Recovering Failing SEO Sites with the Kitchen Sink Method

Ryuzaki

お前はもう死んでいる
Moderator
BuSo Pro
Digital Strategist
Joined
Sep 3, 2014
Messages
6,229
Likes
13,099
Degree
9
I started typing this as a reply in another thread and realized it deserves a repository of it's own in it's own thread. This is mainly for failing SEO sites, but if you're serious about your SEO then there's zero reason not to run through this for your own succeeding projects or ones that need to bust into the next level of growth.

The Problem​

Your site was growing. At a certain point you realized it plateau'd. And then it started to lose traffic. Or it was doing a meteoric rise and then nosedived off a cliff. You have no clue why.

The reason you aren't sure is because these situations aren't usually the fault of any one page or set of pages. It has to do with "everything but the kitchen sink". It's often you coming under fire of a dozen sub-algorithms and offline filters. I almost called this "The Nuclear Method" because we're going to dig into every atom of your site.

The Solution​

The solution is to follow what I'm calling The SEO Kitchen Sink Method. Shout out to Glenn Gabe at GSQI.com for turning me on to the philosophy that "There's Never One Smoking SEO Gun."

Estimated Time to Recovery: 6 months
It will take time to complete (a month), and then it will take time for Google to recrawl your site (a month or two), and then it will take time for Google to collect the data it needs (a month or two), and then you'll need to wait for Google to use that data in the next refresh off the offline algorithm filters (hopefully within the 6 month period). If you miss the window, you'll have to wait even longer. It's important to do your part fast.

Six months is nothing. Here's why: I had this problem with a site I worked on for like 5 years straight on a new domain from the ground up. Built it up to DR55, 500 posts or something. I assumed I hit a natural plateau and eventually moved on to other things until I realized this wasn't right and it was algorithmic. So I applied The SEO Kitchen Sink Method I'll describe below, which played out in the time frames mentioned above.

The Result​

In case you think six months is too long or that I'm exaggerating the results or that it's not worth doing or whatever other mental resistances you may have, just look at the estimated traffic charts from the site here (the reality is much better):

w1Xbsxs.png

You can see the plateau and eventual erosion of traffic and then the recovery. Now envision another exponential curve from day 1 up to the present day, and you can see that I gained back every single bit of momentum and growth I should have had.

You can get it all back, too. Here's how:

The SEO Kitchen Sink Method

The philosophy is that you don't know what the problem is. You'll likely find out later like I did, but for now you don't know. It could be one thing, it could be a dozen things. We're leaning towards "it's everything" and we're going to treat it as such.

We're going to overturn every single stone and see what's beneath it. We're going to improve every single variable we can. We are going to manipulate "everything but the kitchen sink" until your money maker is back and printing cash.

You will need to learn things here. You'll need some tools you've never used. You'll need a spidering tool (Xenu Link Sleuth, Screaming Frog, Integrity [for Mac]). You'll need a backlink index (Ahrefs). You'll need to be able to set up child themes and do a lil HTML / CSS / PHP and maybe JS.

If you don't know how to do something, keep working through The SEO Kitchen Sink Method and do as much as you can and take notes on what you need to go back to. But tear through what you can first. The key is to move fast at the start. A partial recovery is better than none because you bogged down.

You'll be doing the big stuff, the small stuff, the tedious stuff, the stuff you think is a waste of time. Every single stone will be upturned. And it's worth it because you're going to improve every aspect of your site, which will only multiply the revenues once it's back. And there's a huge chance that you'll find a few smoking guns, too, like I did, that were hiding way deep in the jungle that is your site.

Let's break this up into the three main disciplines of SEO plus another: Technical SEO, On-Page SEO, Off-Page SEO, and Branding Signals.

Here's why you need to deal with each one:
  • Technical SEO - I believe that if you save Google resources, they will return you the favor. Plus there's user experience tucked in here.
  • On-Page SEO - If you aren't on the playing field, how can you win? If you over-optimize or under-optimize, you're dead in the water.
  • Off-Page SEO - Now you're on the playing field and it's time to win, but there might be some toxic sabotage going on.
  • Branding Signals (EEAT) - This is all about Authority and Trust, without which you have no business ranking highly anyways. If your site falls over in the forest and nobody hears it... nobody cares but you.
Let's do it, one at a time. There's no particular order, but do them all. If you need help with anything or want to probe deeper, ask in this thread and the community and I will help you. It's what we're here for and we're inviting you to ask. The (free, just click the link) Digital Strategy Crash Course will be your best friend for a lot of this. Most of your problems, if you aren't lazy or greedy, exist in the Technical SEO side of things.

The part below will be purposefully more "checklist-y" than a novel from here out. As more pop in my head, I'll come back and add them.

Technical SEO

Note: Create a child theme before you do anything else, if you aren't working on a completely unique and custom theme already, or you will lose your work when you update the theme. You may want to do a lot of this on a staging server and then port the site back to the live server at the end.

1) HTML, CSS, JS, & PHP Errors
Find them and fix them. Learn about your browser's developer tools and the JS console. Learn to look at error logs and print the PHP errors to the screen. Fix everything. Validate all HTML and CSS. No errors allowed! It's a good time to check if your jQuery is using deprecated functions, etc. Don't worry about ad scripts, you can't fix those. Do a Schema test while you're at it.

2) Update the Server Stuff
What I mean here is to make sure you're not running PHP 5 when PHP 8 is on the horizon. Make sure your plugins and theme are compatible first. This is a tune-up under the hood for speed. Look at your MySQL or whatever database technology you're using. Get everything up to speed, because that's what this is about.

3) Page Speed Optimization
This is a gigantic section. Learn more about Page Speed Optimization in the Crash Course. Do it all, from speeding up the site to the core web vitals. This is a mobile index and if your site is slow on mobile, what do you expect? Google does NOT want to rank sites with bad user experience, and speed is a foundational aspect of that. All of the stuff you need to target is in that crash course post, freely available.

4) Indexation Bloat & Quality
Do a site:yourdomain.com search on Google and note the total indexed pages there. Look in Search Console's index reports and check the total there. Do they match with your number of published posts and pages? If they're low, you have a quality problem and Google doesn't want to waste resources on your crap. If it's high, then you have a bigger problem which is also dragging down your Panda quality score. Find it and fix it. I go as far as to noindex paged archive pages that are paginated, like /category/page-2/ but always allow the first page to be indexed. Always keep them all dofollow, because the page rank will still flow.

5) Sitewide Schema
Are you using Schema at all? I don't mean stuff like 5 star reviews on product pages, but typical sitewide stuff like Geolocation data (for Local SEO), all the branding and meta descriptions and meta titles and published dates and updated dates and all that. Yoast does a great job with this out of the box. Are you associating your social media pages using the "sameAs" schema for branding signals? Make sure you're using this stuff and that it's accurate. Use Google's Schema tester to make sure there's no errors.

6) The SEO Pebbles
Use your spidering tool and crawl your whole site. Does every page have an optimized meta title of an appropriate length? What about the meta descriptions? Does every single frickin' image have an alt text? Do all of your form fields have a <label> for screen readers? Dig up info on these seemingly tiny things and fix them, because they aren't tiny.

7) Nofollow Affiliate Links
Are all of your affiliate links using rel="nofollow" or ugc or sponsored tags? Do your blog comments that allow links tagging them with ugc? Fix this crap. You can't have dofollow links that are paid or incentivized (by affiliate commissions).

8) HTTPS and HTTP/2
Is your site SSL complaint? Is it HTTPS? And while you're at it, are you running it on HTTP/2 for a much faster experience? Make sure you're getting the secure lock in the browsers, meaning you aren't loading any 1st or 3rd party HTTP resources from a non-HTTPS domain.

9) Responsivity & Mobile Friendliness
If you don't have a responsive design by now, you deserve to have tanked. Again, it's a mobile index now, you have to be responsive. But you also want to use the Google Mobile Friendly test (don't rely on the buggy Search Console notices) and make sure there's no errors. But even that Mobile Friendly test is buggy and won't load CSS and JS, so you'll have to manually check for problems they claim are there, for a sanity check.

10) Sitemaps & Search Console
If you aren't using Search Console, sign up NOW and let it populate the data over a couple weeks while you do other things here. Submit your sitemap index and it will give you great info about 404's and other problems with canonicals and anything else. But the main thing is to have a dynamic sitemap that updates the dates when you tweak your content so Google crawls it. It also helps them revisit every page over time. Crawling matters less when they have a literal list of your content to start from.

11) Click Depth
There's not much you can do about this once your site is huge other than work on your category pages, provide sub-category archives, and show a lot more posts per category page. The more the merrier. I'll go up to 50 on mine. The point is to share the page rank around better and to help spiders encounter all of your pages.

12) Content Pruning
This is another big task where you'll decide what content is not high enough quality to be in the index. By deleting low quality pages, your quality score will rise. You will NOT decide based on traffic or links. You'll decide based on quality and if the content is serving the intent of the queries. Improve them, delete them, or 301 them. You probably want more feedback about this so please ask here in this thread.

13) Self-Referencing Canonicals
Every single page needs to declare itself as the canonical version of that page (unless we're talking eCommerce, in which case ask for more details in the thread, especially with faceted navigation, etc.). Yoast makes this simple if you're using Wordpress.

14) Breadcrumbs
Use them, get them setup with the proper schema in them, and spread the page rank around. Google will display your site's URLs in the SERPs differently if you do this right. It's good for users, so good that Google displays it. If Google displays it, it's probably a part of the algorithm, no?

15) Accessibility
Even people who are completely blind are on the internet and they're meant to be able to use it as well as anyone else. You need to account for this and fix any accessibility issues. It's not just blind people but partially blind, color blind, deafness, people who need extra tools to help navigate like screen readers and other devices (not a keyboard and mouse), etc. Work on basic stuff like text and background contrast ratios and go as far as labeling links with title="" tags and form labels and the works. Videos should have closed captioning. All of it. Even text size and line-height.

16) Broken Images
If you have a broken image, you have a 404. Make sure to find these (use a spider to crawl) and fix them. Also, make sure Wordpress isn't creating "image pages" where nothing exists on a page but the header, footer, and image. That's thin content. Yoast will redirect all of these back to page the image is embedded on. This is a super good thing.

17) Broken Links - Internal & External
If you have broken links on your site, you're causing Google a headache and wasting their resources. Crawl all of your external links and remove or replace any 404 links. I go as far as to even fix ones that are 301-ing, like from http:// to https:// now. For internal links, do the same. If you 301'd posts around, update the old internal links to go direct to the new URL rather than through a redirect chain.

On-Page SEO

The goal here is to strike a balance of being optimized without being over-optimized. Over-optimization is typically the problem here, as well as not concerning yourself with QDF (mentioned below), and other issues. This is not a question of how to do on-page or where to do on-page. This is just about the things you need to check that aren't already covered in the Technical SEO part.

1) Interlinking
Stop worrying about manipulating the flow of page rank (worry about manipulating the flow of relevancy and topical authority). Stop worrying about silos. Just interlink every chance you get. This also helps reduce click depth. Interlink in the main content contextually, do it with a static list of related posts at the bottom of the post. Do it out of your footer and sidebar. Let the page rank flow so your average tide rises and lifts all of your boats (posts).

2) Google Fred: BoFu & Above-the-Fold
Long story short, concern yourself with two things here. If you don't have your content above the fold, get it there. Giant splash images are okay now, and so is monetization. It's not about whether there's monetization above the fold, it's whether there's content above the fold. But don't have display ads above the fold, not on mobile these days. Also, if your site is one of these Top10BestProductReviews.org sites that only works the bottom of the funnel, RIP. I doubt you'll recover. Do not publish only "money pages". You need more top of the funnel (ToFu) and middle of the funnel (MoFu) content than bottom of the funnel (BoFu). Or you're a spammer in Google's eyes. Yes, some sites survive, so what. 99% of them did not.

3) Serve the Intent of the Queries
Do not bait and switch users or Google. Do not optimize for information queries and then use the content like a "money page". Google isn't dumb. Also, stop creating gigantic mega-posts. Each post needs to target one tightly-woven basket of related queries and only those, and then the content should serve those queries directly. Mega-posts do not work any more. Serve the intent as directly as possible without fluff. I can't under-emphasize this.

4) Queries That Deserve Freshness: Updated Dates
In this post under the "Self-Sabotage: Queries That Deserve Freshness" header, I show why we do this and what the results are (+20% traffic for a while as the Freshness decays over time). I also provide the solution on how to do this there. The point is, stop showing published dates and show "Updated:" dates instead. You don't even need to edit the content, just tweak the date by pressing "update" in your dashboard (or use a plugin to update all dates at once) and reap the reward. The point is, most of your queries deserve freshness and you don't realize it. This could solve a big chunk of your eroding traffic.

5) Topical Authority
As you recover (not now), stop spraying and praying, especially if you have a more generalized site. You need to publish enough content on one single topic and interlink it all until Google sees you as a topical authority for that tight topic. Over time you'll gain this authority over broader topics and entire categories on your site. When you do, publish for every keyword you can and watch the money roll in with easy rankings. Then move horizontally to your next category and repeat. You may be eroding traffic from having no authority for any of your medium-to-high-competition topics.

6) Keyword Stuffing
You'd be amazed how many people still do this, and you may be one (or your writers). Stop thinking about keyword density in percentages and think about keyword incidences. For a 1,000 word post, 5 or 6 times is enough. For a 1,500 word post, 5 to 8 times is enough. It's about WHERE you put them: Title / H1, H2, First 25 words, Last 25 words, in the featured image alt text, and then your last 2 or 3 should be spread equally spaced through out the rest of the content.

7) Grade Level Reading Score
Unless you're doing medical content or whatever, stop writing like a scholar. Always, always, always write like you're in a casual conversation with a friend. 90% of queries demand elementary school reading comprehension levels. People are dumber than you. Write like you're talking, and talk like a normal person. I'm not saying to have all of your content re-written, but future posts should be in normie-speak.

8) Enhance Your Content
This is very simple. Googlebot is a spider, and I call this spider food. The more unique HTML tags you use, the better. So enhance your main content with lists, tables, embeds, videos, images, quizzes, tables of contents, resource sections, sub-headings, data blocks, whatever. Do it up! You'll find it easier to get natural links this way too. The more NERD entities you can use, the better, too.

9) Use Your Headers Correctly
Your H1 comes first, then H2's. H3's only come under H2's before you return back to the next big H2. Stop mixing and matching. There should only be one H1 header, probably built into your template. Nest them correctly so Google doesn't have to decipher your nonsense, because it won't work out in your favor because you're costing them resources.

10) Use Images & Don't Be Lazy
Images are spider food and offer more chances to use alt texts which gives you chances to put keywords in there (don't count these in your incidence usages as mentioned above. Use your long-tail variations and one exact match). Don't just have your one featured image and then a wall of text. Try to use an image once every 500 words maybe, depending on the structure of the content. List posts can have one every 200 words. The more the merrier (but optimize the resolution and pay load).

11) Over-Monetization
Not only has Google cracked down on this on their own but they joined the Coalition for Better Ads. Thankfully most big ad companies have joined too and have set maximums that don't exceed what the CBA and IAB allow. But that doesn't mean you aren't going HAM yourself. Look into this.

12) Keyword Cannibalization
If you have multiple pages aimed at the same keywords (perhaps by accident), Google will continue to test them all to see how they perform. And most won't be as qualified as the page you want to rank, so you end up with situations like this:

DEVcOmr.png

That's like 10 pages competing for the same term on the same site. You end up seeing cumulative results that look like this:

nOGzMdM.png

Fix this by de-optimizing the accidentally optimized posts, 301 redirecting them to the main post, or setting the main post as the canonical.

13) Outbound Linking
The web is meant to be interconnected. Don't be greedy about "juice." Google wants to see that you're referencing high quality sources so your content can be more trusted. Craft a killer outbound link profile to .gov, .edu, to huge trusted sites, to topical experts like local business blogs, etc.

Off-Page SEO

This is all about dodging the Penguin filter, meaning not being over-optimized or toxic.

1) Ignore Most Automated Links
Trust me on this one. I wasted a lot of time constantly disavowing until I decided to trust that Google can ignore the obvious bad stuff. Alexa lists, keyword research lists, image scrapers, site valuation lists, wikipedia copies, etc. You can ignore this crap. But you will be disavowing so it won't hurt to disavow what exists if you want to play it safe. But most of it goes offline a month later anyways.

2) Disavow Your Toxic Links
There is stuff you'll want to disavow. Sometimes PBN owners will have used you as a co-citation source to distract Google or get them to trust their PBN site. That didn't work and hurts you too. Some automated web 2.0 spammers will create 100's of sub-domain web 2.0's and link to you. Disavow those. I've seen and recovered a penalty from this with 100% links to image files. Yes, an algorithmic penalty for web 2.0 links to image files (not posts). Don't be greedy. If a PBN linked to you, disavow it, no matter how strong.

3) Check Your Qualifications
It may be the case that your niche got more competitive while you rested on your laurels as Google tweaked the "link power knob" to get rid of spammers and boost brands (this is what YMYL is, by the way). If you're targeting keywords where the top 10 is full of DR70 sites, and you're DR25, then there's a huge problem. You need powerful links in a hurry and need to publish for easier keywords in the future.

4) Over-Optimized Anchor Text Profiles
You want to get Penguin'd? Have your exact match keyword's ratio usage exceed 10%. I don't even feel safe at 5% because these numbers are going to get tweaked in the future. The way I do it is only ever use one exact match anchor per page. One and no more. If that's your only link and you're at 100% usage, that's fine because you don't have statistical significance to be seen as a spammer. But stop this greedy anchor text usage. It will kill your project.

5) Low Grade Chef Cuisine Dog Food
You may think you have a great backlink profile. But really it's all nofollow blog comments, forum posts, wiki links, profiles, and crap Google ignores. Ahrefs might tell you that you're DR70 because they think johnssite28.blogger.com is DR85 when it's really DR ZERO. Get honest about your backlink profile and how trash it is. The only links that move the needle these days are contextual links inside the main editorial content (with exceptions of course).

6) Link Relevance
Did you build your new bicycle site on an expiring wedding domain, see success, and then tank? That's because Google realized the content relevancy of your backlinks does not match the content relevancy of your posts. Nice links are only nice if they're relevant. If they aren't, it can actually be worse for you, because Google understands relevancy by more than just what's on your site, but what's connected to your site (search for my discussions on "mini-nets" on the forum).

7) Lazy 301 Redirected Domains
Related to #6 above, maybe instead of building on an irrelevant domain you redirected domains in instead. Maybe they're not relevant, or maybe they are but of poor quality. The content on those sites was bad, the links are bad. Maybe you were lazy and just did a global redirect to the homepage instead of mapping topic to topic, contact page to contact page, and so forth. Google knows that trick and acts on it at times.

8) See More Below
The Branding Signals section below has a lot more to do with off-page SEO. Keep reading and use those too.

Branding Signals (EEAT)

Google has and has had a spammer problem. They solve that with two methods: delaying or randomizing results for a period of time, and by promoting brands. This is how you become a brand in Google's eyes.

1) Populate Social Media
You need to have a Facebook Page, Youtube, Twitter, Pinterest, and Instagram profiles as non-negotiables, and you need to populate them with content. Yes, you should try to get engagement and all that, but realistically you can share your own content and fill it up. Continue to do this as you publish in the future. Each shared post links to your site. The profiles should also link to your homepage. While you're at it, make the profiles designs match your site. Use your logo and colors. Keep using these to keep them active.

2) sameAs Schema
You need to tell Google directly that these social media profiles are related to your brand. You do this by adding the sameAs schema in your sitewide <head>. Yoast makes this easy. Do it, it's non-negotiable.

3) Dofollow Social Links Sitewide
Figure out how to link to your social profiles sitewide with dofollow links from the footer of your site. Do not use a plugin for this or your site speed will tank. There are means and I'm happy to share in this thread on how to do this and keep it fast. But the point is to associate these profiles with your brand and juice them up a bit.

4) Homepage Links with Branded Anchors
Go look at a real brand's backlink profile and see how many of their links go to the homepage and mention their brand over and over. You need this going on, too. A greedy SEO links only to the inner pages. A wise SEO hits the homepage too (and a good marketer attracts these naturally) and lets the juice flow through the site, raising the average page rank of all pages. With enough of this, you'll auto-rank for lower competition keywords and can scale publishing and focus on topical authority for a while, which is almost like free money. There's tons of ways to get these under the "Other Easy Wins" section in the Off-Page SEO day of the Crash Course. Just make sure you get them crawled and indexed and accept many will be nofollow, but the boost to your referring domains is worth it. Just treat the process respectfully and not like a spammer.

5) Get Links Closer to Seed Sites
Google theoretically has a list of trusted seed sites and flows something like "TrustRank" from them. The fewer hops you are from these, the better off you'll be. A good way to think about this is .gov and .edu sites and other official organizations. They link out to real brands, you know.

6) Get Links on Sites Where People Are
Brands are followed by people, and real brands get mentioned and linked to from sites with high traffic. You should have a link from Wikipedia, from Reddit, from the big forums in your niche, from all the social media sites, from news sites. Make a list of targets and make it happen.

7) Unlinked Mentions / Becoming an Entity
This isn't about wasting link opportunities, but you can get yourself turned into an "entity" using everything above (especially appearing on Wikipedia with an actual page for your business, not just links). A great way to do this is to drop unlinked mentions where you'd otherwise get moderated for dropping a link. Blog comments and forums are easy targets. Non-social media social sites like Reddit with static URLs are easy. You should be such a marketer that you start earning these naturally anyways.

8) About Us / Contact / Terms of Service / Disclosures
If you don't have this boiler plate content, you aren't real. You're possibly a spammer in Google's eyes. Real companies let you reach out to them.

9) Author Boxes
You don't have to link to author archives or anything like that. But every post needs to be attributed to an author. Having an author box (where each author name links back to the About Us page where they are listed is good), with a little blurb about the author, is likely a trust signal, if not just a correlation that the algorithm has picked up on. Have a face and a name and keep it consistent. Multiple authors are fine, too. Just attribute your posts to anyone, even a pseudonym. It'll populate data into your schema too.

10) Prove You're Real
One thing Google is concerned about is "is there someone to be accountable for this info?" Which is why we're adding Author Boxes. How else can we do this? Beefy author pages, if you desire, that link out to social media profiles and are much more substantive than what goes in author boxes. Show experience and credentials, too. Another option is to list an address, phone number, and every contact method you can think up to at least the About and Contact pages. Look at the things local businesses do, and emulate as much of it as you can. Build name/address/phone (NAP) citations to the site.

11) Prove You're Worthy
List "As Seen On" style logos at least on the About page, if not near the top of the homepage. List awards and accolades. Again, list credentials and experience. Do things like talk about your editorial process, how you value "buzzword" over anything else. Make use of your sidebar for things you want seen by everyone/bots on each page (your team, awards, as seen on logos, popular posts). Label them in such a manner to get rid of disambiguation.

12) Use Schema & Microdata
Don't do all the above stuff about authors, dates, people, contact info, addresses, and so forth, without shoving it down Google's throat. Mark up everything you can (without going overboard into the weeds, because they provide markup for all kinds of nonsense). Yoast can get your "organization" or "person" data into your <head>, and the rest you can add in custom HTML boxes or within your PHP templates. Use the Schema.org Markup Validator and Google Rich Results Tester (both here) to make sure you did it right. It's really convoluted and involved, so make sure you test it. It's a nightmare, so be ready.

-----

That's what I just brain-dumped in one go. More will come to me and I'll return and edit them in.

If you can go through this, you'll fix the Technical SEO problems that are likely plaguing your site. You'll stop over-optimizing with on-page, off-page, and monetization. And you'll become a brand. Your trust and authority will rise. Your user metrics will rise. All of this gets fed back to Google through the clickstream, through Analytics, through Search Console.

And hopefully, if you were thinking critically as you went through this, you found one, two, or even three actual smoking guns like I did. The benefit of going through all that mess above is that not only will you escape your impending or already-arrived doom, you'll come out the other side much better off than you ever were. The site I showed that recovered above not only gets about 5x the traffic it was but is also converting about 2x better on everything than it was. Yes, it's making around 10x the money it was when it was circling the drain (and before).

If you do this and it succeeds, please share your story here. If you do it and fail, let the community help find other things for you to investigate (keep and share your notes). There's always more for me to add above and more for all of us to learn. No site is doomed. There is always a solution (except in some cases like YMYL or Fred [I've not seen a recovery yet despite the efforts of friends]).

One final word of advice. Do not delete your site in a fit of rash decision making. Let it ride if you can't get to it or can't solve it. Sometimes the dials get turned and you pop back, or you'll learn something in the future that'll fix it. The worst case scenario is you don't recover immediately but your site gains more links, content, and most importantly time in the index, which leads to more trust.

Hang in there and put in the work!

Remember, the goal here is to work on everything but the kitchen sink. As you improve every little tiny thing (and big things) you'll encounter other problems and hopefully discover your smoking gun problems too. Not only will you pop free of the algorithmic penalty but you'll grow faster than ever before with better converting everything.

Put your hands on everything, improving it all, and the core problems will be found and fixed.

That's The SEO Kitchen Sink Method.
 
Last edited:
When I read montage posts by you and carter I think that I should just hand my ftp credentials over and beg for a cut of the scraps from what ever gets done with one of you in the drivers seat.

Love a lot of the line items, I can’t even imagine being organized and motivated enough to attempt the full Monty.

If you were only going to do 2 or 3 items on this list for a mass articles site. What would you do?

If you had some domains with major links and brand traffic but little content and a lot of institutional inertia to doing things what would be the top couple items that don’t involve adding tons of content.
 
Last edited:
If you were only going to do 2 or 3 items on this list for a mass articles site. What would you do?
I mean... this really isn't for the self-admitted and self-condoned lazy person. So I'd either accept that my site is done for or I'd pay someone competent to go through all this.

But to answer your question, since we're talking about failing sites, which implies that they were once succeeding, I'd look for the big problems first, especially those that involve algorithmic penalties or at least demotions.
  1. Indexation Quality - Look for big problems like blank pages being indexed, security issues with auto-generated spam pages, search query pages being indexed, and extremely low quality content being published.
  2. Link Problems - Accumulating, without your awareness, crap like PBN links, automated web 2.0 PBN links, bad neighborhood links, etc. Also over-optimized anchor text profiles.
  3. Page Speed - I'd improve this as much as is reasonably possible. It can act as a ceiling on your potential rankings, because Google doesn't want to send their users to sites with bad user experience.
Page Speed isn't a smoking gun but the other two most certainly can be. We're essentially talking about Panda and Penguin there. I'd also look at Google Fred issues, which can also be a smoking gun.

By working on all of the various items above, the idea is that you may uncover some smoking guns that weren't apparent before or that you weren't aware of even existing.

But another point is also that this can be a cumulative issue, where 50 things that are wrong are combining together to drop you below a certain quality score, or pushing you over a radar threshold, etc.

If you approach this lazily, that's the kind of results you'll get. If you go through the whole procedure, when you do recover you can add a multiplier to your already boosted revenue.
 
Offer this as a service?

I'd say the money is more in buying and fixing, but yeah, give away half of what they can improve the earnings by.

Also, if your site is one of these Top10BestProductReviews.org sites that only works the bottom of the funnel, RIP. I doubt you'll recover. Do not publish only "money pages". You need more top of the funnel (ToFu) and middle of the funnel (MoFu) content than bottom of the funnel (BoFu). Or you're a spammer in Google's eyes. Yes, some sites survive, so what. 99% of them did not.

Does this go for keywords or for onsite factors?

If I do info content, but have a "top 3 products" ad above the fold, does that detract?
 
Does this go for keywords or for onsite factors?
A part of what Fred seemed to attack was over-monetization in general. But the whole "thin content" penalty which seemed to have been rolled into Fred was about content intent (focusing only on the BoFu). You could essentially consider that as being about the keywords, especially if you target one main one per article.

If I do info content, but have a "top 3 products" ad above the fold, does that detract?
I'd be careful doing this anyways because you're mismatching the intent of the query with the intent of the article.

Thinking about things like the "reasonable surfer" patents and all that, Google knows that what's above the fold and at the start of an article contains the essence and emphasis of what the full article is about. So they could potentially think your content, even if serving the informational intent of the query, is commercial in intent.

And that's something they've whacked a long time ago, the old bait and switch where we'd write commercial content for informational queries, long before they understood topical relevance and intents. I'm sure they understand it even better now.
 
Cool, yes, I checked my analytics and it doesn't seem to work very well, at least not on a per visitor basis. It is always tempting to do this with high value products, because you just need one conversion to net you a good profit.

I think I will replace them with cta links to money articles.

As an aside, I'm considering if my kitchen product site, that I wrote about here, seeming to fail strangely, might not have been hit by such a penalty. It is basically only commercial content and not just that, only "best of" and "cheap" content.

Which is weird, because I also have a site about furniture which is basically just 20+ products listed ad nauseam, but I guess those could match the intent "lots of products to be inspired by".
 
This is another big task where you'll decide what content is not high enough quality to be in the index. By deleting low quality pages, your quality score will rise. You will NOT decide based on traffic or links. You'll decide based on quality and if the content is serving the intent of the queries. Improve them, delete them, or 301 them. You probably want more feedback about this so please ask here in this thread.
Is there a fast path to success here? I'm a bit confused on the what and how with regards to this. If it's not based on traffic or links, then what? Can I have more details here if possible? Maybe some examples?
 
@mrpotato, if you want a framework, you could do something like this:
  • Pull all of your pages on Ahrefs by "Best Pages by Links" and sort by best and most links.
  • Make sure those are all up to snuff. They probably are, but if you can improve them, do so.
  • Then filter all of your pages by most traffic, and de-dupe the list by removing the ones you've decided are fine from the earlier steps.
  • Make sure all of the remaining pages getting good traffic are up to snuff. If not, improve them.
  • Now you're left with pages getting little traffic and little links. Visit them all and mark them as one of three options:
    • Improve, 301 to relevant page, or delete & 404​
  • When you're done, you'll be done. You'll have pruned the bad content or improved it.
301-ing to relevant pages should really only be done if there's okay traffic and okay links, and otherwise the content can't be improved or it's keyword cannibalizing better pages, etc.

Sorting by links and then traffic just gives you an order in which to work and shows you which pages are higher priority and probably better quality. You're really just getting those out of your way. Those steps should be fast.

It'll be the low-traffic, low-linked ones you want to really look at. But that does NOT mean they're low quality. But it is a sign that they aren't performing for one reason or another. It could be a competition issue. Don't assume they're low quality.

A page with zero links and minuscule traffic can be a great and delicious page to Google and serve as a relevancy-pusher and interlinker, etc.
 
Excellent info!

I have a question. Let's say my inner page /Acme-Candy does have an over-optimized backlink profile. Are there other ways instead of disavowing good strong links to correct it, like changing the URL to /Acme-Candy-Company and keep the old page and make a contextual backlink to the new page, but change title, and headers, etc.. of old (or is it starting over the aging process with the new)?
Thank you
 
It'll be the low-traffic, low-linked ones you want to really look at. But that does NOT mean they're low quality. But it is a sign that they aren't performing for one reason or another. It could be a competition issue. Don't assume they're low quality.

A page with zero links and minuscule traffic can be a great and delicious page to Google and serve as a relevancy-pusher and interlinker, etc.
This could be an entire thread on its own. It's so counter-intuitive...

Cool, yes, I checked my analytics and it doesn't seem to work very well, at least not on a per visitor basis. It is always tempting to do this with high value products, because you just need one conversion to net you a good profit.
I think on info pages you are better off "warming up" the visitor first anyway.
 
This could be an entire thread on its own. It's so counter-intuitive...

I think on info pages you are better off "warming up" the visitor first anyway.
Sure, but with a 30 day cookie and high commissions, my reasoning was that it would work enough to make it worthwhile.
 
Let's say my inner page /Acme-Candy does have an over-optimized backlink profile. Are there other ways instead of disavowing good strong links to correct it, like changing the URL to /Acme-Candy-Company and keep the old page and make a contextual backlink to the new page, but change title, and headers, etc.. of old (or is it starting over the aging process with the new)?
You're operating under the assumption that disavowing a link nullifies it's anchor too. I don't believe this is correct. As far as we've been told from the horse's mouth, disavowing simply causes Google to add a nofollow tag to the link the next time they crawl it (internally, not in reality).

And what I know for certain is that nofollow anchors still count. Do they contribute to over-optimization? I think so, but I haven't tested that in about 10 years. But they did in the past. I used to use it to do an SEO trick. I'd purposefully get over-optimized for a short tail or one-word keyword. Like nearly 100% usage, all with nofollow links. The page would never rank for that term, but that wasn't the target term. The target was the one-word + another word. I'd get one dofollow anchor with the exact match of the target term and slam to #1. Because I was hyper-optimized for it. I don't care to explain further but it should make sense if you think about it.

The best way to correct an over-optimized anchor text profile is to build more links. But if it's completely bungled then sure, you could do what you've described by creating a new URL for the target term, de-optimizing the old one, and linking to the new one from the old. But that's only if the old one is so totally screwed there's no saving it.
 
You're operating under the assumption that disavowing a link nullifies it's anchor too. I don't believe this is correct. As far as we've been told from the horse's mouth, disavowing simply causes Google to add a nofollow tag to the link the next time they crawl it (internally, not in reality).

And what I know for certain is that nofollow anchors still count. Do they contribute to over-optimization? I think so, but I haven't tested that in about 10 years. But they did in the past. I used to use it to do an SEO trick. I'd purposefully get over-optimized for a short tail or one-word keyword. Like nearly 100% usage, all with nofollow links. The page would never rank for that term, but that wasn't the target term. The target was the one-word + another word. I'd get one dofollow anchor with the exact match of the target term and slam to #1. Because I was hyper-optimized for it. I don't care to explain further but it should make sense if you think about it.

The best way to correct an over-optimized anchor text profile is to build more links. But if it's completely bungled then sure, you could do what you've described by creating a new URL for the target term, de-optimizing the old one, and linking to the new one from the old. But that's only if the old one is so totally screwed there's no saving it.
Now this. This is interesting. U think similar to how I do.

Something they used to meme a lot but don’t anymore is google was a page algorithm.
 
@Ryuzaki Thanks for such a detailed write up. Not sure where you get the time or energy to write up these beasts as well as manage your own sites. Though I'm not sure why you call it throwing everything but the kitchen sink at it. Surely it's "throwing the kitchen sink at it"?

Anyway...

I'm obviously biased because I have a "top10bestproductreviews.org" site, yet I have to disagree with this:

Or you're a spammer in Google's eyes. Yes, some sites survive, so what. 99% of them did not.
Far more than 99% of these sites survived. For every one of these so-called spammer sites that got whacked in December, one got rewarded. The overall notion that Google can truly determine what is good content (especially in the AM space) is demonstrably false.

I can point you to countless examples of absolute garbage replacing good content, half decent content, or even just slightly less garbage content. Sites with mixes of ToFu Mofu BoFu percentages. Sites with ATF ads.

Google's nest of algorithms clearly still uses certain metrics as proxies for content quality, without actually getting it "right".

But, I guess that's the point.

If you address all of the things above, you should, in theory, clean up the metrics that Google has weighted more heavily this time around.
 
This is a great list I agree with everything here except for this

"Stop worrying about silos. Just interlink every chance you get. This also helps reduce click depth. Interlink in the main content contextually, do it with a static list of related posts at the bottom of the post. Do it out of your footer and sidebar."

Following the recent updates, a lot of the big interlinking sites got hit like gearhungry which yes had many other factors but one big one was they interlinked to everything, for example if you found a page on people camping and they mentioned one time in the morning they brushed their teeth that page would be linked to "best toothbrush".

I personally have been switching over to the kyle roof method of reverse silos where he daisychains his supporting content and only interlinks to very relevant content. Matt D mentioned this aswell in his algo update video. I have seen some good boosts from switching to this.

Great list, amazing content as always RYU
 
Following the recent updates, a lot of the big interlinking sites got hit like gearhungry which yes had many other factors but one big one was they interlinked to everything, for example if you found a page on people camping and they mentioned one time in the morning they brushed their teeth that page would be linked to "best toothbrush".

I think he meant this for niche specific sites. If you have a site about dogs you can link to any page and it will be relevant.

It may be different if you have a pet site and you link from a dog article to a cat article.

Your example is different because those are "multi-niche" sites that talk about pretty much anything.
 
You can also be careful how you interlink and instead of exact match in a paragraph, do a "Also Read:" link.

This is the golden standard of larger publications and is better for usability. When I see exact match interlinking, I've come to view it as spammy personally.
 
I've literally been consolidating all my processes into a master 'Why It's Not Ranking' checklist and almost everything here was on it! I found a few nuggets that I've now added.

Thanks!
 
12) Content Pruning
This is another big task where you'll decide what content is not high enough quality to be in the index. By deleting low quality pages, your quality score will rise. You will NOT decide based on traffic or links. You'll decide based on quality and if the content is serving the intent of the queries. Improve them, delete them, or 301 them. You probably want more feedback about this so please ask here in this thread.
Ok, but isn't quality in this case what Google thinks quality is and, if that is the case, shouldn't we look at traffic as the deciding factor? For us small site owners I'm afraid that would wipe out a large percentage of our pages, so the question is: how do we improve pages with no traffic so that they start getting some (change keyword focus of posts to similar but lower competition keywords- KGR even-, TF-IDF, last modified dates...)? Or is that a lost cause and we're better off spending our time planning and producing new content?

Great, very helpful post!

And, in the case of pages we think should go anyway, what does the process look like exactly? I know how to delete a post in Wordpress but how to tell Google not to crawl it so that our quality score increases?
 
Ok, but isn't quality in this case what Google thinks quality is and, if that is the case, shouldn't we look at traffic as the deciding factor?
No, quality != traffic.

If that was the case then CNN and FoxNews and all these "news" sites should delete 99.9% of their content that no longer gets organic traffic.

A post can go viral on social media and get 10 million views and never rank organically because it doesn't have on-page SEO done to it or even cover a topic people are searching for.

"10 funny gifs from times dogs dropped their tennis balls down the stairs" isn't getting searched for but can be extremely high quality and pull tons of traffic.

Forget about social media and all that stuff, or even any traffic at all. If you wrote a 5,000 scholarly essay with 10 unique images on "The Rise & Fall of the Staples Easy Button" that was chock full of entities, factoids, history, data tables, graphs, charts, etc... and nobody in the world but Google ever saw it because you didn't do on-page and didn't publish it on a domain with any page rank power and didn't share it on social media or even tell your friends... is Google going to think it's low quality? I would say "no".

Because you can look at the flipside of that. There are sites doing millions of organic visits a month from Google that routinely get deindexed because they're low quality. Like database sites, black hat spam projects, etc. If they get traffic, and that means they're high quality, then why is Google deindexing them?

And, in the case of pages we think should go anyway, what does the process look like exactly? I know how to delete a post in Wordpress but how to tell Google not to crawl it so that our quality score increases?
You do NOT want to tell them not to crawl it. That would involve restricting access to the pages in the robots.txt file. But if you do that, then how will they crawl it to know that the page is throwing a 404 (gone) or 410 (permanently and purposefully gone) HTTP status code? That's what you want to tell them. That the page is gone, and thus they should deindex it. Or give them a noindex directive, which also requires crawling.

It's not about not crawling, it's about deindexing. All pages in the index are evaluated to form your indexation quality score. To deindex a page requires crawling. There's nothing wrong with having 404's and 410's on your site.

But if you restrict crawling, you'll end up in a worse situation than having low quality content in the index. You'll have blank pages in the index which will absolutely tank your quality score.

It's like Google is walking down a hallway with 3 doors. They open the first 2, write down that one is a bedroom and one is a bathroom, mention what furniture is in the room, what it's used for, how clean it is, etc.

Then they reach the 3rd door but you've locked the door with robots.txt. Their job is to tell users how many rooms (pages) there are on your site, but not necessarily what's in them. So they mark down that another room exists and index it as blank (unknown contents).

Maybe your jokester buddy snuck in the house and hung a sign on the door that says "dlangpap's barbie collection". That's an anchor text. Google will trust that and give your blank page a title of "dlangpap's barbie collection" and say "no meta description available". And on their side there's no content behind the door.

That's a zero quality webpage and what you get when you block crawling. Blocking crawling will make the problem infinitely worse.
 
For content quality prunes I take say posts that have got less than 10 Google organic referrals in the last 30 days and then x-ref those against pages that got less than say 30 views from any source in the last 30 days. If a post is on both lists it gets deleted.
 
Hey @Ryuzaki thank you for this post. Very helpful.

My ecomm site that contains a blog I got hit hard with the latest update and have been working to recover. Crossed many of the items off the list here over the past few months and what's left is what I consider lack of authority - links from seed sites and their closest network as you mention.

Question for you: today the content with the exception of an article or two are written exclusively by me. Author bio and all that, but no diversity in authorship + minimal authority. It's a health related niche, so my gut was telling me in addition to authority links, I should be diversifying my authorship, getting articles from different writers, especially credentialed writers who I can highlight in their bio.

Thoughts on this moving the needle?
 
I am 80% through the list.
I am at 118 posts and 600 images and this is hard. Can't imagine doing this for a 1000-post site. Random thoughts:
  • Tried Link Whisper to speed up interlinking after hearing it being so hyped up by everyone and their mom. It turned out to be useless. It tries hard to find word clusters that seem like titles of other posts on the website but not once did it guess correctly. What it did was miss a lot of linking opportunities. Deleted it and interlinking everything manually. Tough beans. Promising myself I'll always interlink before publish from now on...
  • AAWP produced a JS error on some pages. I sent the error to support and they asked for admin privileges. Respectfully declined and troubleshooted it myself. Turns out when there's no AAWP box on the page the affiliate link tracking feature does a hiccup for some reason. Turned it off and now I'm error free.
  • For the first time in my life, I have a disavow list. I thought it was only for the nerds, but here I am.
 
Back