How we have recoved almost 100% from Last Penguin Update.

[newguy28]

Our site was hit pretty hard with penguin some 3-4 months ago. We went from #4 - #94 overnight. It has been absolutely killing our business. We do High Quality Web Design, Web Application and Software development and have been down to almost no business since we were hit. We regularly got $20k+ web development/software Projects via SEO.

Being a Graduate level software engineer , I always take an engineering approach to SEO and have been quite successful in VERY competitive local niches.

Some of the the stuff I see here about SEO and what Google does/can do/are able to do are just ridiculous. There are very few "manual Penalties" and most 90% are just algorithmic. There are billions of sites on the net and my little site about web development will NEVER get goggles attention enough for a manual penalty. Google is a software company, just like any other software company they want to do things with software. ie Low Cost and Automatic.

Google is simply trying to serve the best possible results for their users, and sometimes SEOer pervert that, so they make adjustments(Penguin)
.
You have to remember that Google wants to do things as efficiently, fast and as low cost as possible. That is the whole point of software.

So looking at our site from an Engineers perspective I was able to correct and RE-RANK on page one in 1-2 months.

This is what I did

Asked Questions, Alot..in this forum and others. I watched ALOT of you tube videos on penguin recovery and such, using my engineers mind i filtered through what I felt Google did. From That I figured these things out from 3 months of research and Trial and error.

1) The penguin went after those who google felt were actively SEOing their site, that means UNNATURAL BACKLINKS. Noone at google is going to waste time looking at every single page, although Google has the ability to write software to do this, it is very heavy on processors, therefore time consuming so they will look to save time by just trying to come up with an algorithm that will stop most cases. This is why they went after those who's Anchor tag profile was not diversified. Google has figured out that very few real and natural backlinks actually come with exact match anchor text. Naked URLs and Generic ones such as "Click Here" are much more prevelant. From my research I feel that 20-30% of your backlinks should be exact match anchor text. One of the easy fast ways we fixed that was to change the "[Tampa Web Design] by [APN Software]" link we put on all our finished projects to just [Web Design] By [Apan Software] that diversified our links by 50% right there.

2) Social Signals are becoming important and I feel that in the next update they will be even more important. Google has to find other metrics to rank sites, remember that Google simply wants to serve up the best companies/sites to their users, Facebook, Google reviews etc are an excellent way to do this. We linked our Google Plus page to our main page, with a URL in the G+ page and a direct link to that G+ Profile on the bottom footer of our site.

3) We DECREASED the keyword density of our money keywords by 1-1.5%

4) We INCREASED the amount of text on the main page.

These things have brought us back from a #94 ranking to where we are today #7, in 2-3 weeks we will be definitely back to #1-#2


Thanks!

 

[blackbeans]

Do you think more Google Circles (Google Plus) shares can boost rankings?

 

[newguy28]

Yes I think that there is an aggregate algorithm that checks social signals...I think one particular thing isnt going to help, just a combination of facebook, google plus. Just having the page linked helped us. We do not have many shares, likes, etc.

 

[SEORasta]

Thank for the insight but anchor diversity has been talked about for close to 2 years. If others are not doing this already then they will never undestand. ALso per Google (if you belive or not) G+1 will NOT help or hurt rankings. Now using more then just G+1 like Retweets, Likes etc may help but alone G+1 will not help you rank. My 2 cents is the #1 factor to ranking is having an perfect or near perfect on page performance. That alone will get your foot in the door while you use SEO to squeeze you whole body the rest of the way in

 

[newguy28]

Keyword diversity wasn't that much of an issue as it was in the last update. There has always been an issues with keyword diversity, but this late update focused much more on it.

Another thing we did, but I do not have any evidence that it help was make sure the home page was W3C Compliant.

 

[ficfroc]

Op have you been building any new links ? Have you also added any new content to the website ?
Thanks in advance

 

[sn0rt]

OP, you didn't happen to go to school for software engineering by any chance, did you?





On a more serious note:

Good insight OP. Thank you for sharing with the rest of the community. People have the impression that google is something similar to the muscleteen-- an all-knowing being. That's simply not true, Google is simply something coded by humans. You know what that means? There will always be a flaws and ways to strategically make use of them.

Wish you all the best of luck.

 

[newguy28]

Op have you been building any new links ? Have you also added any new content to the website ?
Thanks in advance

Yes tons....I almost tripled the amount of content on the first page.

 

[newguy28]

Yes I have a M.S. in Software Engineering, some of my ex classmates now work at Google. Google is not a "All Knoing being" you are right. The stuff I see here that people say Google does borders on Non Domain Based Artificial Intelligence, which is ridiculously hard. For example, it is Much easier for you to program software to intelligently do one thing, such as diagnose a medical issues. But to have one that looks at every page and intelligently tell you if the links are real, what exact category it is in, if the text is useful etc, all that stuff is YEARS from becoming a reality. It is extremely hard to do and slow as hell. They could only do a crawl once a year...lol. So what they do is come up with rules...which is much easier, and they can faster process pages. We just have to figure out what those rules are.

 

[nam6641]

Google is worth $300 Billion. You think they got there by doing what was "easy"? Their algorithm has so many variables and they have such computing power that it will never be fully understood by anyone here. As an example, unnatural backlinks, you think this is merely based on keyword frequency of the anchor text? They look at the content on the page also, they look at the username of person posting it (if applicable), they look at when it was posted, what other outgoing links are on the page, what incoming links link to the page, etc... They'll catch unnatural links, linkwheels, spam links, etc...
You may be able to game the system for a few weeks or even a few months but they'll catch on. This is why you never see the same sites staying at the top of the rankings for ultra competitive terms. High churn and burn cuz they get caught frequently.

 

[newguy28]

Here we go, "looking at the username of the person posting". That's simply too processor and memory intensive, i HIGHLY doubt that happens. You are thinking about one single site, now think about how many usernames posting things there are in the vast universe of the internet, in all countries. There are probably a BILLION. No way Google can keep track. I can give you a pass on page content.

All of this complicated stuff such as this is what SEOers use to make ourselves seem more "needed" and "valuable", which is ok when we are talking to clients, but come on man.

 

[newguy28]

They catch unnatural links because they fall into a category where most spammy links would be in. For example "over 100 links on a page" or "unnatural content". Google has to mass categorize things or they could never efficiently search, archive and serve up the net.

Even if google tried to decipher page content and categorize it from the text on the page it would take them years to crawl the net. The faster and better Google gets at crawling the larger the web grows, so they cannot tackle it in that fashion. They have to do it on a Macro level, categories, rules. They cant look at every case separately.

 

[TheMasterOfMoney]

 

[Rank 360]

This is not a simple process really awesome job,,, Congrats:withstupi...

Keep up the work to rank first in the SERP...

 

[sn0rt]

Yes I have a M.S. in Software Engineering, some of my ex classmates now work at Google.

I had a feeling you were.



Lol, I was pulling your chain a little bit because I think it was mentioned a few times throughout the thread. Good post nonetheless.

 

[the-kashish]

3) We DECREASED the keyword density of our money keywords by 1-1.5%

You decreased the kw density by 1-1.5% to ?

 

[timmy79]

Thank, very insightful.

 

[foodtrip]

Great post. Just wondering since you are an engineer, what do you think of the image search option by google, the one where you can literally drag an image file from a folder to the google search bar and google shows several pages that uses that image.

Is this "simply too processor and memory intensive" as you said? If yes this means google has some serious computing power, if not then good for us.

I just found out that from a thread here recently and it amazes how google does that.

 

[ttmschine]

Their algorithm has so many variables and they have such computing power that it will never be fully understood by anyone here.

I doubt it's fully understood by anyone there either.

 

[nanavlad]

This makes perfectly good sense
Thanks for the info