quality of a site is someone acquires. If they buy it then work on a large scale to back-fill links into old content, that is absolutely against Google's concept of legitimate efforts.

Any site that does this on a big enough scale if caught, will go down. Period. No question.

The question instead is - can they get away with it?  Yeah probably. Up until a certain point.  I've seen many go down over the past year and a half.

Just like every other shiny object hack, it all comes down to willingness to risk the entire operation.

I've never trusted compete.com data - every test I've done with their data has ranged from "wildly under-reported" to "wildly over-reported", with the occassional "lucky guestimate" thrown in.

Breadcrumb navigation is always a good idea to have. It reinforces topical focus, and content relationships.

For a blog, Google authorship is good.  As is either the "All-In-One SEO" plug-in or the Yoast SEO plug-in.

Sitemap.xml file and robots.txt file as well.

Then you can learn about canonical tags and get those set up on the site and in the blog.

Really anything in the SEOmoz "Learn SEO" section is worth learning.

Also, a handy reference is the "periodic table of SEO".

I'm in the "yes they have value" camp. We belong to the Vocus small business service and thus send out releases several times a week when we have valid things to share from a non-SEO press release perspective on behalf of our client. Yet we take full advantage of the SEO functionality that comes with it.

Like has been said, as part of a larger PR and SEO strategy, they definitely have their place in the mix, because in spite of all the "junk" scraper sites, a quality release will definitely generate links, views, clicks and even contact inquiries.

We even use them for our own company, not just clients.  http://www.click2rank.com/press/

Money?  Did someone say money?

Actually EGOL something you just responded with caught my eye - "Will the questions be interesting and varied enough to hold the interest of experience people?"

That is a very good question.  Even when forums create "sticky" entries, and FAQs, and such, so many people never read them.  Which inevitably leads to so many basic questions being asked over and over that it tends to burn many more experienced people out.

Yet perhaps over time enough experienced people will stay, or at least be joined by other experienced people on a regular basis.

All I do know for sure, is that the Moz Pro Q&A is one of the best, if not the best system I've seen to date.

First, don't get too lost in the "first link counts" concept in this scenario.  Your site's primary navigation should always be recognized as the most important links across the site, so while you may have custom alternate navigation such as in a sidebar, without sitewide navigation being identified as sitewide navigation, that will confuse search engine algorithms in regard to the most important top level pages on the site.

Next, you can try all sorts of tricks and techniques to obfuscate links, however Google does read JavaScript, CSS, and AJAX so attempting to hide links in those methods would be dangerous at best, and even more problematic at worst.  This is because 1) their system may wonder "why is this site owner trying to bury links across the entire site" and 2) they are far from perfect in how they interpret what they do find in those methods.  That can then cause even more flawed interpretations.

Bottom line - you are gambling with the efficiency of the crawl process and setting the site up for failed identification of proper relationships vital to SEO.

I agree with BOTH of the above suggestions. They are equally important.  And be sure to test speeds across multiple testing tools

• actual Google Analytics "behavior"/"page speeds" numbers (look at the 30 day site-wide average, AND the 30 day speed average for individual pages where you want to put that code).

• WebPageTest.org - run samples by choosing WPT servers in locations where you get most of your site traffic; test for DSL and mobile device emulators, and Chrome as the browser (options on the first screen in WPT). Then look at the "first view" / "fully loaded" time - do it for a sampling of pages on your site.  Also look at the alpha-grades WPT shows for each (a scoring system where "A" is very good, and "D" or "F" is a fail, and where "C" is a red flag that you may have intermittent problems.

You can then go to WPT's "Details" page in that report and see line by line every single process that page runs and where there might be slow-downs for any one or more process.

To add to Christy's well written and quite accurate response, page load time has been a key factor since Panda first rolled out. It may have been a factor before this, however Panda is when it first really took hold as a clear, change = results measure.  For example, I have had numerous audit clients who had severe page load problems, and several of those have seen increased rankings and resulting traffic from organic listings solely by resolving page load speed.

In one client situation, simply by changing site templates (and resulting load speeds), the client saw a doubling of organic traffic literally overnight.  To then test that this was the cause, he then did the same thing with a completely different site that had a completely different model and also saw traffic dramatically increase literally overnight.

However, given the complexity of SEO in 2012, with so many other factors, if there are other serious flaws to the SEO, just increasing page load speeds may not get the same results I've described.

Bounce rates CAN negatively impact rankings, but not purely as a stand-alone factor.  It's when someone comes to your site, can't find what they're looking for, then returns to Google and then clicks to another site and does not then come back to Google.  Which is also the case even if they didn't initially bounce on first landing on your site. They could also have spent more time, or visited more pages before returning to Google to re-search.

Marshall,

Yes, this is moving into the "too much" realm.  So for example, I would suggest that the URLs and Breadcrumb navigation be refined like this:

http://www.pcmedicsoncall.com/computer-repair/

http://www.pcmedicsoncall.com/computer-repair/laptop/

http://www.pcmedicsoncall.com/computer-repair/laptop/screen/

Then in the Page titles, I might go with

Computer Repair | PC Medics On Call

Laptop Repair | Laptop Computer Repair | PC Medics On Call

Laptop Screen Repair | PC Medics On Call

Or depending on keyword research alternately:

Computer Repair Services | PC Medics On Call

(this one above covers "Computer Repair", "Computer Repair Services" and to a lesser degree "PC Repair", "PC Repair Services" and even "PC Computer Repair Services")

Laptop Repair Services | PC Medics On Call

Laptop Screen Repair Services | PC Medics On Call

(of course your actual seeding should be based on keyword research and your particular goals)...

In an ideal situation, the method where you just code the variable into the canonical tag string can work, however it can also lead to problems if the coding framework is not properly planned out or if an upgrade to the system has a bug in it.  I've seen situations where page_uri doesn't function after a system upgrade.

Best practices dictate that you need to ensure the full non-appended absolute URI show up at the source level on a live page so testing is critical if you're going with that option, and be aware of potential unanticipated breakdowns. It's all about what is seen on a view-source or seen the way Googlebot sees it.

The founding principle to SEO is brand identification. The more you do to model your web presence after successful major brands, the more you will naturally earn trust and authority big brands earn. That in turn boosts all other aspects of SEO.

To achieve this specific to subsidiaries, you establish a parent company corporate site, and a stand-alone domain for each subsidiary.  Every site however, needs to utilize the most sustainable SEO methods possible. You can't just slap up sites with a few pages and expect them to rank or pull in highly qualified visitors without serious focus.

Only link back to the parent company site and other subsidiary sites from your "About", and "Contact" sections unless you believe it's valuable from a visibility perspective to link from every page. HOWEVER if you link from every page, they should be nofollow links. if you do mass volume links from site to site and they're not nofollowed, that leaves you highly exposed to potential algorithm penalties.

If you want each subsidiary to succeed as its own brand you will need separate social channels for each as well.  Again though, they'll only be helpful long-term if you have the resources to maintain them in quality engagement ways.

There are many other rules and guidelines (like "keep duplication of content to as near zero as possible") however that's the core concept that addresses your question here.