@icarusvn: Using nofollow to sculpt link juice is ineffective. Google killed it a few years back. If you nofollow a link, the link juice is still deducted from the total leaving your page. It just doesn't help the page you are linking to. You are basically just dumping link juice on the floor.
@dayker: I wouldn't worry much about URLs. Google doesn't put too much weight on them any more. Matt Cutts did a video blog and basically told e-commerce sites not to bother reworking their dynamic URLs because it wouldn't make much difference.
BTW, if you have a large site, and you aren't siloing it, you are a FOOL!!! It's simple logic. Google weighs internal links the same way they weigh external ones. If a site about football gets a link from a site about lipstick, the link is going to be devalued. Same-same on your internal links. If you have a page about lipstick and link it to a page about football, the link is devalued. If that link could pass 100 theoretical points of link juice, only a fraction of those "points" gets used by the target page because it isn't relevant. The rest gets dumped on the floor. If you have a large website, and are interlinking irrelevant pages, you are hemorrhaging link juice.
If you're Amazon - they don't give a shit because they have a huge pool of link juice. If you have a PR3 home page, how you distribute link juice matters - a lot.
For my day job, I was just hired to run SEO for a $400 million ecommerce company. My second order of business is siloing the company's 7 websites that sell over 12,000 products. First order of business was keyword research, which I just finished. What a pain in the balls that was.
The company was targeting bad (too competitive or low volume) KWs, and the sites, which are PR5s can't get ranked for anything because of their linking structure. If you have 400+ links on your home page (which they do), you are spraying link juice everywhere, and nothing gets ranked.
I have a team of developers, and we are pulling out every white hat trick in the book. These sites are going to be themed tighter than a gnat's arse.
@Satans_Apprentice may I ask your opinion on the different methods of building a silo?
Would I be right in my assumption that the type of silo depends entirely on the page or keywords that you wish to rank.
Daisy chaining of topic/category articles was mentioned earlier in this post and will serve the purpose of ranking several pages/keywords, perhaps if the competition is poor in the niche you are targeting.
Wheras the linking to a parent page from each child 'powers' or attempts to push that page higher.
So it is entirely down to the flow of link juice and this can be manipulated as seen fit. This being true, and user friendliness aside, sites can be constructed/modified to 'power' or push single money pages higher.
This is my understanding currently and I will further modify a test site in line with your advice to control the link juice flow 'tighter than a ngats ass'.
I imagine modifying a large e-commerce site would be a real pain. Particularly one in which you need corporate approval for UX changes. Best of luck.
@icarusvn: Perfect silo structure is pretty simple. Pages only link up and down. At the bottom of the silo, all children link to each other and up to the parent. In practice, it's not that easy. The pervasive top nav bleeds a lot of authority. Fortunately, I know about 4 ways to hide navigation from Google, and I have a team of developers chomping at the bit to do it. The plan is to block our top nav to Google, and build an alternate link architecture for distributing link juice. We have the ability to sculpt link juice with incredible precision. If I have a product or category with low search volume or there's too much competition, I can shut off link juice entirely, but still have users be able to navigate to it. We will have a navigation system for bots and a separate nav for humans.
By my calculations, we are going to be running at about 90% link juice efficiency.
It works, and it's 100% white hat. I won't talk about how I do it. It's my secret sauce. BTW, we also block the footer, which also bleeds ranking power.
It keeps a lot of people happy. I don't have to cut back on the links in the top nav, which is always a huge fight with marketing. They can put 1000 links up there as far as I am concerned.
@Seljo@easypeasy: you can silo with WP, but you can't hide the top nav or the footer, and they bleed authority. It's definitely worth Siloing in WP, but it's not nearly as efficient as building from scratch.
the genesis extender plugin for the genesis framework allows very simple removal of the navbar on a per page basis but this is not exactly user friendly except for landing pages... and the cogs turn...
i think you could do it with wordpress too, i think I've done plugins doing this in the past, but can't remember exactly anymore. but wordpress is just not needed. normally you would want to generate the whole site. you don't want to do this manually, and you also don't want to pay for all that content because that's hundreds of articles (at least).
@everyone .. Some interesting info here! .. You have twisted my arm and I'm going to give this silo a try on a new site. Looking for an opinion on this plugin, is it going to set up the silos correctly?
I recently read an interesting article about how on-page SEO is becoming the new off-page SEO and I though it was really interesting. It was written by one of the more known SEOs around today, you should check it out
It looks for me more and more that wordpress and such CMS becoming hard to manage...
About off page stuff what keywords is better to use for anchors to link your man page ?
I think any mix is good, even if you do not have article in page about that. I mean:
If you have page about web hosting then you can easy add any anchor what is web hosting related, even if you do not have article optimized for that keyword.
Example web hosting is very good i think is good anchor to link main page.
Another point is what I'm not sure is if I have website with 30 articles but all of them are keywords I'm using is very close to each other.
Example: kids clothes is main term and all other 30 terms is low searched terms like with 200 searches per month. Do I still need to split them in SILO ? If yes then how can i split them if they are very close to each other.
I can split like kids girl clothes and kids boys clothes, but then i have some articles what is related to both boys and girls. Then what Should i split in 3 categories. Boys - Girls - Together ?
@botman - You can rank one article for boys, girls, kids. But it would be much easier to group keywords in those categories and keep them separate.
For example, category: kids clothes
Keywords: cheap kids clothes discount kids clothes designer kids clothes kids clothes for school kids clothes online etc
You can rank for all of those in one article and sure maybe some are only 200 searches per month, but a) That's what the keyword tool says, that's not a factual number and b) Lots of small searches per month are easy to rank for and combined they offer some decent volume.
After looking at this closely over the last few weeks. I think the Keyword tool gives a good indication as to how you should silo. It groups searches with a similar intent.
So if your seed keyword is "kids clothes" and the keyword tool gives you similar searches for both girls and boys, you can create two subsilos below. Kids clothes Boys clothes Cheap boys clothes Boys school uniforms X5 Girls clothes Cheap girls clothes Girls school uniforms X5
Anyway just want to make the point that google's keyword tool indicates good related topics for siloing and shows the searcher's intent.
My understanding is yes, as it is the domain that is important and not the individual installation of WordPress. However my experience is limited. And if you look at the material here and referenced, it is the links that pass the juice. So were you to manually link your blog posts to your static pages(on topic and relevant) then they would pass juice. I am no expert though, but this is something I am very interested in and currently testing. There was a set of videos mentioned and I have seen the recommended elsewhere also.
Comments
Would I be right in my assumption that the type of silo depends entirely on the page or keywords that you wish to rank.
Daisy chaining of topic/category articles was mentioned earlier in this post and will serve the purpose of ranking several pages/keywords, perhaps if the competition is poor in the niche you are targeting.
Wheras the linking to a parent page from each child 'powers' or attempts to push that page higher.
So it is entirely down to the flow of link juice and this can be manipulated as seen fit. This being true, and user friendliness aside, sites can be constructed/modified to 'power' or push single money pages higher.
This is my understanding currently and I will further modify a test site in line with your advice to control the link juice flow 'tighter than a ngats ass'.
I imagine modifying a large e-commerce site would be a real pain. Particularly one in which you need corporate approval for UX changes. Best of luck.
By my calculations, we are going to be running at about 90% link juice efficiency.
It keeps a lot of people happy. I don't have to cut back on the links in the top nav, which is always a huge fight with marketing. They can put 1000 links up there as far as I am concerned.
For example, category: kids clothes
Keywords:
cheap kids clothes
discount kids clothes
designer kids clothes
kids clothes for school
kids clothes online
etc
You can rank for all of those in one article and sure maybe some are only 200 searches per month, but
a) That's what the keyword tool says, that's not a factual number and
b) Lots of small searches per month are easy to rank for and combined they offer some decent volume.
So if your seed keyword is "kids clothes" and the keyword tool gives you similar searches for both girls and boys, you can create two subsilos below.
Kids clothes
Boys clothes
Cheap boys clothes
Boys school uniforms
X5
Girls clothes
Cheap girls clothes
Girls school uniforms
X5
Anyway just want to make the point that google's keyword tool indicates good related topics for siloing and shows the searcher's intent.
I am no expert though, but this is something I am very interested in and currently testing. There was a set of videos mentioned and I have seen the recommended elsewhere also.