Welcome to the Q&A Forum

When you say "clean up" do you mean removing the links or disavowing them?

You will never be able to get them all removed, so in the end you will need to a Disavow anyways. If your time frame is short, you may want to make Priority One be doing a Disavow for each of the 50+ sites you are working with. Then you can proceed with attempting to get the links removed. I have not heard that there is any downside to having a link removed that already appears on your disavow file...

As for the order of the Priorities, you may want to shuffle them a bit depending on the different situations on the different websites. I suggest you read this Moz Blog article called It's Penguin-Hunting Season: How to Be the Predator and Not the Prey

...and then test a few of your sub-pages that used to rank well at the program used in this article which is called the Penguin Analysis Tool.  I say sub-page because it needs a single keyword phrase you want rank that particular page for so it do the anchor text analysis. And that works better on focused sub-pages than on general homepages. $10 per website will let you fully evaluate two typical pages on each and see which facet of the link profile is most valuable to attack first.

Hi Matt,

It sounds like you ran Link Detox? If not, then it is possible to get a $50 for two evaluations deal with them that will give you pretty reliable list of bad links to compare to the one you created.

I also suggest you read this Moz Blog article called It's Penguin-Hunting Season: How to Be the Predator and Not the Prey

...and then test a few of your sub-pages that used to rank well at the program used in this article which is called the Penguin Analysis Tool.  I say sub-page because it needs a single keyword phrase you want rank that particular page for so it do the anchor text analysis. And that works better on focused sub-pages than on general homepages. $10 will let you fully evaluate two pages and see if the data helps you.

I did a Disavow months ago and got a little recovery, but not 100%. Fortunately I only got nailed for a few keywords that did not convert very well anyways.  After testing a few of my own pages in the tool above, I wonder if maybe I Disavowed too many of one type of link and not enough of another type.

About doing a Reconsideration Request. I tried it just for the heck of it after submitting my Disavow. An hour or two later I got an automated email response saying that their was no manual penalty and therefore no action was needed. So it probably doesn't hurt, but evidently doesn't help either. Basically they are for those who get a notice.

Hello Chiaryn,

From what I see on my campaigns, what you are saying is only true if a site is all redirected to the root domain version, or if both the root and the www version are active.

If a site is redirected to the www version, a campaign that uses the root domain will show faulty data. Please see the images from my Campaigns for my site (which is redirected to the www version) that I have attached for proof.

The old Moz Pro campaign (shown on the left) is set up for the www version of the domain. The Moz Analytics campaign (shown on the right) was set up for the root domain. The data on the right is much more accurate for Page Authority, LRDs, Total External Links, and HTTP Status. The other odd result is that few of the Top Pages match from www-campaign to root-campaign.

Please let us know if I am interpreting these things wrongs, or if I have the campaigns set up incorrectly...

www-vs-root.jpg

Hello David, Thanks for jumping in!

The old URL was  http://www.easydigging.com/Garden_Cultivator/wheel_hoe_push_plow.html  and the new one it 301's to is  http://www.easydigging.com/wheel-hoe-hoss.html   Please reply back here so I know what the situation is.

Since it appears that Index updates only happen monthly, and it takes a few months for data to fully migrate over to redirected URLs, it would be really nice to change the little message that shows on OSE when an analyzed page is redirected (currently something like "Do you want to see the data for page this redirects to") to something that describes the lag, like.. "It takes a few months for Moz data to transfer to redirected URLs, so be sure to check both the old and the new URL since the one with higher statistics is the accurate one"

Hi David. Or it can be done more simply (If you have G+ pages for both your self and you business) by just putting these two lines in the head of your page

where publisher is your G+ biz page.

When setting up Moz Analytics you should target it for whichever version of your domain you use.

If all your pages redirect to the www version, and all the search results featuring your pages have the www in front, then that is what you should set up Moz for.

This is not real easy as the set up screens seem to assume that you want to put in the plain non-www version. So you do have to enter the www version in the "What site do you want to track?" box on the first page of the set up screens.

When you get to the Competitor section, look and see if your competitors search results show up with www or store or blog or anything in front of their domain name and enter them the same way. That way you can compare apples to apples.

Hi again Chris.  Mostly I was wanting to use Moz to evaluate my site for SEO reasons and frustrated that I kept having to use "old" page URLs to get data.

Also I have just started to experiment with the Penguin Analysis tool from Virante, and it uses Moz data mixed with data from other SEO programs - so I get corrupt results there whether I input the old URL or the new URL.  http://www.penguinanalysis.com/

If they really updated the Mozscape index on the 18th, it seems it should show by now...

Thanks Chris. But unfortunately that makes the situation even worse. I did these 301 redirects weeks ago, and your quote say that there was a Mozscape Index update was a few days ago on October 18th.

I just checked again in OSE and the old page data and strength still has not transferred over to the new page. I have attached a screenshot from OSE showing this.

Any ideas?

page-difference.jpg

I redirected a few pages that had funky URL issues (capitals and underscores and useless words) to the same page renamed with proper URLs. The sitemaps were changed also, and re-submitted.

Like this:  mydomain.com/Not_Pertinent_Words.html  >>>  mydomain.com/good-words.html

Google seems to have found them and changed the search engine results listing in about 8 days.

But it's been about a month now and Moz and OSE still have not transferred all the strength and link data from the old URLs to the new ones.

Question 1: How long does it usually take Google to transfer all the link and strength data for a 301?

Question 2: How long does it take Moz and OSE to do the same? Is there something I need to do to tell them about the changes?

It seems I recently read that Safari had some sort of code change that was causing GA to not track visits from those systems. But I have no idea where I read it. Probably worth a google search for something like "safari bug with google analytics"....

I checked the AquaFarm page in your image.

My understanding is that G+ prefers to use schema data as it's 1st choice for presenting, but it does not look like you have schema data on your page.

Then it goes for Open Graph data (like Facebook uses) and you do have that. But when I looked at the image linked with your   meta property="<a class="attribute-value">og:image</a>"   tag it is only 56x56px - which is maybe too small.

I know that the minimum image size has been changing recently, so I do not know exactly what it is right know, but I have been using 300x300px images successfully.

If you want people to be able choose which image to display with their post, you will need "tag" multiple images.

If you want to do Twitter Cards you will need add those tags also and make a 120x120 image for Twitter.

Basically you end up with two sets of pictures: one "visible" set sized properly for your webpages, and one "hidden" set sized properly for you social media accounts.

Hope this helps...

I'll second what Michael said. If you are just interested in searches in or around Melbourne you can limit it to just AU searches. Doing that will keep the cost quite low.

If you run it about two weeks, you will also be able to pull up the full report that tells the exact phrases people searched for and the quantity of those searches. Always some interesting surprises in that report...

I think that on ecommerce sites it is good to display the social media icons only on certain pages.

Basically on the pages people are likely to share: Home, Product pages, on-site Blog pages.

You can probably leave them off things like the Category pages, Contact, Guarantee, and other instructional pages.

I'll second what Tom said. I used Link Detox's $50 for two audits plan a few times and it worked great. I even screwed up the first run by having some settings wrong, emailed their Help Desk, and they re-credited me the botched run.

If you already have collected a list of backlinks from other sources (Moz, Google or Bing Webmaster Tools, Ahref, Majestic, etc) you can also append that list to your audit. Link Detox finds most links anyways, but appending your own list makes it VERY complete.

They even have a feature where later you can append your Disavow file and the Link Detox system marks any matches so you don't have to keep reviewing those bad links. I keep hoping that MOZ and other analysis programs will add a feature where we can append our Disavow file so our link-based statistics will match what Google sees when they factor in the Disavow file.

I've bookmarked that .htaccess 301 redirect generator page. Always been afraid to mess with that file before, but maybe it's time to learn. Thanks again Peter!

Hi Peter,

They are all links from other sites. I have many versions of these types of links hanging around in my 404 report. I have not worked so far to "split" the ones that look like 2 URLs stuck together.

I will try to attach some screenshot bits from my GWT accounts so you can see what I see. Also I attached the form used by my cPanel for doing 301 redirects. Let me know if the image is not clear.

I really appreciate your help. Thanks!

baka-404-errors.jpg

Hi Lewis. If I remember right, we had the same problem and the fix was easy. Just not apparent.

I was the name of the campaign that was causing the problem. I started entering a campaign named something obvious like MyDomain but messed up too many settings and wanted to start over. So I deleted it. Then when trying to set up the campaign again Moz kept refusing and I think telling me it already existed. We fixed it by naming the campaign MyDomain-2

Let us know if this fixes the problem...

Hi Peter and everybody,

Are there some broken links that are just TOO BROKEN to fix because of the URL used?

I have some linking to my site this way: http://mysite.com/http://mysite.com/randomword  that the 301 direct form on my cPanel just will not accept. Perhaps because of the repeated http part?

Also we have some with weird symbols like this: http://mysite.com/">najboljih  and this http://mysite.com/…word/new-word.html   that the 301 redirect form will not accept. Again I guess the % and & signs are the problem?

It would be nice to get rid of these as they are creating a growing pile of 404 errors. Any suggestions?

Hi Vince,

There is Link Risk at http://linkrisk.com/

Also here is a review from back when they were in beta testing   http://www.stateofdigital.com/linkrisk/

I have not used them because when I did my link analysis only Link Detox was available.

If you give it a try, please share your experience.

Hi again Coral. There is a communication gap happening hear around a few certain words. Just fixing that will help you a lot in moving forward. Here it is...

Panda is not the same as Penguin......Panda is not the same as Penguin.....

Panda is a manual action, where you get de-indexed, get a notice, and have to do the whole terrible link removal and re-consideration request nightmare. You do not have this problem. So forget all about massive link removals and re-consideration requests.

From what you describe, you have a Penguin problem. That's just Google's programming noticing some weirdness in your link profile and either not counting the funky links or giving you a slap, a penalty, relating to those funky links (as in too high a % of a certain anchor text = a slap for your ranking for that anchor text keyword).

There are no re-consideration requests accepted for a Penguin penalty. All you can do is fix the links. And there are just a few ways to do that. Here are the ones I know of:

1. if the page the funky backlinks all point to is a sub-page, and there are few good links pointing to it, you can just abandon that original page (make it a 404) and create a new page to replace it. This rarely works because usually the funky links are pointing to the homepage and you can't abandon that.

2. get the links removed. If they come from a site you have some control or influence over, this is nice. But usually they are junk automated sites and there is almost zero chance of having the links removed

3. Disavow the domains that the funky links are coming from. Disavowing domains is usually better that individual URLs. Especially to get rid of sitewide links. You can use it to thin out your number of links using the exact anchor text you are getting slapped for. And to get rid of links from nasty toxic malware and porn sites.

Google gave us the Disavow option for this purpose. It can be used wrong, but most tools can. A good SEO group can check out your history and link profile and tell where the problem lies. I would suggest that it is not critical that the SEO company be UK based, but it is critical that they be good and that they are willing to work with small businesses affordably. It does matter that they speak english, so UK, Canada, US, New Zealand, or Australia are all fine.

Hi Rochelle. Hopefully a Moz Help Desk person answers. I have a similar problem with Google+

I can add my personal G+ profile, but I can't get it to accept my business G+ profile. There a space there for it, and I enter the page name and URL, but it never does anything with it.

I got it to at least recognize my personal G+ profile by already being signed in to G+ before opening the page in Moz Analytics where you enter your social media account.

I've used it with no bad effects, and probably good results.

I have (or had) a keyword on one of my sites with very over-optimized anchor text links. That keyword suddenly dropped from page 1 to page 7. Just that one keyword, which points to a algorithmic (Penguin) penalty. So I analyzed my link profile (not any easy job) and did a Disavow listing enough domains containing those anchor text links to re-balance my link profile somewhat. That's all I did. No reconsideration request, no removal of links. After about 6 or 8 weeks, that keyword suddenly popped back from page 7 to page 3. Page 3 still sucks, but at least that makes it easier to do the work to scramble back to page 1.

Was it because of the Disavow? I don't know for sure, but it sure doesn't look like it hurt anything.

This stuff is often a bit vague and voodoo like, which is why it helps to get a site review from a professional. Check your PM's as I sent you a reply to your reply.

I'm still a little confused. Perhaps because we are thinking of different things when we say "category page"?

I think of "category page" as a special page in the flow a customer goes thru when they start on your home page. Like this:   Home page (has links to all the categories)  >>> Category page (has links to all the products in that category) >>> Product page (has complete description and the Buy button)

So in my view, it is perfectly normal for the Category page to have a link going out to each Product page it serves. That might be 5 products or it might be 50 products.

If you feel there are too many products showing on the Category page, then you can use Sub-Categories to decrease the number of links.  Like this: Home page >>> Category page (has links to all the sub-category pages)  >>> Sub-Category page (split logically, like Men's bikes and Women's Bikes, or Under $200 and $200 to $400 and Over $400, etc)  >>> Product page

You may also want to check with your e-commerce system provider. If they don't have an arrangement that you like, there are lots of other e-commerce systems out there...

Hi Coral. I sent you a PM (which is accessed by clicking the little human shape next to the question mark and the magnifying glass at the top of these pages)

Everything I've read has said to redirect everything to either the www or non-www version. Usually in the htaccess file.

It may be time to dump GoDaddy and get a more full-featured hosting company. I used to use Yahoo for my hosting and they also did not allow things like access to the htacess file and 301 redirects. Once I went to a better hosting company I was so much better off.

Now I use GreenGeeks.com for all my hosting. They have a full cpanel and all the good features. They will do your site migration for free. If you don't know how to do the techy Server stuff, they have great free phone support that will actually do things like alter the htaccess while you are speaking to them (which they have done a few times now for me).

Hi Ron.  If the keywords and phrases for your niche are REALLY LOW volume (like some of mine are), then I suggest using Google Adwords by running a few short PPC campaigns.

A two week run with your keywords set as a Broad match will gather enough data to activate an interesting Adwords report. This report is found on the "Keywords" screen of an individual campaign. Click the "Details" button then select "Search Terms / All" on the dropdown. This will list exactly what phrase people typed in the search box, how many times it was searched (Impressions), how many Clicks those phrases got, etc.

Just peeked at one of my campaigns and for the last 30 days it only had 2400 impressions. That's pretty low volume and those small volumes do not show up well in any free tool that I know of. Though they used to in the old Adwords Keyword Tool. I miss it...

Your welcome. Here's a little insight on the two shopping cart systems I mentioned. Note that both systems have added (and are promoting) the standard e-commerce hoopla of auto-generated product pages and the sort of stuff systems like BigCommerce do. But they do both also have the simpler option of just using the cart and checkout system and allowing you to put your Buy buttons anywhere you can reach.

I started trying to set up AmeriCommerce because it had the most features. But as I got in deeper, the complexity of the features and the poorly written technical documentation overwhelmed me and I gave up. If you have some brainy tech guys they can probably understand the documentation better than I could.

I then switched to UltraCart and have found it much easier to use. You can see the standard cart appearance on some example pages on their website. My cart at EasyDigging.com is pretty heavily customized. They will soon have a easily modified HTML/CSS based cart available. There documentation and tech support is pretty good.

Sam wrote: "Each site is unique in their own way and are already ranking well on their own."

Another possibility is to leave the sites where they are (since they already rank well) but let them all share one single Shopping Cart / Checkout system. I know of a few e-commerce system that allow this. UltraCart and AmeriCommerce are two. You can put the "Buy Now" and "View Cart" buttons on any site and some social media sites (you don't even have to own the sites, which opens interesting possibilities for guest blogging with a shopping capability)

Then you could just re-style the three sites appearance so that they blend well together yet stay distinct. Like three old neighboring houses that were all built in the same era. And freely offer navigation between all three so it pretty much acts like one big website. Maybe put a logo in the header of page saying something like "Part of the Gizmo Group of fine stores". Many people would barely notice that they are moving from one domain to another.

The upside would be safety. If one of the three sites gets a penalty, the other two can continue to pull in customers while you repair the penalized site.

Just tossing around ideas. I've considered doing this with some unused domains we own. Hope it helps you...

Hi David,

As far as I can tell, a campaign does not have to have GA connected to it. But it behaves differently when it does. Also weird things happen when old Moz Pro campaigns are viewed in the Moz Analytics dashboard. Here's what I am seeing in my MOZ account...

1. Moz Analytics and Moz Pro behave very differently.

2. a week ago I created a quick campaign in Moz Analytics just to test something, I did not associate it with any Google Analytics account. It has data in it now, but not as much as other campaigns that are associated with a GA account.

3. also a week ago I re-activated some old Moz Pro campaigns that do not have GA associated with them. They now appear in the Moz analytics dashboard - but with the spinning icon and the "We are collecting your data now. Check back within 24 hours" mesage.

4. to view those old Moz Pro campaigns in functioning manner, I have to skip the Moz Analytics dashboard and view them through the old Moz Pro dashboard which is here  http://pro.moz.com/campaigns/

Hope this helps you.

Hi BeytzNet.  It would help to know what scale of e-commerce you are curious about?

I think the online retail industry needs some new terms. The brick-n-mortar retail industry has a nice collection of easily understood terms. We all know the difference between a big box store, a mom-n-pop shop, a franchise, a boutique, a chain store, a convenience store, and a resale shop.

But online retail seems to just have "e-commerce sites". An online store with 50,000 products is a whole different beast than an online store with 50 products (or at least it should be) - and they should have different names. Maybe Mega-e-commerce for the 50,000 item store, and Micro-e-commerce for the 50 item store. Then we could fill the gaps with Mini-e-commerce for the 500 item stores and Middle-e-commerce for the 5000 item guys.

And boy is it hard to find good e-commerce systems for the 50 item stores. Something like Magneto or BigCommerce is overkill. But systems that are friendly to Micro and Mini-e-commerce are out there if you dig deep enough.

Hi Peter. Maybe I'm reading your description wrong, but it sounds like a pretty standard ecommerce arrangement.

Is the following true?

Search Volume (ranked highest to lowest):

1. Generic Products (mountainbike, commuter bike, racing bike, etc)
2. Product Brands (Schwinn mountainbike, Trek mountainbike, Hurley mountainbike, etc)
3. Brand Models (Scwhinn 851, Schwinn 3400, Schwinn 4xtc, etc)

Page linking plan in question:  ( >>> means links going to a collection of pages)

Generic Product page >>> Product Brand pages

Product Brand page >>> Brand Model pages

Although this is a pretty typical ecommerce arrangement, I personally think it is not the best plan for sites only selling a few products. When there are only a few products I think it is better to go straight from the Home page to the specific product pages.

I too think that C is best, but instead of listing the inventory in ALL the stores perhaps you can have them select their local store and then just show the inventory level at that ONE store for the rest of their visit to your store (or until they select a new local store)

Showing the inventory at all stores may be a bit distracting on the page (bad for CRO) and may lead to lots of customer phone calls asking if you can have something from store A sent over to store B for them (which could be a pain in the butt)

When I first read your question, the first thing I thought of was just recycling the URL's each sports season...

For instance, each year the Bears play the Packers, so in 2013 you write up your prediction on the page: mypredictionsite.com/bears-vs-packers.html and the page hangs around until 2014 when you re-write and re-publish the page for that year's game. Be sure to use schema and other tags to assign a recent Date for updating the page, and put something big and bold at the top of the page so people know what season the prediction is for. (Maybe also a tally of how well you did predicting their previous matches.)

That way any links it picks up over the years are pertinent to the Bears playing the Packers, and that could help with ranking. Also you don't have to keep track of perpetually growing collection of 301's.

Just a thought...

Hi Bruce,

I have two small ecommerce sites, so your question is very interesting.

I haven't heard before the phrases you used: "short head and chunky middle keywords". I guess keywords are described differently in NZ

We started out adding content by adding "article" or "guide" or "how to" pages to the site. These pages do not contain any Buy buttons, but do contain image and text links to our store pages. You can see an example here: http://www.bestdryingrack.com/How_to_dry_flowers.html

These article pages (on both sites) tend to be some of our most active landing pages, though they don't feed people into the store pages at a high rate. But even at a low rate, they are a super cheap way of letting potential customers know about our stores.

But we have since switched to adding content in a blog format. The blog is of course hosted on the site in a /blog/ folder so it helps are total link count. We made this switch because it made it much simpler to write about a broader range of topics. Before we spent weeks planning and writing a 3 page "guide" and connecting it into the site. Now we can do a single page blog post in a day anytime we come across something interesting or amusing. And it doesn't even have to closely relate to the products on the ecommerce site. Here is an example: http://www.easydigging.com/blog/apple-peeler.html

We have not been running the blog long enough to know at what rate the pages will feed people into the store, but I assume it will be similar to the article pages.My hope is that we can post weekly and therefore create more blog pages than we ever used to create article pages. Doing so would feed more people into the store and also keep increasing our link count.

I had a Moz account earlier this year, let it drop for a few months, then decided it really was important to have one. So I'm back.

But there is still something about it that I really find frustrating. I can't seem to find any answers to what seem to be basic questions. Maybe they are too simple, maybe they are just dumb questions, but I sure would appreciate it if somebody could lead me to another source of information. Maybe somebody has written a "Moz for Dummies" book?

Here's an example:

I'm setting up campaigns for my websites, and I get to the Brand and Mentions section. It tries to autofill the space using my campaign name, but that's not accurate. So I go look at the Help Hub for Brands and Mentions, and it gives an example using MOZ as the brand name. But not many sites have a three-letter domain name, do they?. One of my domains is EasyDigging.com (2 words) and the other is BestDryingRack.com (3 words). So I look and look, but can't find any examples of longer domain names.

So I have no idea how to enter my Brand. Should it be the whole "EasyDigging.com" or should it just be "EasyDIgging" or should it be broke up into "Easy Digging"?

This is just 1 example. I find these sort of unanswered questions almost every time I try something new.

Is there a collection of examples anywhere?

I really hope so. Surely there are others who learn best from examples, who hate having to guess based on slim or vague instructions. Keeping my fingers crossed that somebody can lead me to the goldmine of good examples. Thanks!

All of our Toxic links seemed to have good reason for being labeled such, and I did Disavow those domains. There is LOTS more info in the downloaded CSV files than ever shows on the Link Detox screen.

Some of the High Risk ones were kept. Usually it was the thing about them coming from a de-indexed site or blog. When I looked at the site or blog it appeared to be a nice site that just happened to get hammered by Panda or Penguin. I figured that the owner of the site was working hard on getting the penalty lifted and eventually it would be a good site again.

I also disavowed a few marginal domains to help re-balance my anchor text profile. Which was out of balance enough to get Penguined for the over-optimized keywords.

Successful? We will not know until Google runs a new Penguin update. Hopefully they do it real soon...

Answer to First question: I don't think anybody knows yet as Disavow just hasn't been around long enough yet. But what I read is that it causes Follow links from the "bad" domains to list to be considered NoFollow links and therefore harmless.

Answer to Second question: Yes, yes, and no.  I wrote about this in another posting yesterday here: http://moz.com/community/q/google-disavow-file-update

Good luck!

Yes. I did keep in most of the old domains that had been in the original Disavow file.

Since the things I learned after doing the original file showed that I had mistakenly disavowed a few domains that were OK, I did omit them from the new file. And I added some that had not been in the original.

Nobody but Google knows for sure, but my understanding is that a particular Disavow only has an effect while that particular file is active on your Google Webmaster account. So if you delete it or upload a new one, then you will eventually see the old effect disappear and be replaced by the effect of having no or a different disavow file.

I just wish the effect wasn't so slow on Google. The effect of disavowing linking domains in Bing Webmaster Tools seems to be very quick. But it's rarely needed there since Bing doesn't have nasty Pandas and Penguins to battle...

Hi Daniel,

I also submitted a Disavow file, then learned new things and updated the file to reflect the new knowledge.

You can only have ONE disavow file. So you simply create a new file (which may or may not contain many domains that were in the original) and then go to Google Webmaster Tools and delete the original file and upload the new one.

This article has good info on how you should create your Disavow file: http://moz.com/blog/google-disavow-tool

Especially this part from the article:

Google rejects many disavow files because of bad formatting, but webmasters usually never know. Guidelines state the file type should be .txt only and “must be encoded UTF-8 or 7-bit ASCII.”

Good luck!

Just wanted to send some love to all those that go beyond page 1 

As the owner of a small ecommerce business who has been pushed off the first page by all the big box store (Sears, Walmart, Costco, etc) results, I'm always overjoyed to hear about those who have made it a habit to go beyond page 1.

Maybe there should be some sort of Go Beyond viral campaign like "Google Beyond page 1 for the good stuff!"  (pretty awkward, needs some work, where are all the bright copywriters?...)

Hello speedingorange,

You wrote:  "I have always been under the impression from reading online that an algorithm update would see a site destroyed for most terms and a notification generated in webmaster tools, however the site still seems to still rank for some terms, others however it has fallen off the face of the earth for. "

I think you have it backwards. The Panda update, which is NOT algorithmic, is the one that would that would "see a site destroyed for most terms and a notification generated in webmaster tools".  The Penguin update, which IS algorithmic, is the one that would cause "site still seems to still rank for some terms, others however it has fallen off the face of the earth for".  And a Penguin penalty does not result in a Google Webmaster Tools notification.  (That is my understanding. I'm sure someone will correct me if I am mixed up, And I am OK with being corrected.)

Two of my sites are suffering from what I believe is Penguin damage. For both the damage is a lot of rank lost for only a few keyword phrases. Other phrases continue to rank normally. When I did a bunch of keyword research (using Moz, Ahrefs, and Majestic results blended together) I found that the few keyword phrases that I severely lost rank for had both:

1. a high percentage of Follow links using exactly those phrases as anchor text, and
2. links from sources that didn't really pertain to the anchor text or to our website.

My understanding is that these are the type of things that Penguin will bite you for. So we did remove or change a few of the anchor text links, and did a Google Disavow on most of the remaining links. We did leave a few anchor text links that were from good pertinent sites. This was done in early June, and we still have not seen any results. I have read that we will not see any big results from submitting a Disavow file until Google updates Penguin again. So we are keeping our fingers crossed until then.

Couple of quick things to explore if you are going to try a similar rescue. First, we did not pay any attention to our NoFollow links since they supposedly do not effect rankings. Second, we did create a nice file of follow links from all our link research, and used it as an additional submission with the Link Detox tool (the spreadsheet report from that tool is very valuable when checking for overoptimized anchor text). Third, we disavowed entire domains rather than individual links because we heard that was most effective.

Good luck!

Hi Jen. I'm curious to hear what answers you receive. Especially since they are NOFOLLOW links...

My understanding is that NoFollow links do not effect ranking at all - whether they are good or bad or even all have the same anchor text. That is supposedly a featuire of Google's Disavow tool - it let's you turn bad links into NoFollow links and make them "disappear".

I just spent many days working on a Disavow project, and the first thing we did to our spreadsheet of links was sort out and delete all the NoFollows. What a mess if we misunderstood the effects of NoFollows!

No research. Just memory of doing searches with and without an S for my own keywords and noticing that the results were fairly similar.

I just checked garden and gardens - many of the page 1 results are the same.

Then I checked tool and tools - very different results because of the band "Tool"

Checking garden tool and garden tools takes it back to many similar page 1 results.

The original poster just asked for a Rule of Thumb. So perhaps the answer is "It depends on the keyword. Google it and see what happens."

When the plural is made by just adding an S, then Google seems not to differentiate the singular or plural. You can verify it by opening two windows and searching for the term both with and without the S and seeing if the results are ranked differently.

But if the plural is a whole different word, like Goose and Geese or Mouse and Mice, then you will definitely have to makes a decision on which to use.

Couple of things...

1. You service is focused on Grand Rapids, but no where on your home page do the words "Grand Rapids" appear. You need to use the phrases and the locations you want to rank for. Get something like "Home Care in Grand Rapids" as an h1 way up near the top of the page. Then sprinkle appropriate words like "Grand Rapids, Michigan" at least 4 times through the rest of the home page.

2. If your site is brand new, it may not rank organically on Google for quite awhile. Last I heard, Google had a 1 year "sandbox" that most new sites were stuck in. The sandbox is there to make sure you are a serious site that is going to stick around. Supposedly it is VERY HARD to rank organically above page three while you are in the sandbox. During that period you can use PPC and social media to drive visitors to your site.

Thanks you Chris and Gagan!

Your responses were fast and valuable

I have two e-commerce stores (one using Yahoo Merchant Solutions and one using UltraCart) but neither of them are set up in the generic Categories/SubCategories/ItemPages format. So take what I say just as different opinion...

First the techy answer: What I have read is that it would be good practice to either:

1. pick your best selling "Armani Jeans Polo Shirt" color and set the rel=canonical tag on all the other color pages to point that one best selling color

or

2. use the  "Armani Jeans Polo Shirt" subcategory page (the one all the colors variations are linked from) and set the rel=canonical tag on all the other color pages to point that starting subcategory page

Now the different opinion:

You wrote "We do not do it with colours (the dropdown) because the cateloge of clothing can become quite thin when viewing. That and becuase the majority if not all major retailers do not do it this way."

I think the reverse is true... Having separate pages for each color dilutes your SEO effort by making most of your pages "thin" with no really unique content. Your blue page will not be any different from your green page (except for different pictures and the words blue and green interchanged). Also if you are a newcomer to selling these clothing items, perhaps you should do it differently than "all major retailers do"? You can't really compete with them on their own terms - they are too big and too well established. So perhaps using a different store format will help, and putting all the color and size options on a single strong unique page may be the way to do that. It may open some unique marketing or presentation possibilities.

Good luck!

Setting up canonical tags for an old site. I really need advice on that darn backslash / at the end of the homepage URL.

We have incoming links to the homepage as http://www.mysite.com (without the backslash), and as http://www.mysite.com/ (with the backslash), and as http://www.mysite.com/index.html

I know that there should be 301 redirects to just one version, but I need to know more about the canonical tags...

Which should the canonical tag be???

(without the backslash)

or

(with the backslash)

Thanks for your help!

I believe that your choices will be restricted by the CMS that your particular e-commerce system uses. Does your e-commerce system provider have a user's forum where you can ask long-time users for recommendations?

I see you are not doing separate pages for sizes. I assume that you are handling that with a drop-down selector or something similar. How about doing that for the colors as well?

Otherwise make good use of the rel=canonical tag to avoid having lots of essentially duplicate content...