Fine tuning on-site seo

[Tucky]

Hey folks

So I just ran a site of mine through ahrefs.com and it comes up with 78% on their Domain Health which I would like to improve on. I'm not sure how much these affect the rankings but from a personal improvement angle I'd like to increase the score anyway.

So as a background, the site is a fairly basic wp site using the Salient theme. It has the Yoast seo plugin running so I may be able to do some additional tweaks there, but here are some of the points the ahrefs dashboard suggests fixing...

Duplicate title and descriptions on /blog and /blog/page/2 - obviously this is where the posts go onto the next page so how can this be fixed?

Missing H1 tags on those pages too. I don't think you usually have an H1 tag at the top of your blog page do you?

Missing meta descriptions on post category pages

Any ideas how to fix these please?

 

[animalstyle]

Duplicate title and descriptions on /blog and /blog/page/2 - obviously this is where the posts go onto the next page so how can this be fixed?

This is the only one I am not sure about 100%, maybe @Ryuzaki can check my work here:

Yoast should be specifying the same canonical URL for each page beyond /blog/. That means no matter how many /blog/page/X's you go, the canonical URL always points back to /blog/. Yes there are going to be duplicate descriptions, but if you're telling the search engines that it really starts at /blog/ then you should be OK.

Missing H1 tags on those pages too. I don't think you usually have an H1 tag at the top of your blog page do you?

If you're using header tags properly, you should. The H1 tag declares the title of the page and carries the most weight of the header tags. What html tag is surrounding your page titles/article titles, or do you even use titles for these pages?

Missing meta descriptions on post category pages

Visit these pages and click the edit button while logged in to WP. You can go and specify these meta descriptions in Yoast.

 

[Tucky]

Thanks

Ok so the main blog page just shows 10 post previews and the titles are showing as H2's. This makes sense as 10 H1's would be suicide. There is no other content above the posts apart from the hearder but I'm thinking maybe ahrefs is just basing it on a standard page, so it doesn't really matter if you see what I mean?

I have found the category pages. There is a normal content area where you can fill info in (including images!) but perhaps the one to fill in is the Yoast snippet editor where it has space for a description? Or maybe I should just make the categories nofollow?

 

[animalstyle]

Ok so the main blog page just shows 10 post previews and the titles are showing as H2's. This makes sense as 10 H1's would be suicide. There is no other content above the posts apart from the hearder but I'm thinking maybe ahrefs is just basing it on a standard page, so it doesn't really matter if you see what I mean?

In this case I like to put 'Blog' as an H1 at the header. Then if I have categories that same H1 becomes 'Category Name' or 'Blog - Category Name'. Its all about usability, if someone landed on your blog would that title help them?

I have found the category pages. There is a normal content area where you can fill info in (including images!) but perhaps the one to fill in is the Yoast snippet editor where it has space for a description? Or maybe I should just make the categories nofollow?

That is what I was referring to, the Yoast snippet editor, yes.

When you say you 'found' it I wonder - are these things that you want people to see? If not - noindex makes sense. If you want people to see them, then leave as index.

 

[Ryuzaki]

Duplicate title and descriptions on /blog and /blog/page/2 - obviously this is where the posts go onto the next page so how can this be fixed?

Using Yoast, what I like to do is simply set anything past /blog to no-index. Note I'm not saying no-follow. Just no-index. There goes all of your duplicate content woes and the possibility of ending up with 100's pages with duplicate snippets from across your site. Panda-proof. Having those in the index does nothing for you anyways.

I like to use the category description in Wordpress as a piece of static content that never rolls off to page 2 like the snippets do. You can add a paragraph, links, pictures, whatever. I'll also set up if statements to show something different on page 2 to infinity that has a link back to page one, maybe just a sentence here. This static content helps if you intend on trying to rank these category pages.

You can also use this same description as your meta description. That may even be default in Wordpress. I can't remember at the moment. Like @animalstyle said, you can edit the category and find the Yoast options too.

Also, I'd just my category title as the H1 and leave the post titles as H2. In your case, you just need to find the proper template and add that in there. I think this is it off the top of my head:

<h1><?php single_cat_title(''); ?></h1>

And that should solve all of the problems. That's how I do it anyways.

 

[c4yrslf12]

Using Yoast, what I like to do is simply set anything past /blog to no-index. Note I'm not saying no-follow. Just no-index. There goes all of your duplicate content woes and the possibility of ending up with 100's pages with duplicate snippets from across your site. Panda-proof. Having those in the index does nothing for you anyways.

Couple questions..

1. Does it really hurt to have those /blog/page/2 and /blog/page/3 pages? I don't see how it's creating a Panda issue... maybe you could explain a little further.
2. Assuming no use of Yoast, how might one 'fix' the issue?

Thanks!

 

[Tucky]

Thanks for that guys. I have decided to no-index / dofollow categories and tags (seem to remember reading it's not a good idea to have tags indexed either). I've also added an H1 title to the blog main page and a subtitle. This was available from within the theme.

 

[Ryuzaki]

1. Does it really hurt to have those /blog/page/2 and /blog/page/3 pages? I don't see how it's creating a Panda issue... maybe you could explain a little further.
2. Assuming no use of Yoast, how might one 'fix' the issue?

On a large enough site set up like a typical website, yes. Even more so if you put posts in more than one category and tag at a time. You can end up with thousands upon thousands of pages showing the same snippets with no static content and the same meta descriptions over and over again. These are auto-generated pages and are against the Google Webmaster Guidelines. It's vague and arguable, but here's what it says concerning this discussion:

• Stitching or combining content from different web pages without adding sufficient value

Of course they don't slap a ton of people for not having the background to be able to fix it. But if it gets crazy enough, Panda will jump in and make you consult a coder. You can add sufficient value to page one easily, but you can't do that to thousands of pages. So I suggest no-indexing them. You can do some searches across the SEO community's blogs and find tons of examples of people getting pinged by Panda and them fixing it, with almost all of the scenarios being similar to what we're talking about.

Yoast uses some if-statements to apply the meta tag to the appropriate pages:

<meta name="robots" content="noindex,follow"/>

You would have to manually set up PHP if statements in your templates without Yoast to do it the same. Or if you notice how Wordpress is set up, /blog exists but /blog/page/ and /blog/page/1 404's until you type /blog/page/2. So you could create a rule in your robots.txt that says don't crawl anything in my categories including /page/ and beyond. Assuming you've submitted a sitemap you shouldn't have issues with crawling where Google can't find your pages. However, what you're doing is creating page rank holes this way, where it's flowing out and not landing on your pages. The best way to do this is how Yoast has done it with the meta tags.

I have decided to no-index / dofollow categories and tags (seem to remember reading it's not a good idea to have tags indexed either)

Let me clarify my stance in case I wasn't clear above. In your situation, I'd choose the main taxonomy and leave it indexable. Let's say that is categories. I'd no-index ALL tags, but leave them followed. For categories, I would only leave page 1 indexable, and anything past that, such as /blog/page/2 would become no-index. I would at least index the first page of your main taxonomy and add a static paragraph or so to make it worthy. These are things people might actually search for.

 

[Potatoe]

"Stitching or combining content from different web pages without adding sufficient value"

Of course it's phrased in that typical Google way where even after you've read the terms, you still don't 100% know if you're in the clear or not.

Cos if Google asked me whether or not having a way for my visitors to see every single article I've ever published that mentions Blue Widgets, or every article that mentions Red Widgets... I would say that's more than sufficient value...

Having said that - I'm going to re evaluate because I have links from my menu and on every page linking to category and tag pages (They each have their own uses, no overlap between cats and tags) but that's a lot of juice flowing all over the place.

 

[Ryuzaki]

Cos if Google asked me whether or not having a way for my visitors to see every single article I've ever published that mentions Blue Widgets, or every article that mentions Red Widgets... I would say that's more than sufficient value...

You could argue the same about dynamically generated search pages, and if you don't block robots from crawling those, you'll see them finding and even generating countless pages like homepage.com/?=search-term . These are valuable to users but not valuable as entry points into your site.

Same goes for pages 2-999,999 of your categories. Page 1, on the other hand, is valuable. A user will find Page 1 or any page with a search bar as valuable as the other 999,999,999 pages.

It's obvious why Google wouldn't want to index all of that. In a perfect world they'd know how to ignore it. I find it silly that they instead try to enforce their own rules on the internet. I also find it silly that Wordpress and other CMS's haven't made this a default feature.

Another way to look at it is based on their idea of Doorway Pages:

• Pages generated to funnel visitors into the actual usable or relevant portion of your site(s)
• Substantially similar pages that are closer to search results than a clearly defined, browseable hierarchy

That's definitely not the purpose of paginated taxonomy pages, but Google's robots can't always tell the difference.

Another fix, using canonical tags for paginated content:

• Use rel="next" and rel="prev" links to indicate the relationship between component URLs. This markup provides a strong hint to Google that you would like us to treat these pages as a logical sequence, thus consolidating their linking properties and usually sending searchers to the first page.

Google basically tells you right there, they only want to index the first page (the only one that's truly useful as an entry point).

 

[CCarter]

I don't agree with noindexing and nofollow things like tags and categories. The only pages I nofollow/noindex are pages with forms on them, so i don't get spam. If a tag page starts ranking then it indicates to me i have potential I didn't know about and can do some better interlinking to get things going. But by removing pages from the index you are putting bigger holes in your fishing net yet still trying to fish. Tags and categories are meant to help Google also understand what relates to what.

 

[RomesFall]

I use WordPress for 95% of my sites because I've got lazy and it's easy, but it's not that great out of the box for SEO as we know.

I'm more on @Ryuzaki ' side when it comes to the logic I'm using for my sites...

At the moment my newest sites I've totally ignored the tag / category pages, meaning I don't use them and the default pages are just NoFollow / NoIndex and not in my sitemap. I strip the links to them in the theme files. They effectively don't exist. The same goes for date based archives. You can still display the date, just remove the link. Author pages can be NoFollow / NoIndex and keep the link if you think your visitors are likely to use the page to see their favorite authors posts, but if you only have 1 author then what's the point of it at all?

Setting up like this is better from the outset because if all link flow considerations were equal (they're not, but we can assume they are for this kind of work) then you've already given important pages more juice than they'd get with using WordPress as it is.

This isn't perfect, but I prefer it and I've seen better results by ignoring some of these poor SEO features of WordPress. I prefer to build a custom UX for my visitors.

Instead I'm making do with a silo-like setup... /health/nutrition/ might be two pages I have, then I'd have unique content on each page, the health page is an overview and introduction to all sub-categories. It would display popular posts, or posts I want to boost in the SERPs and links to the sub-categories themselves.

The actual posts don't sit within that URL Structure because in WordPress you can't make a post a child of a page as you know, but this isn't the end of the world. Instead you can get the benefit of relevancy, but I also believe depth level of a URL is important in terms of the weight a search engine gives it.

So by linking to /post-name/ from /health/nutrition/ I'm actually telling the search engine that /post-name/ is more important than /health/nutrition/ rather than telling it that it's a direct descendent which would imply that it's not as important as /health/nutrition/.

As I said this isn't perfect, but I've seen better results doing this than setting up WordPress as is.

Here's an example of that on a site with a brand new piece of content I had last year that I ended up pulling the plug on:

The site this was on had a handful of links from my social media profiles and that's it. I think setting up WordPress in this way has resulted in most of my content indexing in the SERPs at positions <100 for up to most medium comp keywords (assuming content is good as well) on brand new sites.

 

[David]

Instead I'm making do with a silo-like setup... /health/nutrition/ might be two pages I have, then I'd have unique content on each page, the health page is an overview and introduction to all sub-categories. It would display popular posts, or posts I want to boost in the SERPs and links to the sub-categories themselves.

Hey @RomesFall, thanks for sharing your insights on this. What technique do you use for creating pages like these - do you create custom templates, or are you simply making a page in a WYSIWYG editor?

If you display popular posts - doesn't it make a duplicate content? I.e. the same thing that you were trying to avoid by removing regular Categories?

 

[Golan]

Ain't sure how i overlooked this thread.

My 2 cents: I hate SEO plugins, so for noindexing purposes i use Ultimate Noindex Nofollow Tool.

 

[jäger]

Just to extend on what RomesFall said about date and author pages etc.: The Yoast plugin has options for this, and also to completely deactivate them and simply redirect to the main category or wherever.

 

[Handel]

@RomesFall, you described very clearly of how you're manually crafting category pages. Are you linking from category pages to post of different categories? Theory says you shouldn't, each category is like a separate silo structure. Links can only be to other category's head, but not to somewhere deeper.

But let's take for example all those magazine style websites. They have their categories, that's fine. But when you browse the category down, all the related posts appear from same AND other categories. From usability and UI point of view this is completely justifiable - it keeps users busy and makes them spend more time on the site.

So, is it always a tradeoff between usability and SEO? Or am I missing something?

 

[Shimmy]

@Handel Have a look into custom post types. Having one for each silo means that posts from other post types (i.e other silos) won't be shown.

Mine are set up to be standard posts but have a hierarchy which lets my lower-tier posts only show their siblings for the next/previous page links. And as for the archive pages, you can set it to false and rewrite the slug to be a page (which acts as your top silo page).

 

[RomesFall]

So, is it always a tradeoff between usability and SEO? Or am I missing something?

If a post fits two categories I just add it to the one it fits most strongly...

Good site structure is good SEO and it's good Usability IMO.

This is how Google see's your pages from linking.

Level 0 - Homepage
Level 1 - Whatever you just linked to

This is how Google see's your pages from URL Structure.

Level 0 - domain.com
Level 1 - domain.com/category/
Level 2 - domain.com/category/post-name/

You should try to align these two as best as you can, which can be tricky... Even a typical blog homepage is going to give your most recent post a link from Level 0 to Level 1, while your category tells them it's Level 2. Does this impact rankings? Not in my opinion. That's the exception to the rule for me.

If you have two categories that are both Level 1 and both link to a Level 2 post that's where it gets tricky from a parent/child context, which could be confusing for a crawler.

A crawler is going to be able to more efficiently crawl your site if it's not going round in circles.

The reason it could be confusing is because you can only have one category in the URL Structure. If like me you want to keep those tightly themed for relevancy signals then it's better to just focus on the most relevant category.

I think anyone with relevancy in mind should start from thinking about site structure and looking at that through the eyes of the crawler.

I'm probably over-thinking it, but it's never harmed my sites and it's been an overall improvement for me compared to when I did things how most people do. Higher rankings from first indexation etc.

 

[Golan]

I like to use the category description in Wordpress as a piece of static content that never rolls off to page 2 like the snippets do. You can add a paragraph, links, pictures, whatever.

Ryu, any simple way to do that?

 

[Ryuzaki]

@Golan , assuming we're talking about Wordpress:

1. If your theme already includes category descriptions in the category template, you'll want to click through the left hand side menu in the dashboard where you create your categories. There's an option to give them a "description" where you can give it a name, a slug, etc. That description can take HTML, links, pictures, whatever. You can slap as much and whatever you want in there.

2. If you do that and find that it's not appearing in your category pages (usually theme designers will put it below the title and above the rolling posts, then you need to add a snippet of code to the category template:

<?php echo category_description(); ?>

That's be generic with no parameters, meaning that you put it in once and it'll pull the description for every category depending on which is being viewed at the time. Quick and easy!

 

[Handel]

If a post fits two categories I just add it to the one it fits most strongly...

@RomesFall, thanks for an in-depth answer. But you slightly misunderstood my question. My posts are per single category only. What I meant was that some themes have kind of Infinite Scrolling feature - after initial post they load other posts from different categories as you scroll the page down.

This is definitely OK from users point of view – it keeps them busy on site. From SEO side, I see it as a shortcoming, because it's linking from one category straight to other category's posts. This is where my question came from: tradeoff between usability & seo.

 

[Golan]

@Ryuzaki Wow i should have missed this some time in the past years when this was introduced. So many thanks.

 

[Ryuzaki]

@Handel , to give you another perspective on the matter of silo-ing...

Think about all of the other links on the page, such as your main navigation menu and footer links, sidebar links. If you control all of the contextual links, each page is still likely linking to the other categories through those navigation elements.

You could argue that Google knows the difference between navigation, supplemental, footer, and contextual. And thus those extra links don't matter. Or you could attempt to be very strict with the silo linking by leaving out navigation, which would be horrible for usability.

We know Google's entire purpose is to rank pages based on how useful they are to the user. We also know that, while they try to impose their will on the web designers of the world, sometimes we don't care.

An example is the last few years where "Hero Images" became a huge thing. Hero images are those giant splash images people are putting at the top of their posts, either above or below the title. They eat up a TON of "above the fold" real estate, which is an absolute no-no for Google if you want to rank... They want readable content above the fold.

However, they had to back off of that to a degree. The fight became less about the percentage of the Above-the-fold real estate being text and more about it just not being a ton of ads. Designers wouldn't stop using hero images and the users liked them. They grabbed their attention fast and told stories and stirred up emotions.

My point is that Google evolves along with the internet when they can't control the tide. Case in point is Silo's. These were powerful and relevant in the flat-file HTML days of simple web design. Now 99.5% of people are using CMS's that are by default chock full of interlinking all over the place.

So how did Google adapt to this to still gain an understanding of content relevance? Partially the content linking to your content, the anchor text being used, etc. That's internal and external.

But for silo's specifically, I literally think all you need to do is have breadcrumbs with the proper schema-markup. Also using a nested URL structure helps. Otherwise you should interlink to relevant content that will provide the best user-engagement metrics possible.

I can't find any examples in the SERPs to show you without outting my own websites. But if you look beneath the titles you'll 99% of the time see the full URL to the page:

htt p://www. asdf. com/asdfasdf/asdfasdf/asdfasd/

That's because they have no mark-up. They might want to force this understanding on Google using interlinking. But if you look at a site with proper mark-up you'll see under the title something like this:

Home > Books > Genre > Fantasy > Dark

This schema markup stuff isn't just for pretty displaying in the SERPs. It's just like the ole meta-keywords tag of old. "This is exactly what's going on, Google."

Anyone reading this and interested in more should check out CCarter's write up in the Crash Course: Day 26 - Schema Markup

 

[Handel]

Thank you @Ryuzaki, your answer with some historical perspective made the whole thing much clearer.