Journal SEO Contest: How I Won Several Top Prizes...

Discussion in 'Case Studies and Journals' started by anthony80, Jul 22, 2016.

Share This Page

  1. anthony80

    anthony80 Member

    Joined:
    Mar 11, 2016
    Messages:
    11
    Likes Received:
    13
    I've been asked to share my experience and techniques from the SEO contest where I won several of the top prizes. First, I'd like to congratulate all the SEO's that competed-- it was a good and a challenging battle! I'd also like to specially thank this forum and the sponsors who made the contest possible. A Big Thank You to all the Sponsors:
    [​IMG]

    My SEO Contest Results:
    I feel honored to walk away with several of the top prizes from the contest. However, the biggest prize for me was being able to rank the main keyword higher than anyone else in the contest and to have ended with most days ranked in Top 5. Other prizes or awards included Most Traffic and Best Final Performance.
    [​IMG]

    Who am I?
    On this forum I chose an anonymous username of anthony80 but you may know me by spaxton1 or Scott Paxton. Since I do a lot of work and consulting with top SEO guys; I felt it smart to not place an SEO target on my back for this contest; thus the anonymity. I've been doing SEO full-time since 2006 and I'm very passionate about it. I love SEO!

    Most of my income comes from ranking my own sites and brands. However, I also do a fair amount of consulting/ coaching with SEO Firms, national brands and top SEO's. I specialize in big competitive niches; but also have a handful of local SEO Clients.

    I'm best known for my proprietary SEO software. I've invested nearly 1/4 million dollars into various SEO applications which gives me a slight competitive advantage in ranking organically. For this contest I relied mostly on: Serp Avenger, Serp Code and PushbuttonPBN.
    (Disclaimer: My software is exclusive and I only open it up to my waiting list every 3 to 6 months)
    [​IMG]

    Enough about me... let's jump into how I ranked!

    The SEO Contest
    In case you missed it; the contest was open to anyone in the SEO community. Participants had to "register" before the deadline and agree to the terms set by this forum and the sponsors. The nuts and bolts of the contest:
    1) We were all provided the same keyword to rank
    2) All participants must register a brand new domain
    3) All participants must keep private their site and the keyword until the end of the contest.
    >> any and all methods to rank were considered fair game.

    The Keyword
    First, let me comment that no one participating in the contest had any say over the chosen keyword. We were at the mercy of the contest. The keyword provided was: Club Penguin Walkthrough
    My first impression: What??!
    To this day I'm still not sure why that keyword was chosen. I suspect it was a play on the 2012 Algo Update known as: "The Penguin Update" which turned the internet upside down and caused SEOs big and small to hate the little black and white critters.

    Besides the obvious animosity most SEOs have against "Penguins;" the keyword itself was an odd choice. If a client approached me with this keyword I would advise them not to waste time or money on it; because the cost wouldn't justify the results. With that said... anytime you throw a big group of active SEO's at any phrase it gets competitive and interesting. ... still I would have like to rank for something "bigger" (more on that later)

    My Domain

    I find domains don't matter much at all. However, I still prefer top level domains such as .com, and .net. Before picking the domain, I spent a couple days researching the niche "Club Penguin." I found that the game was broken into "Missions" so I settled on the domain: walkthroughmission.com

    Side note: I intentionally didn't use "Club Penguin" anywhere in the root domain as that is a Disney Trademark which could have resulted in DMCA Takedown for trademark infringement. Even ruthless competitors will fake DMCA Takedown request to knock out competitors. Generally it's not worth the risk.

    My Site
    Step 1: Before I built my site I first wanted to learn everything I could about the sites that were ranking currently. I dove fairly deep into site structure, content, inner linking, keyword optimization and keyword ratios. To speed this up I ran an onpage analysis of all ranking sites using my SerpCode.com software (this can be done manually).

    My Analysis helped me identify a patterns in site structure and content that seemed to be favored by Google for this phrase. Here are the main takeaways:
    Inner Page Ranking: 100% of sites ranked were "Inner Pages" and not homepage.
    Inner Linking: Top Ranking Sites shared a similar inner linking scheme. Main "Walkthrough" Page >> linked to individual Mission pages.
    Keyword Optimization: Exact keyword was found "blended into phrases" rather than stand alone.
    Content: Keyword score was low within content.
    Additional Optimization: Broad Keyword phrase match was common within Tags, Metas, Images, and Headings

    Inner Linking Structure:
    From this data I planned out a simple site layout that would best match what google was looking for:
    Homepage >> Link to Target Page http://www.walkthroughmission.com/club-penguin-walkthrough/
    Target Page>> Link to Mission Pages Mission 1, Mission 2, Mission 3 and so forth through Mission 11.
    Mission Pages>> Link back to Target Page: Club Penguin Walkthrough Tips
    (Fairly simple inner linking set-up... but required a lot of content and image creation)

    Page Content and Elements:
    Based on the pattern of currently ranked sites; I noticed most had "mission" optimized images followed by "mission" optimized snippets of content. So on my Target page (the page I was trying to rank) I created an image for each of the 11 missions and optimized each image tag for its related mission. (Note: I also optimized image URLs) I then created a paragraph or two about each mission

    Image Optimization: Optimized for "missions" not for target keyword as part of inner linking plan
    >> Alt Tag: Such as Mission1
    >> Image URL location: Such as /mission1.png
    >> Image OCR: (Scanable words within image) Such as: Mission 1 + Title of Mission

    Content Optimization
    : Optimized for "missions" not for target keyword
    >> Content Keyword Ratio: Exact keyword phrase was not used within content. Synonyms were uses within content.
    IE: tutorial instead of walkthrough
    >> Titles and Metas: The "Exact Keyword" was used in phrases throughout, but not by itself. Added to: Meta Tags, Titles, and Inner Anchors
    Note: This keyword had a very low keyword ratio; so I took steps to not "over optimize" for the "exact keyword."

    Step 2: Designing the Site
    Nothing special here. I created a basic "helpful" fan site. There was no point in making the site pretty as it has no real effect on rankings (other than bounce rate-- but that's a topic for another day) The site was built using WordPress; mostly because its quick and easy to setup, so that's what I used. The main goal was to get the site live and indexed as fast as possible.

    [​IMG]

    Indexing The Site
    Normally I don't worry much about indexing; as it should happen naturally. However, in a 12 week contest everyday counts. There are several methods to index a site. Pinging the site is the most common; but I prefer sending "traffic signals" to the site as it seems to be the fastest method. Throughout the day I trickled in about 12 "Natural Traffic" signals. When I checked back later; the site was indexed.

    The SEO Plan

    During the first week I completed all the onpage SEO work and content. I took the time to research each of the 11 Club Penguin Missions. I'm sure I could have cut corners here; but I was treating this project like I would a paying client. I wanted to learn everything I could about the niche so I could be more competitive. ... and for the record; I'm still not a fan of penguins!

    Off-Page SEO
    When it comes to offpage SEO... there are many techniques and methods. I've been doing SEO full time for over 10 years now and I've seen a lot of trends come and go. In this contest I saw quite a few people using pretty outdated and short-term methods to rank. Which I guess can work well in a short-term SEO contest; but it wouldn't make sense for real world SEO (at least not in competitive niches).

    Monitoring the Competition
    Its clear Google uses variables within its algorithm. Meaning that what works in one niche might not work in another. Google is clever and knows these variables confuse new SEOs (or lazy SEOs) who use a "One Size Fits All" approach to SEO. When analyzing offpage elements; I look for patterns until I understand what variables Google is using for the niche. Once I have that information I can create a linking plan that will get results. I never build a link until I fully analyze and understand the variables Google wants.

    Analysis of currently ranked sites:

    For the analysis I used SerpCode.com to look at several backlink patterns. I wanted to see things like: Anchor Ratios, Link Types, Social Signals, Redirects, External Links, and Historical Data. This helped me create a link plan that was based off of what the current algo was looking for.

    Anchor Ratios

    For this niche I saw a higher than usual "blended" and "related" anchor ratio. But more importantly I saw a low "exact match" anchor ratio-- which is common since the Penguin update. Branded and raw url anchors were nothing of note either.
    Exact Match = Keyword Phrase IE: club penguin walkthrough
    Blended Match = Keyword found inside other words IE: this club penguin site has good walkthroughs
    Related Match = Related or Synonym Phrase IE: club penguin tutorials
    Branded Match = Brand Name IE: Penguin Boy 007
    Raw Match = Plain URL IE: walkthroughmission.com
    >> the plan: build links using a mix of mostly Blended and Related Match anchors.

    Social Signals
    Rarely do I find social signals factoring much into rankings. For this niche some sites had decent social signals but not enough to create a pattern of importance.

    External Links

    This can be debated all day long about whether this helps or not. However, the analysis did show a pattern. Ranked sites were often linking out to related sites in the niche. I find linking out to related authority sites can help with initial relevancy; but isn't much of a signal to make a big impact in rankings one way or the other. For this contest, I linked out to the official club penguin site from the homepage.

    Link Types
    I looked at several types of links and saw that image links were more common in this niche. I built in "image links" into my linking plan. Other link types such as site-wide links, and redirects were not important to sites currently ranking.

    Other Signals
    Many other signals were looked at and considered; however they were of minor importance.

    Continued Serp Tracking and Analysis
    I think its important to note that constant analysis and adjustments to the SEO plan is important. Google is constantly updating and changing which ranking factors have more weight. Understanding and adapting to those changes in real-time can protect rankings. For this contest I paid close attention to new sites that started to rank in top 100. But even more vital I paid closer attention to sites who were dropping in ranks. I tried spotting patterns in newly ranked sites to see why Google was favoring them. I also looked for negative patterns in sites that were falling in the serps. In a nut shell, I looked for favorable signals and avoide the negative ones. Serp Tracking was set-up through my software.

    To find which sites were from people in the contest; I simply checked their domain registration dates. If the registration date was after the start of the contest I knew the site was most likely from another participant in the contest. I kept a list and watched participants closely. I cross referenced the weekly contest results with the serp results to try to identify what site belonged to what user.

    The Link Plan
    After I analyzed both onpage and offpage elements it was time to implement the linking plan. Moving forward I simply had to check my daily serp tracker and spot any issues or concerns in my current plan and adapt best I could. I felt pretty confident with the current plan and was ready to jump into the exciting part. Getting Results!

    The Strategy

    When it comes to link building; I like to look at the big picture. I discovered long ago that "short-term" SEO makes no sense. While some people believe in churn and burn; I don't. Most of my sites (or clients) sites cost 10's of thousands of dollars to build and can't afford low quality methods that can lead to penalties or hurt existing traffic. Since I was treating this contest like I would a paying client; I decided I would build out a custom link network consisting of relevant sites. The plan was simple; find and acquire sites in the niche and have those sites link back to my target site.

    In the real world; this strategy makes more sense. But in the "Club Penguin" world where its practically impossible to monetize sites; the number of "related sites" in this niche proved difficult to find; but not impossible. Using my PushButtonPBN.com software I was able to find several club penguin forum sites. But it wasn't enough... so I branched out and found "Online Game Related" sites.

    Throughout the contest, I continued to find authority sites and rebuild them. Once I acquired a new site I waited for it to index and then I used the site to link back to my contest site. I carefully considered the anchors for each link I built and kept within the anchor ratios explained previously. Due the non competitive nature of the niche; I noticed that many of the Club Penguin Sites I found were low quality and not worth using.

    Final Thoughts
    For me SEO is about pattern analysis. If you can identify onpage and offpage patterns you will win and outrank your competition. I love SEO and hope some of this has been helpful to you.

    PS: If you're interested in any of my proprietary SEO software; join my waiting list:
    http://serpcode.com/


    PPS: If you'd like me to do SEO for you; add me to Skype and let's chat: My Skype ID: spaxton1

    Thanks to everyone that participated! Best of success to you all!

    My Best,

    Scott Paxton
     
    Last edited: Jul 22, 2016
    msharron, John Stover, red and 5 others like this.
  2. red

    red New Member

    Joined:
    Jan 2, 2016
    Messages:
    1
    Likes Received:
    1
    Scott-

    Congrats -

    This is excellent - I would love to get in on Serpcode - we do analysis manually and it's a PITA! PM me if your allowing new memberships! thanks
     
    anthony80 likes this.
  3. Wazza

    Wazza Active Member

    Joined:
    Mar 6, 2016
    Messages:
    100
    Likes Received:
    46
    Great write up! Thanks :)
     
    anthony80 likes this.
  4. JimKata

    JimKata New Member

    Joined:
    Feb 24, 2016
    Messages:
    13
    Likes Received:
    7
    Very cool -- thanks for sharing!

    What do you mean by "find and acquire sites in the niche and have those sites link back to my target site."?

    Buy existing sites? Dropped domains? Create copy-cat sites with similar content?
     
    anthony80 likes this.
  5. Paul

    Paul Member

    Joined:
    Dec 15, 2015
    Messages:
    39
    Likes Received:
    41
    Excellent explanation! I can see why you finished in the top 3.

    Quick question about: '>> Image OCR: (Scanable words within image) Such as: Mission 1 + Title of Mission' Is there anything specific you do to make the words scanable?
     
    anthony80 likes this.
  6. anthony80

    anthony80 Member

    Joined:
    Mar 11, 2016
    Messages:
    11
    Likes Received:
    13
    Thanks for the replies. I'm not real active on this forum; so I apologize if I miss some responses in the future.

    Sure, I created SerpCode.com for that very reason. Its a pain to manually research each site to understand what the algo is looking at. Anyone who uses any of my software knows that's its pretty exclusive. I open up to my email list every once in a while (usually every 6 months or so). I'll probably open in another couple of months; but even then it usually sells out in a short time. ... the logic for this is that I want my software to remain effective so I limit the number of people I allow to join. I make a lot more money doing SEO on my own sites than I do from selling access to my software; which is partly why I don't care to promote it much. But it is helpful and makes life much easier.

    I can't save spots for anyone; but if you join my email list I promise to email you once it opens again. You can join my email waiting list here: http://SerpCode.com

    Thanks! I hope it adds some value to your own SEO work.

    Another good question. Yes, Yes and no. There's really no wrong way to acquire sites whether that's from buying existing sites, auctions, drop catching, domain buying, etc. They're all fair game; with exception to the copy-cat sites (that's not something I'd recommend). There's a bit of technique here and experience will help when it comes to acquiring the right kind of sites. Authority, Age, History, Content, Traffic are what I mainly consider.

    In most cases this probably won't make a huge impact on rankings; but it can help. I like to research Google Patents to get a better understanding of Google's technology and practices. A fairly recent Google Patent included an interesting insight where it says it might use "OCR" to read text in images. (by the way, these little insights can be extremely important for anyone doing competitive SEO.)

    With that in mind; Google can and does scan images deeper than most people think. This is why using images that are on lots of other sites can hurt rankings. Duplicate images, where the same image exist in other places doesn't seem to be a major issue yet, but it can be in the future. When possible unique images are best practices when creating a site. All images for this SEO contest were unique and not duplicated.

    So back to your question...
    To make the words scan-able, just use high contrast from the background (For example, lighter text will give high contrast on darker images. Darker text will give high contrast on lighter images) Make it easy for Google as they do have limitations as to how they can decipher text in images.

    A side note to consider... conversion rates are more important than the slight ranking benefit this can give. So consider how the images are going to impact the user experience on your site. There are plenty of ways to optimize images without worrying about the OCR stuff-- it can help; but it should never be at the cost of losing conversion points. ... this can be easily split tested.
     
    Last edited: Aug 20, 2016
    JimKata likes this.
  7. JimKata

    JimKata New Member

    Joined:
    Feb 24, 2016
    Messages:
    13
    Likes Received:
    7
    Thanks for the info. How many backlinks did you create in total?
     
    anthony80 likes this.
  8. msharron

    msharron Member

    Joined:
    May 7, 2016
    Messages:
    25
    Likes Received:
    13
    plain old match and exceed. awesome write up.
     
  9. anthony80

    anthony80 Member

    Joined:
    Mar 11, 2016
    Messages:
    11
    Likes Received:
    13
    It was less than 100. I'm a big believer in "less is more." The exact number of links was somewhere in the high 80's going off memory. Since the contest ended I've since stopped doing SEO on this site and plan to just let it die out. Many of the links have been removed or "dehosted." There's not much I can do with the site as its not worth monetizing and it doesn't make a good flip.
     
  10. JimKata

    JimKata New Member

    Joined:
    Feb 24, 2016
    Messages:
    13
    Likes Received:
    7
    Can you explain ""Natural Traffic" signals"?
     
  11. anthony80

    anthony80 Member

    Joined:
    Mar 11, 2016
    Messages:
    11
    Likes Received:
    13
    Sure. When I talk about traffic signals I'm referring to mimicking real vistors. I've tested this quite a bit and found that traffic signals can be used for several things. In this test I used them to get the site indexed faster.

    Now in order for traffic signals to work Google has to be able to detect the traffic through one of their properties such as through Google Chrome, Google Search, Google Fonts, Gmail or even 3rd party services they own or control.

    Dont quote me on the date but somewhere around 2013 when Yandex (Russia's most popular search engine) announced that in the Moscow area they were going to rank sites entirely off of traffic signals; Google came out and said they tried that and found that links were still providing the best results for ranking sites. ... This is when I started testing traffic signals as it told me Google was considering them to some degree. Still today traffic signals can help in many scenarios. In fact this is why often sites will see a temporary ranking boost after sending emails out to their list.

    We could discuss all day how traffic signals can be used. But in this scenario I simply used them to get my site indexed faster. I did use them a little on my linked pages too; again mostly for indexing faster but I also used them to boost some of my links authority.

    Side note: be very careful with traffic signals. Google is good at detecting fake traffic signals which can give a negative effect. ... For example I've found that using "data center" proxies will hurt. But those are easy to detect so it makes sense that Google would use them as a fake signal.

    Not that I do much with Yandex... But it's interesting to note they've quietly gotten away from using only traffic signals. The results were worse when they used them because once everyone knew about the change they were quick to find ways to game traffic signals.
     
    Golan likes this.
  12. msharron

    msharron Member

    Joined:
    May 7, 2016
    Messages:
    25
    Likes Received:
    13
    so...at a guess plugin wmt/analytics.

    twitter/facebook/Gplus =3
    niche forums = 5+
    social sponsored posts/ppc = 2+
    misc social platforms/ppc = how long is a peice of string.
    niche blog commenting
    you could even use some traffic exchange platforms mech turk or bots to simulate user interaction.

    am I close?
     
  13. msharron

    msharron Member

    Joined:
    May 7, 2016
    Messages:
    25
    Likes Received:
    13
    Side note: be very careful with traffic signals. Google is good at detecting fake traffic signals which can give a negative effect. ... For example I've found that using "data center" proxies will hurt. But those are easy to detect so it makes sense that Google would use them as a fake signal.

    lol.....yeah found that out the hard way.
     
  14. JimKata

    JimKata New Member

    Joined:
    Feb 24, 2016
    Messages:
    13
    Likes Received:
    7
    Ok, you've explained *what* they are... but are you intentionally being vague with respect to how to generate them?

    Other than paid ads, what's your preferred method to generate "natural traffic"? Thanks.
     
  15. anthony80

    anthony80 Member

    Joined:
    Mar 11, 2016
    Messages:
    11
    Likes Received:
    13
    There are several services that provide traffic signals. However, I found them to vary in their effectiveness and often cause more problems than its worth. This is mostly due to how they function. For example the "crowd" sourcing that is used for them has inherent problems. One of the bigger issues is that many encourage proxy usage. There's nothing natural about getting traffic from proxies registered and owned by hosting companies. Those IPs are intended for web hosting, software, etc. ... its very easy to detect IP ranges that aren't "natural." This is why using the wrong service can hurt.

    I do admit there is reason for being somewhat vague here. I've spent a consider amount of time and money investing in my own proprietary solution for this. Currently I use it strictly on my own sites and for my client sites with no plans to release it publicly. Without getting into specifics; my own service which I call http://SeoTrafficSignals.com is too costly to scale into subscription based tool. When considering the value it offers to rankings, its helpful for ultra competitive keywords but not something most could or should justify. Its something I use internally and will continue to as long as it remains effective.

    [​IMG]

    I know that's not really helping you out much; but some things are best not discussed on a public forum. What I can say about it is that Google is smart. If you remember back to when Google had some issues with "Fake Clicks" to adwords; you'll remember they created some pretty advanced technology to detect what they called invalid clicks. Based on my testing; I'm sure that same type of technology has been used to detect traffic signals.

    To do it correctly takes a lot of resources and creativity. Unfortunately, this is one area I can't expand on much more.
     
  16. Bender Bending Rodríguez

    Bender Bending Rodríguez Senior Member

    Joined:
    Mar 10, 2016
    Messages:
    507
    Likes Received:
    338
    I think its funny how the person who won first just built spam links and a PBN and called it good whereas the person who won third was worried about things that no one has heard of.
     
    vlachs likes this.
  17. anthony80

    anthony80 Member

    Joined:
    Mar 11, 2016
    Messages:
    11
    Likes Received:
    13
    What's more funny is how the 1st, 2nd and 3rd place point scoring worked.

    1st Place only ranked 15 Days in Top 5.
    2nd Place only ranked 3 Days in Top 5.
    3rd Place Ranked 32 Days in Top 5.

    At the risk of sounding a bit arrogant; I did rank higher than anyone else in the contest and I ranked in Top 5 Twice as long as anyone else. Not to mention I grabbed the "most traffic" award. I mention this because your response implies that I somehow didn't do as well as the guy that "spammed and called it good."

    In real world SEO; its how high you rank and traffic that counts. So maybe these things you've never heard of might matter. The point system favored short-term quick rankings and didn't have weighted points for top position. But it doesn't matter much to me because I took the awards that I wanted-- I wasn't doing this for the prizes (come to think of it; I still haven't gotten most of the prizes yet).
     
  18. sites

    sites Active Member

    Joined:
    Jan 2, 2016
    Messages:
    89
    Likes Received:
    75
    Well, "first place" ranked many, many days in the top 10, all alone there, and it took a really long time and some heavy negative SEO for others, including "third place", to catch up with him.

    So whether you calculate rank or traffic or actual revenue, vlachs won by a mile. Because, being on the first page matters.

    You have a point there. It would have been interesting to see what the long term rankings are. For the next contest, I hope we will have a medium/high volume keyword with high commercial intent.
     
    vlachs likes this.
  19. anthony80

    anthony80 Member

    Joined:
    Mar 11, 2016
    Messages:
    11
    Likes Received:
    13
    Sorry, but your comment doesn't match the weekly results. Using your "Page 1" logic both 2nd and 3rd place had more days on page one than 1st place. There is a big difference between being on "Page 1" and being Top of Page 1. I'm not sure how you "win by a mile" when you have less traffic/ revenue and lower rankings. Regardless; it was a good competition and I'm happy with the results.
     
  20. Bender Bending Rodríguez

    Bender Bending Rodríguez Senior Member

    Joined:
    Mar 10, 2016
    Messages:
    507
    Likes Received:
    338
    ... So the scoring didn't truly reflect your abilities, the contest is not the real world, and, in the real world, it's traffic that counts and you got the most of it based upon the traffic scoring system, which you won. Either way, you're satisfied so you arguing doesn't matter to you.

    OK dude. Congratulations. Let's move this on to something more productive, ok?

    Do you have any proof that text in images are being used as a ranking signal? It sounds like speculation to me. Even if they have a patent, it doesn't mean they're using the patent.

    Traffic signals, do you have any proof of this either? Yeah, I've heard people talk about it but, again, it seems like speculation to me. Even if it's being used, IMO, one wouldn't be sending fake traffic signals when one can just simply generate traffic to a site and get real signals.

    For example, running a weekly giveaway, where visitors gain extra points for referring friends, would generate exponential growth. No bots needed.