The index.html file name offers no value for SEO nor users. I recommend not using it and directing users to a "clean" URL (eg seomoz.org/ instead of seomoz.org/index.html).

As long as you properly set this up and consistently link to the same URL there are not any duplicate content issues. If you wish to go one step further you can add the canonical tag to your page.

would this drop have been an automated deal?

In my opinion 99.99% of ranking drops are automated. Google has clearly indicated their desire is for their algorithms to perform all the work automatically. In cases where the Spam Team manually takes action, it only makes sense that they then work with other teams to automate the process of catching the issue.

Google makes approximately 500 changes to their algorithm each year, so it is possible for your rankings to change without making any changes to your site. Based on your situation, it is likely that one or more of the changes you made negatively impacted your site. There is not any way to know which change caused the drop. That is intentional. Google wants you to examine your site and fix any and all issues, not just one.

You mentioned adding several directory links. Links from quality directories such as Yahoo, DMOZ, BOTW, etc. would generally be considered "good' links. On the other hand, most free directory links are low quality and many are "bad" links. Ensure any links you build to your site are of a quality nature. For now, treat your page as if your rankings are valid and work to improve them. I know that is not what you want to hear at all, but it is the best way for you to get your old ranking back.

@TheBackBurner, the algorithm has thresholds. It's like a balloon. You can puff 30 times and it doesn't matter but then that last bit of air and poof.

How does a canonical work? Does the robot read the canonical and immediately go to the canonical URL or does it continue to read past the canonical tag and get to the no index, follow tag if there is one present?

The first thing to understand is the canonical tag is a suggestion, not an order. While a search engine will usually honor the canonical tag, there are instances where Google or other SEs may determine the canonical tag is not being used correctly so they disregard the canonical tag. Based on this understanding, yes the robot will read the entire page regardless of the canonical tag status.

Is it necessary to have both a canonical tag and no index, follow tag in place? Or should the canonical tag be sufficient to avoid duplicate content?

The two tags you mention conflict. You would never use both tags on the same page.

Noindex means you do not wish the page to appear in the search index. The canonical tag means you do wish the content to be included in the search index, but use the canonical URL in the index.

if both a canonical tag and no index, follow tag are in place, should they be in a specific order?

The order of meta tags does not matter. If a page was marked with both a canonical tag and a noindex tag, the noindex tag would take effect and the page would not be indexed, so the canonical tag would not have any effect.

In short, you want to use the canonical tag to resolve duplicate content issues, not the noindex tag.

To specifically answer your question about the differences between WordTracker and the AdWords keyword tool, I examined the WordTracker site. I performed a keyword search for the phrase "depression and bipolar link". It showed 34 searches. To better understand what that result meant, I searched the site and located the following explanation:

"For the Wordtracker data, the Search count is the number of times each keyword appears in our database of searches over the past 365 days. This constitutes just under 1% of all US search, and the data is gathered from metacrawler.com and dogpile.com."

There are two key differences between AdWords data and WordTracker. AdWords clearly has a much larger data source so it should be more accurate. Also, AdWords data is presented based upon monthly searches, where WordTracker uses yearly searches. The AdWords result for "depression and bipolar link" would be 3 monthly searches. Since the result is less then 100, Google rounds the result to 0.

You are reaching for very long tail phrases. You will capture other keywords and shorter phrases in the process.

For example, while Adwords shows no traffic on "depression and bipolar link" the phrase "depression and bipolar" shows 165k monthly searches with medium competition. If I were to create a page, I would focus the article on "depression and bipolar". If you really wish to keep the focus on "depression and bipolar link", you can do such knowing you will capture traffic from other versions of the phrase.

The disadvantage to keeping a canonical tag on a page which does not require it would be, as a rule, you want to present your web page with the least amount of code possible. Unnecessary code causes extra confusion and adds to the processing time of web pages.

I use the canonical tag on all pages, but not everyone agrees. If you would like further support, SEOmoz uses the tag on all pages as well. If you use any CMS, ecommerce software, forum software or any system which generates pages dynamically then I would highly recommend a canonical tag on every page. At times a system will generate pages which you might not be aware of, but a crawler will find.

Sometimes a page will offer a print version, the ability to sort on ascending/descending, and numerous other changes. You might think you only have one version of your page but have many versions which you do not realize exist. A proper canonical tag ensures the correct version of your URL is always offered for indexing, and you avoid duplicate content issues. With that said, if you have a basic html/css/php site and you are 100% confident in your programmer, then it is not necessary.

EDIT: In your case, it seems the canonical tags are performing a necessary function. From your home page I clicked on your featured item and I landed on the following URL:

http://www.nathosp.com/product/r1212_c

You have the identical page offered under another URL: http://www.nathosp.com/product/r1212_c/hotel_towels.

If you were to remove the canonical, you would have duplicate content issues on your site.

I agree with Steve but wanted to add a bit.

You don't see your true PR. Google updates PR internally on a frequent (?daily) basis, but we do not see it. They update the toolbar that we see 3 or 4 times each year.

The PR number you see is truncated. Your prior PR may have been 40 and showed as 4. Your PR may now be 39 and show as 3. So a 1 point drop may be extremely minor.

Some details from Matt Cutts: http://www.youtube.com/watch?v=h3Jup5R1MGY

@Panu, I believe you might be missing the relevancy of the Panda affected sites' links and their impact on PR. Most of the sites negatively affected by Panda have links to other sites. The value of those links are now devalued, so the PR of the pages receiving those links would also be impacted.

what do I need to do to get out of this "SERP lock"?

Without looking at the site and the keywords involved, we can only offer generic advice.

I would suggest examining all aspects of your onpage factors. Some specifics are:

• page title: focus a single keyword

• header: focus the same keyword

• content: the first sentence of your content should also focus the same keyword

• site internal linking: when appropriate, other pages of your site should provide links in content to other relevant pages.

• url: clean, friendly, static urls which offer appropriate use of keywords is helpful

Wikipedia is a great example for many of the above steps.

There are other items to check. My point is link building and site promotion is an ongoing process which happens over months and years. On page changes have the ability to instantly and dramatically change your ranking. There is a good chance you are stuck due to onpage factors.

The first part of the battle is knowing your links. Each source of backlink information, including Google, has advantages and disadvantages. I have been resolving manipulative link issues for clients since 2011 (i.e. before Penguin) and what has worked in 100% of cases is a backlink report comprised of data from the following 5 sources: Google, Bing, AHREFs, Majestic and OSE. This combination of backlink data has always worked. That is not to say there are not other sources of backlinks which can be used.

Once you have a comprehensive list of all linking domains, the next step is to visit each site and determine if the links are manipulative or organic. Covering that topic fully would be a very lengthy article but in brief:

1. Closely examine Google's Quality Guidelines.

2. Remove any links you, or anyone else acting on the site's behalf (i.e. "SEO", link builder, developer, etc) built.

3. Remove links you created to the site from other sites you own

4. For each link ask "How did this link get here?" If the answer is the site owner independently created the link for their own reasons, that is great. This type of link is a naturally EARNED organic link. If there was any undue influence from the site owner which caused the link to be created, it is likely a manipulative link.

5. Take a look at the other links from that site. If the site offers links to other spammy sites, that is a very bad sign. Would you be ashamed for your customers to see this link?

6. Where is the link? Is it in content (which is strongly preferred) or is it in a sidebar or footer where spammy links are more likely to be found.

There is no hard and fast rules. There are guidelines. Ultimately you look at links and determine if they possess the qualities of an organic or manipulative link.

With respect to automated link identification services, there are two problems. First, they will typically identify some organic links as manipulative. Secondly, they will identify some manipulative links as organic. Both issues are highly problematic. You get what you pay for.

Good Luck.

I'm afraid there is no answer to that question. Here is why.

Each blog article you offer will cover a specific topic. More specifically, the article will focus one or more keyword phrases. The question search engines will decide is which web page is most likely to satisfy a user's query for that phrase.

If your article topic is "Houseism defined" then 100 words is probably enough. You could expand the article to 1000+ words by offering examples, explaining the history of the term, share instances where a House-ism was used in media outside of the television show House, etc.

The more content you share, the more thorough your topic coverage is of the keyword BUT the more opportunities arise to go off topic or dilute your message.

How many words should be in a blog post for it to be optimal for the search engines?

Enough to fully cover the specific keyword phrase target of the page. Always examine the top 5 SERPs for the target phrase. Search results are a competition. Some low quality articles rank as #1 and other high quality articles can't break the top 3 due entirely to competition.

My first suggestion is to drop the ".html" extension at the end of your URLs. It offers no value to you nor your site's users. It just makes the URLs longer and less readable.

My next suggestion is to separate words with hyphens. Use /home-furniture not /homefurniture

With respect to a part number, that is a disadvantage many larger sites have which smaller sites don't experience. Do you NEED to have a part number in the URL? Does it help your employees or customers? Or can you do well with just the product name? If a part number is required, I would at least recommend keeping it down to one number. Your example of "93630" seems fine but your other example of "slpt758-f13" is not desirable.

I would also try to work on your category and product names to ensure they don't duplicate each other. /market-umbrellas/california-umbrella seems unnecessary.

To sum it all up, I would suggest the following for your URLs based on the examples above:

www.vistastores.com/market-umbrellas/california-red

www.vistastores.com/home-furniture/winsome-wood

If product ids are required then add them as -93630

I am working on such a big eCommerce website with 7000+ products. Website is live since more than 3 months. But, Google have indexed only 94 URLs.

It is important to determine the root issue which is causing your pages not to be indexed. It is not going to be your URLs. Check the following:

1. Have you submitted a sitemap to Google? If not, do so as there are clearly some issues.

2. Have you performed a test crawl of your site? Use the SEOmoz crawler. You don't need a 100% complete report. You are trying to review your site for crawl errors which impact your whole site.

3. I would suggest examining your site's navigation. With only 1% of your site getting indexed, there may be a problem.

4. Do you have any content on your product pages? If you just throw up an image with a product name, the page will likely not be indexed.

5. Ensure your robots.txt file is not blocking pages which should be indexed.

6. Thoroughly examine your Google WMT for errors and issues.