The What Really Ranks? Experiment thread. Find out what IS really working in SEO..

[Mike Anthony]

I been meaning to do this for awhile now. SO much talk about this or that kind of link working. Do profiles still work what about article directories? What about contextual links and PR? But what are the google search results saying? I mean its right there right now in the search results what is working and what is not. So heres what I want to do for this experiment. Its not like the normal experiment done here where someone takes a domain and ranks it.. We want to learn a lot more than just what one site can tells us. so we want to look at a wider range of niches and sites.

On the other hand we don't want to expose anyones sites and links on an open forum. SO here is what I want to do. You give me some of the top keywords in IM and I will look at the sites that are ranking in the top 5 (BUT NO PARTICULAR SITE AND NO BACKLINKS EXPOSED) and report on the following

What kinds of backlinks were used by the sites that rank.

Forum profiles (and percentage)
Article directories (and percentage)
PR contextual links ( and what percentage)
Blog comments (PR and what percentage)
Blog rolls (PR and what percentage)

(welcome to suggest any other stats to report on and yes based on the backlinks that are showing and the amount that can be pulled up in a checker - nothing is 100%)

I already have a couple of the terms I would like to check but need your suggestions of other popular and extremely well known Keywords people go for in IM (only the popular ones so as not to be accused of exposing anyone's keywords).


"Make money online"
The all too popular example of "Dog training"

are in the bag. but please ad your own (do not post your own niches). Would like to have at least ten to go with. This way we can get a sense of what is really working out there not just for one site and one niche but an overview of whats working for lots of sites ranking at the top of Google in some of the most common IM niches.

IF theres anything you don't understand about this approach let me know. if not then fire your top dream niches at me

LIST SO FAR
"Make money online"
"dog training"
"car insurance discounts"
"online savings account"
"Weight loss
"SEO service"
"online dating"
"affordable Vacations" (unless someone has a better idea for travel niche)
""mesothelioma lawyer"

Only need 1 more and we are good to go

[adamv]

This should be very interesting and something I am going to keep my eye on.

I would like to see some research into difficult keywords but also moderate to easy keywords. I'm interested in what it takes to rank for "car insurance" vs. "car insurance discounts" vs. "cheapest car insurance for teenagers".

It would be interesting to see if links that may not be very effective for the ultra competitive keywords would still be effective for ranking the easy to moderate keywords.

[Mike Anthony]

Hi Adam,

We can go moderate also. I won't go down to easy because if theres no competition at all I have found any link can help. So we can throw in "car insurance discounts"?

Still need seven more. Don't know if i should put Car insurance in - thats ultra difficult and not something I think the average Imer is going for (at least realistically).

Originally Posted by adamv

This should be very interesting and something I am going to keep my eye on.

I would like to see some research into difficult keywords but also moderate to easy keywords. I'm interested in what it takes to rank for "car insurance" vs. "car insurance discounts" vs. "cheapest car insurance for teenagers".

It would be interesting to see if links that may not be very effective for the ultra competitive keywords would still be effective for ranking the easy to moderate keywords.

[michael scott]

here are 3 keywords I'm currently going after. One of them is already on page 1, position 3

penny stocks
online savings account
cruises

[michael scott]

For "cruises" you can ignore the top 2 and focus on #3 and down because you'll see more juicy stuff.

Also, I noticed you didn't mention internal linking as a factor for ranking.

[eljeffe77]

Interesting thread, just responding so i can follow it easier. :-)

As for me, I'd love to see what's working for "weight loss". I'm not even in that niche, but as it's one of the more 'insane' ones in IM, it'd be interesting to take a look at what's working for these folks...

[JOleng]

This should be great experiment. If I could add a keyword :

"SEO service" let's see what top ranking SEO guys are doing in terms of SEO

[Mike Anthony]

Originally Posted by michael scott

Also, I noticed you didn't mention internal linking as a factor for ranking.

thats because we all know on page SEO has no value

No I was concentrating on off site linking

[JoshuaG]

I'd love to see some travel related keywords make the list.

[Mike Anthony]

Originally Posted by JOleng

"SEO service" let's see what top ranking SEO guys are doing in terms of SEO

Which is why I stay out of SEO serps myself as they are the number one pages reverse engineered. Anyway added.

[Mike Anthony]

Originally Posted by JoshuaG

I'd love to see some travel related keywords make the list.

Well go for it and suggest some.

[Clyde]

In-contextual high PR blog posts links, blog rolls, all still work, give most juice.

Article directories (non-spun) has limited juice, do it once and don't bother. Same goes with Social Bookmarking links, a one-off and that's it.

Forget forum profile, Web 2.0, and blog comment spam for now, unless you have an aged domain names and established portfolio of backlinks already.

For your case study, please make sure none of the keywords has an EMD ranking top 5 because that can skew the results.

My keywords vote:
viagra
cialis
xrumer service
backlinks

[Mike Anthony]

Originally Posted by Clyde

For your case study, please make sure none of the keywords has an EMD ranking top 5 because that can skew the results.

What ever really ranks is what the experiment is about . If there are any EMDs then I will note it not exclude it but given its a cross section of top five in each serp that would dilute some of that effect.

[LiamP]

Something like "online dating" or "get your ex back", so that all 3 of the big IM niches (health, money, relationships) are covered.

Vacations are a biggie too.

[Fraggler]

iphone 5

[patrich]

Originally Posted by Mike Anthony

I been meaning to do this for awhile now. SO much talk about this or that kind of link working. Do profiles still work what about article directories? What about contextual links and PR? But what are the google search results saying? I mean its right there right now in the search results what is working and what is not. So heres what I want to do for this experiment. Its not like the normal experiment done here where someone takes a domain and ranks it.. We want to learn a lot more than just what one site can tells us. so we want to look at a wider range of niches and sites.

On the other hand we don't want to expose anyones sites and links on an open forum. SO here is what I want to do. You give me some of the top keywords in IM and I will look at the sites that are ranking in the top 5 (BUT NO PARTICULAR SITE AND NO BACKLINKS EXPOSED) and report on the following

What kinds of backlinks were used by the sites that rank.

Forum profiles (and percentage)
Article directories (and percentage)
PR contextual links ( and what percentage)
Blog comments (PR and what percentage)
Blog rolls (PR and what percentage)

(welcome to suggest any other stats to report on and yes based on the backlinks that are showing and the amount that can be pulled up in a checker - nothing is 100%)

I already have a couple of the terms I would like to check but need your suggestions of other popular and extremely well known Keywords people go for in IM (only the popular ones so as not to be accused of exposing anyone's keywords).


"Make money online"
The all too popular example of "Dog training"

are in the bag. but please ad your own (do not post your own niches). Would like to have at least ten to go with. This way we can get a sense of what is really working out there not just for one site and one niche but an overview of whats working for lots of sites ranking at the top of Google in some of the most common IM niches.

IF theres anything you don't understand about this approach let me know. if not then fire your top dream niches at me

LIST SO FAR
"Make money online"
"dog training"
"car insurance discounts"
"online savings account"
"Weight loss
"SEO service"

Only need 4 more and we are good to go

lol Mike, I saw the thread title and thought to myself, "Oh god, another one of those threads where someone that got started in seo yesterday preaches about profile links or something!", but then I read that it was you and what you were doing, so I am sure you have got it covered, but if you need a second pair of eyes feel free to give me a holler. I am happy to help.

It does sound like a neat experiment, I usually monitor these types of things on my own, its important to watch the changes within the industry rather than running in blindly hoping something works. Which is what most guys are doing these days.

Anyways, off to subscribe to the thread.

[RosieCain]

I am not sure whether it is approprite or not to post here as I am not a war room member when I see all war room members posting above.

If at all possible, I would like to suggest a comparison for the following metrics:

ratio of no. of backlinks to corresponding no. of unique domains

& then,

ratio of those unique domains to no. of unique c-class IPs.

From my un-systematic and random observations in the past, I always believe there is a correlation of the above to the ranking especially for competitive keywords.

Anticipation:
low ratios in both above will tend to give better ranking

Thanks

[adam westrop]

Very interesting, some questions though:-

(1) - Are you going to record age of backlinks/discovery date? Age of links seems to be very important these days
(2) - Are you recording internal links? All the popular link analysis tools do a rubbish job of picking up on internal links
(3) - Should you not do a end cross sample of links in the Google index link checker in Scrapebox to see what % of a sample of 100 are indexed? So we know what type of proportionate figures to deal with?? I feel this is quite important as someone could have 10's of 1000's of profile links in their backlink profile and could be ranking high, however none of the profile links are indexed in Google and its other links doing their job propelling their ranking.

Overall, this test is unable to be conclusive (but then is any in SEO).

But big big big big big props to Mike for starting this... It will provide some good indicators.....

So thats 1/40 'value' threads in the SEO section I've seen today.

[JawadAshraf]

Do whatever keyword analysis you want to. But I just want to add only 1 thing from my own SEO experience and would like everybody to have a note of it that whenever you use auto back-linking process your work will go to hell forever. I have this bad experience with lots of my sites. That's why STAY AWAY FROM AUTO BACK-LINKING process. One thing more DON'T ever BLAST Direct back-links to your site, it doesn't work any more. Although ALL types of back-linking is working RIGHT NOW i.e blog commenting, forum profiles, article submission, social bookmarking, link wheel etc etc but if you do everything MANUALLY & with a slow process. Otherwise you are just wasting your time, money and energy.

Good luck.

Jawad

[Mike Anthony]

Originally Posted by adam westrop

Very interesting, some questions though:-

(1) - Are you going to record age of backlinks/discovery date? Age of links seems to be very important these days

I can add that. Thats a great addition

(2) - Are you recording internal links? All the popular link analysis tools do a rubbish job of picking up on internal links

I'll see about that but for that very reeason it could make that a bear of a job. In fact at the moment even in coming links is going to be more of a challenge than a few months ago now that almost every tool is in transition due to Yahoo closing their API.

(3) - Should you not do a end cross sample of links in the Google index link checker in Scrapebox to see what % of a sample of 100 are indexed? So we know what type of proportionate figures to deal with?? I feel this is quite important as someone could have 10's of 1000's of profile links in their backlink profile and could be ranking high, however none of the profile links are indexed in Google and its other links doing their job propelling their ranking.

Problem is Google is notorious for not reporting links that ARE in their index. Thats why people have always relied on other engines. So I don't know that such a proportion check would even tell us anything. Also I am making people pick the keywords for objectivity but I'd be shocked if profile backlinks play a big part in many of these search results.

Overall, this test is unable to be conclusive (but then is any in SEO).

True but like you said what is? Maybe not ten but I do find searching Google and looking at the top rankers VERY Conclusive. I do like threads where people rank their individual sites but you never know with those whether factors relevant just to those sites and domains influenced them. Here we will be looking at ten keyword results across 50 or more sites.

[Mike Anthony]

Need just two more guys. I am passing on sex related keywords and definitely not skewing the results by including specific kind of link keywords

[Vlad Romanov]

So you will look at sites that have already been ranked using the methods u mentioned above? What exactly is it going to prove or show about what works now? Unless I'm seeing this wrong all you will get is useless information about how sites used to be ranked several years ago if not several decades ago...

[Mike Anthony]

Originally Posted by Vlad Romanov

Unless I'm seeing this wrong all you will get is useless information about how sites used to be ranked several years ago if not several decades ago...

Blown away every now and again by the crazy ideas some people have about SEO. So the top five in the search results now reflects how sites used to rank several years ago?????? or decades ago???

I thought people said SEO was easy. then why do people continue to have these way out ideas despite being here months?

Anyway I am getting under way. at some point I know someone will add the last two needed keywords as we move along.

[Vlad Romanov]

Originally Posted by Mike Anthony

Blown away every now and again by the crazy ideas some people have about SEO. So the top five in the search results now reflects how sites used to rank several years ago?????? or decades ago???

I thought people said SEO was easy. then why do people continue to have these way out ideas despite being here months?

Anyway I am getting under way. at some point I know someone will add the last two needed keywords as we move along.

Thanks for such a contrastive reply. Instead of being blown away you could clear some points for me.

Let's say you look for "making money online" as the keyword. You will find old sites which have been ranked years ago. It is obvious that since then they have received natural links from all over the place... In the end this study will have no real conclusion since you will not be able to break down if certain backlinks were used to rank, are natural, etc. My point is your study has too many variables to provide a meaningful conclusion.

Sorry that I had to make it clear for those who don't understand SEO and have been here for a while...

[Mike Anthony]

Originally Posted by Vlad Romanov

Thanks for such a contrastive reply. Instead of being blown away you could clear some points for me.

I'll try but you are pretty committed to your lack of knowledge and I don't want the thread totally hijackedby arguing presumptions that are false before things even begin

Let's say you look for "making money online" as the keyword. You will find old sites which have been ranked years ago.

Thanks for making this quick and painless. Perfect example. A warrior here just ranked for THAT TERM THIS year and a just few months ago. Age did not make it rank #1 but the placement of some links this year did. That fact alone proves what I just said. You have zero clue what you are talking about - but for brownie points

B. I have ranked newer sites over older sites and many people here have and know how to partly from doing research JUST LIKE THIS. I may be doing this fresh for these keywords but my business as a SEO has involved doing this for years. The number one way for anyone to rank is to determine how the competition rank and go out and get links towards that purpose. By your rationale it would be impossible to rank because the other site will always be there because of age. SEOs do what you claim is impossible every week.

C) there is no defined rate of organic or natural link pickup based on age. A new site can have a piece of content that goes viral, picks up editorial link gets featured somewhere that gives it mad links in a short time etc etc even negotiate links for their site. Your premise that age necessarily relates to number of organic links is off by a mile. YOu can go out and get links to match almost any link profile out there. the fact that you may not know how to do it hardly makes it harder for someone who does. Age does pplay a part but it is hardly insurmountable and its through looking a t the link profile of the sites that you determine whats needed. If you haven't figured it out - thats exactly the purpose of this thread.

Lastly your appeal to variables is pointless. There has never been and will never be a SEO study regarding Google that does not have plenty of variables. No one but Google knows all that is in their algo. IF we used that kind of thinking we would do no tests at all and be even further in the dark. If it gives people pointers and some direction then thats fine. The test does not have to meet your condition of conclusiveness to be helpful.

Now if you still think that you know what you are talking about then please feel free to skip the thread but I won't bother responding further so as to derail it with correcting people's deep misunderstandings of SEO.

I admit you did make it very clear that you don't understand SEO but this is not the thread for me to straighten out all that you don't understand about it.

[patrich]

off the top of my head, how about something like "mesothelioma lawyer" and "bail bonds" as your last two? I know the mesothelioma niche has some really high cpc's which would be of interest to the adsense guys. And you can make some big money in the bail bonds market with referrals.

[patrich]

Originally Posted by Vlad Romanov

Thanks for such a contrastive reply. Instead of being blown away you could clear some points for me.

Let's say you look for "making money online" as the keyword. You will find old sites which have been ranked years ago. It is obvious that since then they have received natural links from all over the place... In the end this study will have no real conclusion since you will not be able to break down if certain backlinks were used to rank, are natural, etc. My point is your study has too many variables to provide a meaningful conclusion.

Sorry that I had to make it clear for those who don't understand SEO and have been here for a while...

Uhmm, the number 2 site(on my end) for that term was registered just over a year ago and I briefly looked at their link profile which consists of a lot of blog networks and comment spam. Which kind of kills your argument.

And it is very easy to determine if the links were natural or not. At this stage in the game you should be able to tell whether or not a link was given "naturally" or if it was bought/spammed/etc.

[redtaksaka]

How about 1 geo targeting keyword? I'll go with "UK dedicated hosting".

[Mike Anthony]

Originally Posted by redtaksaka

How about 1 geo targeting keyword? I'll go with "UK dedicated hosting".

Hosting niche kind of veers off the normal IM kind of keywords unless you mean some affiliate marketing.