commit: 6edd53c - #922 (2014-07-29 19:00:33 +0100)
Source: Google Search Quality Team Drops the Ball - YouTube
Wow, didn't realize how angry I was getting at the end of the video. Sorry @MattCutts and Team, I still love you.
That's a very interesting video, and I imagine we've all had moments where we've wondered what on earth was happening to Google.
You didn't address this in your video, but could your client have received a manual penalty?
Thanks Ophelie. I monitor their GWMT like a hawk (no warnings, ever). There were a few questionable links (likely negative SEO), but there were quickly address and disavowed. I'm more concerned about the spam appearing ahead of the top brands (Gazelle, Nextworth, etc).
...this is new? Examples like this (multiple examples most of the time) are found in practically every competitive SERP.
Steve, I know it might sound weird, but I don't believe it's an issue that can be easily solved by the spam team. Buying and 'producing' links has worked for so long that is has fundamentally skewed the link-graph. I don't think that what you're seeing is really a result of laziness on the Google Search Quality Team's. It's simply a complicated problem they;re facing. They cannot tackle the problem manually. There's simply too much stuff out there right now.
They could turn up the knob on their spam filters, but that would mean:
1) more false positives (mistaken penalties for sites that don't deserve one)
2) penalisation of sites that are useful but have been using links against Google's guidelines (in some verticals if you throw out all the sites that have been bending or breaking Google's rules you're left with nothing useful in the SERPs)
3) making it easier to negatively affect other peoples rankings
So it's a bit of a catch-22.
You may have got there attention with your video and your client's competitors will now probably get a manual review and penalties, but that's about it. Google will keep introducing new filters and penalties but there will always be those that slip through. Like there always have.
You're right Nemek. I really just wanted to get their attention to the issue and to see if there's a reasonable explanation that I missed. Call it a learning experience.
If they find even one thing that helps cleanup results, it's a win for everyone, not just my client.
I hear you on the catch-22. By bringing these issues up rather than ignoring them we stand a greater chance of resolution. If you watch the video, you'll hear me compliment Matt and team more than once.
Then there's the whole "it's my job" to help my client restore their placement and displacing spam is a part of it.
Thanks a bunch for the feedback, it really means alot me. Have a fantastic weekend!
Steve, thank you for the video, check this: http://searchenginewatch.com/article/2267464/Matt-Cutts-Google-Penguin-2.0-Coming-in-Next-Few-Weeks I think Matt responded.
I saw that - thanks Slava. Keeping my fingers crossed that we'll have cleaner results soon!
Every scummy SEO seems to add Wikipedia links into their content. I genuinely don't think that's how co-citation works - surely you're supposed to link to the same place as the relevant Wikipedia page is linking to?
The remaining power of EMDs is that you can kinda get away with spamming your keywords in your URLs. So branded links.... don't exist for them.
1. Ahrefs ranking system is just a quick guess. I'd assumed they'd used some "machine learning reverse engineer Google's rankings" system that Majestic kind of uses (or did, I don't know how they could accurately give that information anymore). But they don't - couldn't find any info on how it's calculated when I looked last.
i.e. Some of their lame guest blogs like http://www.de-yapi.com/a-guide-to-sell-your-laptop-online/ might somehow be transferring the power of 100 link networks to their page. Kudos to the copywriter on that piece of awesomeness.
> At some point, whether it’s due to an accident or an update many of us wind up having a notebook that’s no further of any use to us.
This is compelling stuff. I'd spend as much as $5 on it.
2. Ahrefs, OSE, Majestic are not omniscient. You can probably block their crawlers quite easily if you want to build your own link network or whatever. Even without that, they can sometimes find less than 10% of your links - http://www.matthewwoodward.co.uk/experiments/backlink-checkers-compared-ahrefs-majestic-seo-seomoz-raven-tools-seo-spyglass/ - it's unlikely we're missing much though - I footprint scraped Google - nada.
3. They appear to be involved in some sort of footer links scheme. A lot of their SEO company's clients seem to have blogroll/"related content" links using keyword variant anchors. None of them seem to be facing the above site - actually they're linking away.
However it's just not a competitive a term. 720-1100 exact searches a month, CPC kind of low. http://www.quicklaptopcash.com/sell-used-laptop is ranking #2 for me, DA of 26.
Who is your client?
Thanks for the feedback - you're right, 3rd party tools are better at providing clues rather than realistic data. Great point!
This particular account is interesting because the "money terms" are all phrases you'd never use in context, such as "sell laptop" or "sell iphone". Using the word "your" hoses the account with each attempt to create on-page awesomeness. Bizarre.
I don't think I'd pay $5 for that content, considering it will hurt more than help. :)
As it turns out, the semantic derivatives of the example term provide 80% of the traffic, but they are all very similar and do actually add up to quite a bit of traffic if you refer to the screenshot at the beginning of the video.
Let's just hope that shady blogrolls have finally seen their day with P2.0.
By the way, I saw that outbound linking to Wikipedia and Google+ too. Weird how an OBL could even be considered a positive ranking signal, especially, as you mentioned, the fact that it's pointing to the root of the domain, not to individual co-citation.
Thanks again for the input - it really helps clue us in on what these guys were doing.
Great! Head over to login and use your credentials there