Welcome to the Q&A Forum

Howdy,

I'm planning out a website and need to plan out the URL structure for best SEO value. Generally I would do something like this:
site.com/widgetssite.com/widgets/large
site.com/widgets/large/blue
etc.

I think this is a pretty straight forward SEO tactic. The issue I have with it is in terms of natural language the "thing" you are searching for in this case is a widget, so typically you would type/search [adjective] [noun], or in this case "large blue widgets."

So one proposal I have is to instead append the "widget" to the end of the URL:
site.com/large-widgets
site.com/large/blue-widgets
site.com/large/blue/square-widgets
etc.

Obviously this breaks the whole silo concept since the square-widgets page is inside the /blue directory but the blue widgets page isn't at /blue it is /blue-widgets. My solution is to setup 301 redirects from /blue to /blue-widgets (even thought there are no site links pointing to that page).

Does this seem like a good idea? Or does this break the whole folder silo concept? What I like about it is that it feels more user friendly in terms of natural language and for certain high value keywords we can get certain pairings of words into the URL more like how a person would type them in.

I'm not sure what the correct terminology would be for this but I'm calling it an ad wall. Essentially an ad overlay when someone enters a website. I see this most commonly on certain news websites. For example when you click on a link to an article on ign or forbes.com you get an ad that you have to close or skip to read the article. What are the SEO considerations if implementing something like this?

I'm wondering if there are any similar to a pay wall in the sense that you want to let crawlers in to see your content and rank it but users get an ad or redirected to an ad and then back to the article page. This link currently does it for me for example

http://www.forbes.com/sites/tjmccue/2012/05/22/spacex-launches-with-15-dreams-onboard/

I set my user agent to google bot and go right through to the article but if it is set to the browser default I get to an ad page I have to skip first. Is this the infamous "white hat cloaking"? Are the other ways to implement the same idea (a modal window that opens via javascript for example) that are more or less risky? I'm mainly interested in doing this based on referrer: people who type a URL directly don't see it but clicking on a link they do see it, for example.

Yes actually you are correct. After I read this answer I tested it on my personal site by adding the tags around some nonsense words. Not only did Google index the pages with the nonsense words making it into their cache of the pages, but my site ranks for those nonsense words. So while it would be awesome if Googlebot honored those tags, they only work for the Google search appliance!

I have a page that currently ranks high for its term. That page is going away for the main website users, meaning all internal site links pointing to that page are going away and point to a new page. Normally you would just do a 301 redirect to the new URL however the old URL will still need to remain as a landing page since we send paid media traffic to that URL. My question is what is the best way to deal with that?

One thought was set up a canonical tag, however my understanding is that the pages need to be identical or very close to the same and the landing page will be light on content and different from the new main page. Not topically different but not identical copy or design, etc.

How do people deal with low value pages on sites which tend to pool pagerank and internal links? For example log in pages, copyright, privacy notice pages, etc.

I know recently Matt Cutts did a video saying don't worry about them, and in the past we all know various strategies like nofollow, etc. were effective but no more. Are there any other tactics or techniques with dealing with these pages and leveraging them for SEO benefit? Maybe having internal links on these pages to strategically pass off some of the link juice?

Thanks for the info!

I'm curious, what does Google mean here regarding internal links avoiding repetition? Wouldn't it make sense that your internal links for a particular page on your site all use the same anchor text for the sake of consistency? Or am I not understanding their advice correctly?

Are Google site links only ever shown on the top website? Or is it possible for certain queries for the site in position #2 or #3 or something to have site links but the #1 position not have them? If there are any guides, tips or write ups regarding site links and their behavior and optimization please share!

Thanks.

Thanks!

Is it possible to download the data in my Moz analytics for use in Excel? I'm specifically after the data behind the graphs in the Links section. I don't see a way to download this...

Also, the more I think about it, I probably had a list of keywords in excel where the column header was 'Keyword' and selected the entire column to copy/paste which is how this keyword got into the campaign. Glad to know it's a known issue!

Thanks. Just to be absolutely clear, cause it's kind of confusing, here is a screenshot of what I'm talking about.

721GA7diNk

Have you tried looking in any of the other link graph tools like ahrefs.com or majestic SEO?

Hi,

In my campaign settings, under manage keywords, I have the keyword "Keyword" listed. I can't seem to delete it. I'm not sure how it got there. Is that there for a reason? When I delete it it just comes right back.

Yes that makes perfect sense. I had no idea that Google would use its own discretion on whether to honor a canonical tag or not! Totally agree that noindex is the way to go then. Thanks!

On a website I'm working on which has a search feature with paginated results, all of the pages of the search results are set with a canonical tag back to the first page of the search results, however Google is indexing certain random pages within the result set. I can literally do a search in Google and find a deep page in the results, click on it and view source on that page and see that it has a canonical tag leading back to the first page of the set.

Has anyone experienced this before? Why would Google not honor a canonical tag if it is set correctly? I've seen several SEO techniques for dealing with pagination, is there another solution that you all recommend?

I do have some experience in this area. If you operate a highly measurable marketing program, some web marketing agencies will agree to a "pay per performance" model of compensation, but you will have to work with them for it to be clearly defined, and they will still want a flat rate compensation for their hours spent. At the end of the day, agencies want to get paid period. And they should be. You may end up paying more for their services going this route, so if saving money is your concern I wouldn't recommend it. If ensuring that your agency can deliver and that they have some "skin in the game" to keep them honest, then this could be a great direction.

A typical setup I've seen is the agency will give you their hours at "cost" or a very low rate as a baseline to cover their expenses and time, then if you have very good past historical performance reporting setup, and they are comfortable that they can do what they say they can, you can define a payout based on "results" such as website conversions from organic search sources. So comparing year-over-year, say you got 100 conversions in October 2012 from organic search, you could say for every conversion we get in October 2013 above 100 you get 25% of the revenue, or something like that.

Also keep in mind, the industry is somewhat in free fall right now in my opinion due to the increase of "not provided" keyword data. In the past, you would do a contract like I outlined above specifying that you would not count branded keywords. The last thing you want is to run a magazine ad which increases searches for your brand 2000% and have to pay the agency for the influx of organic search conversions that you would have gotten anyway! With all the organic search data lumped into one bucket now, I don't see how that will work anymore personally.

(sorry if my response is newbie-ish as well. Not sure what you've tried already but these are basic.)

1. I assume you've setup Google Webmster Tools for both the sub domain and the original domain? Anything in there like trying to fetch both sites as googlebot, blocked URLs, any strange messages or errors? Fetching and resubmitting the entire new site to be indexed? Basically GWMT is your friend in this situation.

2. If you've now blocked the subdomain with robots.txt I think that may delay resolving the situation. It may not be ideal but setting up 301 redirects from the subdomain back to the original domain would be the best course of action and moving the sub domain (I assume for development or staging purposes) to a new subdomain with the proper robots.txt file would be idea.

Actually I see a very similar breakdown of Google as a referrer and direct traffic for safari as a segment in my Google Analytics as I see for my total traffic so I do believe GA is reporting the referrer correctly for Safari. I have no idea if Safari specifically has more of an issue with "not provided" but I do know that Chrome has been particularly bad all year since it started treating any search made in the address bar as a secure search by default. Maybe Safari does too, I don't know. But the traffic is at least attributed back to Google organic as a referrer, not direct, in my analytics.

Just curious what analytics solution you use? I've been a long time Omniture user and just recently started using GA (not by choice). I was definitely seeing issues with the way Adobe was classifying and under reporting not provided searches from Chrome due to this issue!

lJOkDWdi

I've been curious what other SEO professionals have been seeing in their own analytics since the industry news came out that Google has gone to "100% not provided." I've definitely seen a steady rise in 'not provided' data across the board since about July but no sudden spikes and still not 100%.

Looking at one specific website this morning for the month of October to date in GA, and segmenting out organic traffic from just Google, I'm seeing about 85% not provided. Looking at just today (Monday) it is 88%. So over time it is definitely trending up, and I expect it will be greater than 90% by the end of October. Just curious if that is consistent with what others are seeing? Has anyone truly gotten to 100% of Google Organic traffic at 100%?

John,

You just literally blew my mind with that googleon/googleoff documentation! I've been working as an SEO since 2001 and have literally never heard of this! I have so many questions I need to research. I can think of a lot of ways to use this but I'm sure the best practices around its use are more nuanced than just the technical documentation.

Anyway, in terms of my immediate problem, not sure if that will fix it. I should have mentioned that in addition to the message about Javascript, the noscript tag also contains site content, including navigation links, that are not on the page otherwise for non-javascript clients. In other words, this entire website is a singe blank page with no content on it if you do not have javascript without the noscript tags. The long term solution is to completely redo the website, obviously, but I need a short term solution to get some SEO traction. I guess I could always put the javascript message as an image.