Terry Kyle’s Big 60-Day Backlink Experiment

[Terry Kyle]

OK, here’s the 60-day backlinking experiment that I will be running and documenting in this sub-forum starting on Thursday 17 September 2009.

Hopefully this ambitious public experiment will help many Warriors here and you’ll enjoy the ride!

I don’t remember seeing anything like this on Warrior Forum before so it should be interesting and possibly sort out a few SEO/SEM myths from facts (and maybe prove that different Search Engine strategies can each achieve the same objective).

Starting on Thursday 17 September 2009, I will simultaneously run 4 different SEM (Search Engine Marketing) strategies on 3 different NEW GoArticles in the same niche plus one BRAND NEW website BUT with approximately the same search volume for the 4 different keywords.

This is not really an SEO experiment in the sense that I can’t do much with on-page factors like internal link structure and H2 tags etc as the three articles are examples of so-called ‘parasite hosting’ on GoArticles and the fourth will be relying on WordPress SEO power and Ken Fry’s SB101.

Why GoArticles?

GoArticles are particularly responsive to backlinking but I will choose keywords where there are currently no GoArticles on Page 1 of Google USA (though I live in London, UK) or at least no more than one at the moment.

Each article will target a different keyword within the same niche, be just over 500 words and use the exact keyword phrase 5 times. I will try to have a roughly similar search volume for each keyword (at least 100 searches a day) and a roughly similar degree of difficulty for ranking.

Why 60 days?

In my experience, it takes time to get serious search engine traction with a money site (some Web 2.0 sites generally fade out fast) and this experiment is about building a stable, high, long-term rank for each article.

4 different search engine strategies will be used:

Article 1

Normal backlinking using the backlinks of Angela, PJ and my own WSO (including only IRRELEVANT sites to niche) – 50 backlinks per week for 8 weeks (60 days), minimum PR 5.

Article 2

Backlinking only to RELEVANT sites, blogs and forums (are they of higher value given their topic relevance? Let’s see) – 50 backlinks per week for 8 weeks (60 days), minimum PR 5.

Article 3

SENuke Pro including Link Wheels with all feeder sites backlinked using the backlinks of Angela, PJ and my own. Some articles used in each of the Web 2.0 sites will be spun by Power Article Rewriter to an originality degree of about 40% (I usually hit about 43% judging by SEN). Other articles will be spun by Human Rewriter (which I haven’t used before but Jeremy Kelsall rates highly).

Though I’ve had plenty of success with my own backlinking methods, I have briefly trialled SENuke a couple of times in the past without much success – could have been the proverbial ‘nut behind the wheel’ - but this time I want to use this experiment to fast-track my learning curve on SENuke.

Helpful Warrior Tom Brite is my SENuke wingman (he has awesome SE Nuke buyer bonuses by the way) so hopefully we can all learn the nuances and insider tricks of SEN Pro during this 60-day experiment.

Wild Card – Test 4

I will also create a brand new site on Thursday, a dot com site with the search phrase in the domain name, install a WP blog as the main site (not in a sub-folder), change the Permalink post names section to be more SEO-friendly, add more RSS sites to ping, install Warrior Jeremy Kelsall’s Content Blender plugin (more on that later), Google Sitemap Generator, All-In-One Seo and Google Analytics.

I will create 60 highly spun versions of a few articles and set up the blog to publish one each day. More importantly, one inner page post URL (the first post) will be added to Warrior Ken Fry’s SB101 WSO which relies solely on social bookmarking backlinks. This site will get no other juice except from the daily posting of a seemingly original article (with cloaked aff links at top and bottom like the GoArticles), WP’s SEO-friendliness and SB101.

And...

Each GoArticle will have two cloaked affiliate links in the same call to action sentence, using offto dot net, one at the top of the article (permitted in GA) and one in the Bio Box.

Each article will direct to the same Clickbank product (each with a different Tracking ID), a product with a respectable gravity of roughly 100+ (probably) and each keyword should show as anything from “Easy” to “Doable” in SE Nuke and have a green or yellow SOC (Strength of Competition) result in Micro Niche Finder.

I’ll post screen grabs of these when I choose them. I will also choose a niche with universal relevance across the world – that rules out reverse phone lookup, dammit!

A new ClickBank account will be created for this experiment.

My secondary goal at the end of the 60 days (approximately 17 November 2009) is to sell, one, some, or all of the article accounts and the website here on Warrior Forum and/or at Flippa.

Value estimations on site flipping can be volatile but I will look to sell each article account for a minimum of 10x monthly revenue e.g. if one of the GoArticles is averaging $100 a day and $3000 a month, I will TRY to get $30,000 for it but let’s see what happens. That could be too ambitious.

I know that a GoArticle is not as attractive to a buyer as a standalone website and there are always fears about GA changing their rules (on affiliate links, for example) or Google suddenly slapping them (as they have done with Squidoo in the past) so my valuation is lower than the 12-24x monthly profit ballpark with normal websites.

Time will tell there.

However, website buyers are ultimately attracted to revenue, be it from a popular YouTube video or any other form of web property so hopefully that stage in this experiment will prove as interesting to you (and me) as the SEM stage.

By the way, I have absolutely no interest in promoting one method over another and am only interested in what WORKS in this ‘race’ between the 3 articles and the new website.

I will not be favoring any one approach and will be doing all four simultaneously over the 60 days with my outsourcer backlinking Article 1 (IRRELEVANT sites), me backlinking RELEVANT sites for Article 2, SENuke doing the ‘heavy lifting’ on Article 3 and SB101 carrying the hopes of a generation for the brand new website.

I make no income predictions whatsoever but let’s see how the race goes.

As the experiment unfolds, I will post screen grabs of GA article views, ClickBank hop stats, commissions, Analytics stats etc.

Hopefully 200 Warriors won’t jump all over these keywords when the experiment begins which would artificially distort the results of this test.

Being able to show the keywords, site, articles and progress is what will make the whole experiment worthwhile.

So, one niche, three GoArticles, one brand new site and 4 different SEM strategies.

Let the race begin and feel free to chime in with your questions or suggestions (parameters can be modified before the start) as we go…

More details soon as we get closer to the kick-off.

Stay tuned…

BTW, if you've just dropped by for the first time and feel like this thread is of value to you, don't forget to hit the "Thanks" button below on the right!

[Floyd Bogart]

Looking forward to the results...

[CalGolden]

Ambitious undertaking, Terry. Thanks for sharing this with the gang.

got this thread plugged into my tivo. Looks like it's goanna be a thrilling ride.






Cal

[Ross Kenny]

I'll be following this thread closely.

Nice to see threads like this instead of the usual general questions.

Keep it up

[remodeler]

This will definitely be interesting Terry and help a LOT of folks along the way. Looking forward to following along!

[Yuds]

Very interested to see how this turns out.

Thanks so much for doing this!

[olamilekan2]

Brilliant thread, i will be following along, this will definitley help many Warriors.

[Terry Kyle]

Many thanks for the feedback and I want to add that I will be going for a NON-IM, non-Make Money Online niche in the experiment.

Of course, those are valid niches but I want this experiment to get outside the crowded IM pool.

Therefore, I'm looking at niches with a level of emotional desperation in the target market in a niche that I have never been in before (the same situation usually facing newbies)...

[purelight]

I'm really looking forward to your results as a subscriber to Angela and Pauls service.

[Tom Brite]

I think everyone here will be interested in the results!

Me and Terry have been bouncing ideas off each other and perfecting this for the past week before he goes off and does all the hard work and testing for you all!

Hoping to see some very definite differences in results to show which one is the best winner out of all of them.

Good luck Terry.

Tom Brite

[GuerrillaIM]

Hi Terry,

Unfortunately these types of threads get deleted. Real shame in my opinion. Any time anyone even slightly refers to a case study and *poof* thread deleted.

Hope this thread survives.

[JoshuaG]

OMG I've been waiting for this thread my entire IM career.
Can't wait to see the results!!

[m4ster]

I'll make sure to follow this thread! I'd really like to know the result and I have no doubts that you will stay consistent until the end. Good luck!

[hcl_23]

Looking for the results.

[Preciseim]

Will very much look forward to the results.

[Pattaya-Addict]

good luck with your experiment mate

will be follwoing this one

[Terry Kyle]

Thanks for all the support - doing my niche/keyword research now!

[Joe118]

Terry, I think you cannot expect this experiment to turn out well if it is run publicly during the 60 days that count. Everyone who wants to blog about the experiment will be distorting the results by including links. The links from blogs are not going to be worth the same and not going to be distributed evenly, so you lose control over the experiment.

That, and of course also the links from WF itself.

[younghamir]

I will subscribe to this thread and learn from your results. good luck

[LIndaB]

This looks to be a really interesting experiment. I'll be checking back often to see how it goes. However, I would suggest that you keep the domain, the niche, and the article titles under your hat until after the 60 days are over. If you don't, your results are going to be skewed. Plus, I've seen too many times in various private forums that when someone is doing a case study, that all kinds of people assume that it is going to be super profitable, so they jump in as well. If you want a fair test, you don't need those headaches to account for.

[flaminjo]

i will be eagerly waiting for the results... do keep us posted about the developments

[Mr_Julian_S]

Hi Terry and thanks for the wonderful tip. I would really like to give your suggestion a try and even excited to see the results then. Thanks again.

[SaSeoPete]

Hey Terry, good on ya definitely want to see the results.

May I make a suggestion? Maybe do not give the niche away until the end. I bet you are going to get 100's of people jumping on the gravy train here. Just a thought

[saintsguy]

Like everyone else I will be following this and definitely hope this thread is not deleted. It will be interesting to see how successful an IM'er can be using GoArticles. Best of luck, Terry

[Fernando Veloso]

Nice Idea Terry. You have another follower here.

Subscribed.

[Tony Dean]

Good luck with your ambitious experiment - I look forward to the results, maybe all Warriors can learn something from this.

[Terry Kyle]

My thinking at the moment is that without revealing the articles, site and keywords, this experiment loses a lot of interest.

What I'll do is provide some other attractive keywords in the chosen niche for the 'keyword vultures' to attack which hopefully will draw some of them off.

OK, here's my first screw-up in preparation for this experiment and exactly why this whole process is of value to everyone following along.

I was experimenting with Human Rewriter and just got back my first rewritten article (in under 24 hours by using the Express option) and wanted my article to have plenty of spin syntax options so that it would be good for 20-30 semi-original articles. This was for an article in an unrelated niche as a test.

When I set up the article though, I only asked for one rewrite but "Word Spins Per Sentence: 3" and "Spin Options Per Word: 2". These last 2 settings are the maximum you can request.

The produced spun article needed a little correction here and there which is fine but basically - unless I'm doing something wrong - I only have 2 articles: my original and the single rewritten one. Not quite what I had in mind.

In short, if you are using this service, you MUST choose 4 or 5 rewrites when you FIRST submit the article for spinning BUT keep in mind that this will 4-5x the price so you could be looking at $20-$30 for an article that has a 30-spin 'lifespan'.

That's OK as it saves a lot of time at my end and a buck an article is hard to complain about.

Also note that at Human Rewriter, without the express option ("Rush My Order" for 1.25 'credits'), article turnaround time is presently 48-72 hours.

A lot of IM is trial-and-error and my little errors with new services and programs (like this) in my experiment should mean that you won't need to.

If Jeremy Kelsall (or another Warrior experienced with HR) reads this and thinks the above is incorrect, I'd love to be advised otherwise Jeremy (he uses this service a lot)...

[jplanigan]

Originally Posted by Terry Kyle

I was experimenting with Human Rewriter and just got back my first rewritten article (in under 24 hours by using the Express option) and wanted my article to have plenty of spin syntax options so that it would be good for 20-30 semi-original articles. This was for an article in an unrelated niche as a test.

<snip>

In short, if you are using this service, you MUST choose 4 or 5 rewrites when you FIRST submit the article for spinning BUT keep in mind that this will 4-5x the price so you could be looking at $20-$30 for an article that has a 30-spin 'lifespan'.

Why not use humanspinner.com if you are looking for spun articles? At $0.11 per sentence it is sure to be much cheaper. I haven't used them yet, but I have heard good things about both quality and speed.

Patrick

[cashcow]

When I set up the article though, I only asked for one rewrite but "Word Spins Per Sentence: 3" and "Spin Options Per Word: 2". These last 2 settings are the maximum you can request.

I think your writers might have forgotten to do the spin syntax parts.

What you should have gotten (If I am reading this right) is 1 new sentence for each of your originals, but each sentence would have had 3 words spun for 2 different options.

Kind of like this:

The {quick|fast} brown fox {jumped|leaped} over the lazy {canine|dog}.

To me, that would make for a crapload of different articles, wouldn't it? Each sentence would have 8 different permeations, wouldn't it? Or am I thinking wrong?

Lee

[Terry Kyle]

Latest update:

Thanks for the positive feedback, firstly.

Apologies to Human Rewriter as my second and third test articles (for other sites of mine) came back with spun syntax OK - setting was one rewrite with the maximum number of word spins per sentence and maximum number of spins per word (about $6 for a 500 word article); maybe I missed changing those on the very first article

BUT...

when I pasted these two spun articles into SENuke and checked their uniqueness level under the "Social Network/Article Directory Nuke" module, the two spun articles achieved scores of only 10% and 15% unique. 20% is a minimum and 30% preferred.

HR are now showing a 3-6 day backlog so I will look at Human Spinner as an alternative (good to try both of course). Later in the test I will feed in some HR-spun content for my SENuke Linkwheels for Article 3 (2 rewrites per article - about $12 per spun article).

That means that I will spin articles myself in Power Article Rewriter as I have pretty much settled on my niche and keywords for Thursday's launch...

[7digital]

Good luck Terry!

I'll be watching this thread!

[younghamir]

This one is very exciting.

[Terry Kyle]

All set to go for tomorrow and lots of stuff to share then!

[esh]

This sounds interesting, i am going to follow this thread!

[NicholasCarter]

Love to hear the results

[Tom Brite]

Looking forward to it Terry!

You have lots of people waiting to see your results!

Tom Brite

[Terry Kyle]

DAY 1 - PART 1

OK, let's get rocking and rolling!

Even though Keyword research isn't the main part of this experiment, I still wanted to document my process.

In short, I wanted to target a niche that I have never worked in before (like newbies), involves a degree of emotional desperation on the part of potential buyers (instead of a hobby like kite flying or RC cars - actually those RC car guys are pretty fanatical!) and wasn't in 'make money online'.

For the Keyword research part of my experiment, I've bought Keyword Elite 2.0, used Micro Niche Finder (which I've had for a long time) and SENuke (which I've trialled for 7 days only in the past but am using 'properly' now with Tom Brite's assistance).

I am also in Jeremy Kelsall and Donald Van Fossen's SEO Bootcamp (which I do recommend) as I'm always hungry for new tactics to expand my IM repertoire.

Obviously I'm also using my own accumulated IM knowledge from the last year and a half or so.

Now, in order to keep the experiment as 'pure' as possible, I'm going to reveal everything I'm doing here EXCEPT the keywords I'm targeting and the articles and site I've created (at least not until further on in the experiment).

As the experiment unfolds, I will reveal all of these as I think it is necessary for the enjoyment of participation.

Also, my own view towards mistakes, failures and screw-ups as we proceed is that they're great as I can continue to fine-tune my IM strategy and not make those same errors again - trial and error in part.

Now, today the experiment has begun so I've been doing my specific keyword research, using all of the above tools and finding out a few interesting things which I'll share in a moment.

Firstly though, here are the tables of a bunch of keywords from which I will shortly choose my final 4 (for the 3 GoArticles and one new site).

The column on USA Exact Monthly Search Volume is straight out of Google's Keyword Tool BUT underneath that, I've included the Keyword Elite monthly search estimate which usually is VERY different (even though it is presumably based on Google results).

In comparison, Micro Niche Finder (MNF) delivers exactly the same search volume as Google's Keyword Tool.

In the comparison table, I have also included SENuke's estimate of whether ranking is achievable for a specific Keyword.

SOC stands for Strength of Competition and each program makes a guess about how hard it is to rank for a Keyword - the lower the number the better and green means go, amber means treat with doubt (a bit hard to rank) and red means forget it, too hard.

Later in the experiment, I will publish the unblurred table here.

However, in the next post here below, I will outline the dangers of relying on all this information and what else I take into account when targeting a niche.

Oh and another nice tip I accidentally picked up in the SEO Bootcamp of Jeremy and Don (which I don't think they'll mind me sharing as they didn't make a big deal of it) is that when you are using Yahoo Site Explorer to check a site's backlinks, do it with the SEO Quake for Firefox plugin enabled so that you can see the PR and other details about sites linking to the one you are analysing e.g.

Here are my initial Keyword research tables which I will discuss in more detail in the next post below:

[Terry Kyle]

DAY 1 - PART 2

So, as you saw in the previous post, obviously the results from these tools differ in the search volumes predicted (by a huge degree in some cases!) and in the degree of difficulty.

My own shortcut way to make the final choice is to look at the top 3 sites for a Keyword in Google:

[1] are there any Web 2.0 or article directory pages sites there e.g. Squidoo, Buzzle, EZA or GoArticles etc;

[2] how many backlinks are there to those Web 2.0/article directory pages (Google generally ranks pages not sites)?

[3] in the absence of those types of sites in the top 3 (which is usually a big green light for me if they ARE there), how strong is the #1 site in terms of backlinks (to the page, not the site) and authority (e.g. is it a Wikipedia page) which could be harder to budge.

I have just started to look at the inanchor search operator but the numbers usually seem pretty high for these so I'll stick to the above measures until I get better at that inanchor approach.

There's a good WSO: 1000 very low inanchor Keywords with reasonable search volume!

As the above suggests, typically I will IGNORE:

[1] "in quotes" 'competing sites' numbers (the top 3 or 5 are your true competition)

[2] SOC estimates by automated programs (as they don't normally take backlinks into account though Keyword Elite might, I need to look into that more).

Search terms with 4 or more words are usually a less competitive place to start though occasionally you may get a 3-word term that isn't too competitive.

OK, so I've got my 4 terms from the comparison tables on the previous post:

Keyword #1

Keyword #5

Keyword #11

I'm pretty happy with the first three Keywords in terms of seemingly beatable competition but the 4th is less satisfactory. Nonetheless, I'm going with #12 which is the one I'll let SB101 loose on with the new website.

And in order to write my articles (3 for GA under different new account names and one for the new WP blog website), I am going to read 5 or 6 articles in my target niche on EZA and make notes as I go.

That's four x 500 word articles with the Keyword in each 5 times.

Written today.

As I complete each article, I will add it to GA. In my experience, GA has a new article up and running in 24 hours or less. It seems automated and pretty quick so I assume that GA doesn't use human moderators like EZA.

I will also create a new account at Clickbank now and give each article a unique tracking idea:

Article 1 (Angela/PJ/My links only - IRRELEVANT sites only) Tracking Code: ART1

Article 2 (Angela/PJ/My links only - RELEVANT sites only) Tracking Code: ART2

Article 3 (SENuke Linkwheels with wheel sites backlinked using Angela/PJ/My links) Tracking Code: ART3

I cannot create SENuke linkwheels pointing to my money site GA until it has been approved and given its own URL by GoArticles i.e tomorrow (for which I will need a spun version of this article which I will do using Power Article Rewriter though any spinning program will do)

Article 4 (New website/WP blog/SB101 program from WF WSO) Tracking Code: ART4

All links will be cloaked using offto dot net.

Realistically, Test 4, the new website being powered by SB101, probably won't be up and running until very late tonight or tomorrow morning but that slight time lag shouldn't really make much difference over the 60 days of this test.

If one (or more) of these models is successful financially, you only need to replicate the steps that I detail in this thread to hopefully enjoy similar results.

Back to work now...

[Terry Kyle]

First GoArticle up and first Google Alert set up (with a distinctive long phrase from the article in inverted commas for exact match) to track when the big G spots it.

By the way, if you aren't using Google Alerts to track your own progress and that of your competitors, you should be...

[Terry Kyle]

Also, if you find my screen grabs too small and hard to read, let me know.

You can make your screen bigger by holding down the CTRL button and spinning your mouse wheel (in Windows anyway).

[Terry Kyle]

I've also added individual CB tracking code to cloaked aff links in articles e.g.

Art1a (top link)

Art1b (link within article)

Art1c (Bio Box)

So we can track which performs better...more on that later.

[preneurseo]

Thank you for this informative post. I hope you'll update this from time-to-time. We'll be following you.
Thank you.

[Terry Kyle]

Day 2 - PART 1

OK, so 2 of the three GoArticles are now live with their own unique URLs.

When all three are live, I will create a unique RSS feed using this site:

WebDevTips - Code Generators - RSS Headliner

(brilliant trick picked up here on WF)

and submit that RSS feed through BOTH Warrior Big Mike's RSSBot AND individually to FeedAgg, FeedAge, Pingomatic, Autopinger, Feedburner and Feedraider.

I will also use a bunch of bookmarklets on my Firefox toolbar to social bookmark the new article pages (I will do a screen grab soon of that).

These steps are not for backlinks but to get indexed by Google quickly and by doing all three at the same time, I am giving them an equal chance.

Also, I used Warrior Jack Duncan's Ultimate Keyword Theme Extractor which reveals the common phrases in the top ranking pages for specific terms so that I could make sure I was also using them.

This whole LSI debate is for another day but I figured it couldn't hurt!

GoArticles does not offer its authors a unique RSS feed (which EZA does and you should be blasting your EZA Author RSS everywhere!) but the site above lets you get around that. Nice.

More soon...

[Tom Brite]

Lots of work there Terry and nice to see you examining other points too which is cool!

Tom Brite

[rbecgolf]

Good luck with the experiment. It will be fun to watch.

[Fernando Veloso]

Terry:

I am amazed with the amount of work and detail you're going through.

Jesus!!! This is getting a precious thread.

[JMartin]

Guess I'll be reading this thread to see how it turns out for you.

[Terry Kyle]

DAY 2 UPDATE:

Thanks for the support!

OK, so 2 of my 3 GoArticles are live but I'm still waiting on the third (the first were up in under 12 hours).

I also set up the site for Test 4 using namecheap and my hostgator reseller account.

The .com domain was of course already taken so I added a '2' at the end e.g mydomain2.com just like you have to do with a new email address sometimes.

There are no hyphens/dashes in the domain, just one long word string.

I set up WP on the site, customised settings, theme, plugins, permalinks etc and will post screen grabs of all that tomorrow.

I posted my new article tailored for that 4th keyword to my new WP blog (using the free Direct Response theme), added the keywords from all 4 test versions here as category and tag names.

I then 'enrolled' in Warrior Ken Fry's Social Bookmarking 101 program with the inner page post url and I am now waiting to find out if that site/page has been approved to go live in the system. I'll let you know on that.

As the SENuke target article (#3 if you remember) is live, I can start on my link wheel for that tomorrow (Saturday) as soon as my maximum variation article is back from Human Rewriter (all $35 of it) - hopefully my SENuke gunslinger partner Tom Brite won't be out surfing then!

OK, good day's work today - loads of screen grabs up tomorrow and more good fun then, gotta go to Salsa now...

[Terry Kyle]

Oh and if you think this thread is valuable, why not Rate This Thread highly up above in the WF toolbar?

[Rendix]

Terry,

One comparison I would like to see at the end of your analysis is a comparison between googles exact traffic projection and your actual incoming traffic. Also the corresponding SERPS placement for those sites.

Best of luck to you!