This isn't a flame, just a simple observation that what works for you may not work for every site, with every link partner.
It's not the number of links that matters in testing, but the number of sites involved in your test; it appears to be two (one linked from inside pages, one from links pages).
We know nothing about the content or niches of these two sites, We know nothing about the quality or nature of the links.
We don't know what "Links from links pages pfft. Links from non links pages rock" means.
You're happy, that's great - our mileage may vary.
It was tested on an old site. This would not work on a new one. As for types links it does not matter. I can tell you all pages had a cache in google.
The non links pages worked in 10 days( multiple times) the links pages did absolutely nothing in 30 days.
we are talking at least 3000 different sites in each case.
Most links were from individual IP's on both links pages and non links pages.
[edited by: Crush at 12:32 pm (utc) on Feb. 28, 2007]
|The non links pages worked in 10 days( multiple times) the links pages did absolutely nothing in 30 days. |
what does worked mean?
I have found that 4-5 links from an authority/trusted site can have more effect than 100-200 links from "other" websites whether link pages or not.
I would imagine that your data is tainted because those offering links from links pages will be low quality sites and those offering links from their content pages will be quality sites.
You may be right that links pages vs. content pages is better for linking (I believe so myself as do most webmasters) but your test does not conclude that.
A test to conclude such would be 4000 sites linking to one site from their content pages and the exact same 4000 sites linking to a different site from their links pages. Different sites/industries/pagerank/trustrank etc. etc. etc. can open a whole host of possible corruptions in your test.
|Take it from Uncle crush. Stop your link begging requests and if you do please get them on an internal page that is well linked from within the site. G can devalue links pages. |
It sounds like maybe you got links on "link pages" that were "hidden", that is not well linked, on the host (selling) site?
Is that right?
If so, wouldn't that make a difference in the outcome of the test?
Sorry if I'm missing something, but the title of your post mentions recips in a directory structure and the body mentions purchased links either within content or on a links page. Aren't recips two way and puchased links one way links?
|Different sites/industries/pagerank/trustrank etc. etc. etc. can open a whole host of possible corruptions in your test. |
I agree. There are many more considerations to getting links than links page or not.
Trust rank is often a huge factor. I bet a link from a links page on the BBC's web site would carry a lot of weight even if it was a PR0 "further resources" type of page. The best sites to get links from are often the ones that do not sell links.
[edited by: Jane_Doe at 4:30 pm (utc) on Feb. 28, 2007]
So there are lots of schemes out there that sell links. Basically it is a bunch of webmasters who set up a script and they swap links that way. They owners of the script sell links as if.you are a member adding to the community.
Set up like this:
Add link for directory to index page> add links directory> then to swap links. Basially a typical link directory and pretty easy to spot IMO.
So I checked the links on yahoo, they all got cached and showed as backlinks but no serps.
NOw the other type of links I bought ( sorry but I do not want to elabourate) appear on a page with within a site, not on a links directory, they also got indexed but the difference was the urls I threw them at ranked in 10 days or so constantly.
What is a "quality link". It is a link that has many inbounds ( internal or external) to it and no more.
Please do not get trustrank confused with what I am talking about here. The domain I point the links at it trusted. Anyone that throws links at a new domain is an idiot.
[edited by: Crush at 4:56 pm (utc) on Feb. 28, 2007]
Crush your experiment will never be ok for the flame team.
even <snip> links analysis that clearly shows they buy links all around the place is not good enough because "<snip> is huge"
you'd better show off with your car (if you have a fancy one from your superlinks) or sth like that. that would be "evidence"
[edited by: martinibuster at 9:01 am (utc) on Mar. 1, 2007]
[edit reason] See TOS [webmasterworld.com]. [/edit]
Just trying to get people to think differently Idol. I still get far too many begging link requests. It would be a better for all if people started trading links this way,
|What is a "quality link". It is a link that has many inbounds ( internal or external) to it and no more. |
What about relevance? Did you find it didn't matter?
If it mattered, did it apply to both kinds of links?
The 'links pages links' were part of a standard link exchange scheme, and (of course) were rapidly discounted, often with a bad neighborhood. But were the 'in page' links also part of a link exchange, or were they a 'one off' arrangement?
I'm still not flaming (I'm really not!), but I'm interested in how your study was set up, as I think the conclusions you've drwn are very interesting; BUT being a sad old sceptic, I'm interested in seeing if any other conlusions are equally valid.
"What about relevance? Did you find it didn't matter?"
Not AT ALL. Themeing is still a fantasy.
"If it mattered, did it apply to both kinds of links?"
Infact the links pages were themed because they went into teh right category.
"The 'links pages links' were part of a standard link exchange scheme, and (of course) were rapidly discounted, often with a bad neighborhood."
You have been reading too much Matt cutts. "Of course" is supposition. Not to do with bad neighborhood, it is to do with directories and their stucture. G's also can discount them. Bad neighborhoods are really a thing of the past and a scare tactic.
But were the 'in page' links also part of a link exchange, or were they a 'one off' arrangement?
Cannot really say much apart from it does not matter. It could be in a blogroll, could be on page. In fact they were everywhere.
"I'm still not flaming (I'm really not!), but I'm interested in how your study was set up, as I think the conclusions you've drwn are very interesting; BUT being a sad old sceptic, I'm interested in seeing if any other conlusions are equally valid."
Well, really I am not posting here for brownie points. I have had some help here in the past and am helping now. I am way more advanced than most members in WebmasterWorld in terms of the experiments we make. That is not bragging it is just a fact that we push to see what can and cannot be done.
Summary, links pages devalued. Almost all other types of pages OK.
No skepticism here, the word “devalued” is to kind in describing how far the impact of links from that type of format has fallen over the past 18 months.
|Just trying to get people to think differently Idol. I still get far too many begging link requests. It would be a better for all if people started trading links this way |
i am just so glad people don't listen.
or maybe those who reply in the threads are just bots to get more posts in the user profile and people controlling these bots forgot to switch the flame mode off.
Don't be too sensitive; it really isn't fair to suggest that every time someone disagrees with you that they are 'flaming'.
They just see things differently. The whole point of a forum is to discuss and exchange differing views - and to my aged eyes, we're doing exactly that; not a burning bush in sight!
Two glaring things here:
1) I have never heard of 4000 reciprocated links in 30 days. Not even close, doing it the right way. Maybe by joining link networks.
2) One example is not enough to draw any global conclusions.
DomainDrivers, please let me know of your experiments and how you can get to #1 for good serps.
I really don't "experiment". Never have. I haven't changed my apporach since 1997. It worked then. It works now. I have no use for SEO fads.
I simply reciprocate with other sites that are relevant to the client's realm of interest. Not PR analaysis, etc. Relevance. Period. It is branding, pure and simple. The SERPs derive from that.
I am sorry if that is not sophisticated enough for many people on this forum, or that is doesn't comply with the latest theories. I could care less. As I said, I am in the business of branding within a realm of interest.
I can't guarantee SERPs of any kind. In most cases, we are not responsible for the on-site optimization. But, in that environment, we get to see what works and what doesn't.
I am all for focused optimization, based on thorough keyword analysis, with "people oriented content", not off-topic bait or over-optimized babble.
When that is done thoroughly, it all seems to work remarkably well, with substantial SERP stability. I have seen Google rip SEO fads to shreds.
So, I stay away from SEO fads, (link bait being the latest). I simply do what has worked since 1997.
[edited by: DomainDrivers at 6:36 pm (utc) on Mar. 4, 2007]
you are playing with fire. It worked today, but tomorrow you may sink or get banned.
I have to believe that massive link exchange schemes that create similar themed pages on many different sites WOULD be detectable by Google, both algorithmically and in the course of manual checking. Either way, once a network is discovered, those link pages would be devalued.
I think the experiment is interesting, but I don't think 4K automatically exchanged links that popped up in the course of a few weeks can be directly compared to getting links from 4K individual site over the course of a year or more. (Not that one could get that many, or that it would be worth the effort.)
Your point about purchased links from non-links pages is well taken - even though Google would like to prevent link buying, it's pretty hard to catch algorithmically.
How did you "buy" a total of 8000 links? i've reviewed a couple of link buying services and they're charging an avg of $10 per link. So, $80k you spent on this? Or are the majority of these links all from a couple of sites ... just in their header or footer or something?
Also, where did you go to buy these deeply-linked inbound links? I'm trying to find the same thing. Both services I've looked at who sell 1-way inbound links, it looks like they're just doing a fancy 3-way link scheme and all inbounds come from links pages. Iguess they're banking on ignorance of the consumer ... I tend to believe as you believe that those links are soon if not already worthless and contextual (not on link pages) links are far more valuable. Trying to find someone who would sell them to me!
What does "pfft" mean?
Of course - links from content pages count for more than the links page - for the simple fact that on the links pages there are already other links there, so the "vote divides" among them. And when you add the fact that there is not much content on the links pages (on most websites), the vote from those pages dont count for much. It's SEO 101.
But it's free. I would never pay anything for links - I have had a website out of the sandbox and instantly ranking on very competitve keywords with 10 reciprocal links.
Also - I never exchange links with websites on their automated link pages of any kind - I do exchange links with those websites using internal pages. And it's not that hard to make your link page look like a content page - just add some relevant content.
If you want to buy links - sure, do what you want. But you dont have to spend that money on links in order to rank. You can rank #1 with only reciprocals as long as:
# You exchange links with only related websites.
# Be selective to the websites you link
# Put an extra effort into your "links" pages.
These are golden rules for a long-term presence on any SE. Follow them and you can save yourself a whole bunch of money :) And if you dont - well, you will have to buy links, because no one would want to link to you ;)
I think Crush and Domain Drivers are both right, just not discussing the issue on the same wavelength.
I recently sold some links to a related site. He put his links on my pages. I made money. I did not get a reciprocal link, nor did I ask for one.
From his point of view, the links provide traffic. From my point of view, they're advertising for his site, plain and simple, and I am in business to make money.
I have never asked one single site for a link but I got them nevertheless, based on content. It worked for me over the past 7 years, but there's no guarantee that it will continue to work indefinitely. In the long (and short) run, it's all about traffic, period.
Think about how a site is popularized on the internet. We have Goofle and their algo and PR. The algo isn't perfect and PR is vastly overrated, but still valid.
But more traffic comes from advertising than anything else for many sites. Why do you think the very first thing a company does to "monetize" itself is to offer up advertising? Because paid advertising works. If it didn't, Google, Yahoo, ebay, MySpace, etc. wouldn't do it.
There's a new paradigm coming soon. Link farms and link pages are history. Crush is totally on the right track.
So how much did your positions increase? What was the change in pagerank? How many more backlinks were you showing in Google?
How can you tell it was the purchsed links that made a difference and not the exchanged links?
Hello Uncle Crush, I'd be interested to know your methodology for screening, negotiating and paying for 4000+ links, seems like a lot of work still :)
Sorry, I am giving no specifics. The reason I am sharing this is because people still doing links swaps should really stop because there are diminishing returns. We did recips link swaps for years. From 2001 until about this time last year and they worked great. Now you need new types of links. To be honest we have been buying links for ages now and have never been smashed.
Do you think google can see your credit card transactions? I cannot fathom why people crap themselves about buying links. Just look for footprints, even then there is a slim chance of getting caught.
For the crap type you can buy 4000 links for $175 a month from reciprocal link schemes but these are the type that do not work because of footprints like pages called links.html, links.asp etc.
Most important is the domain. Use an oldie with bundles of trust.
[edited by: martinibuster at 10:06 am (utc) on Mar. 5, 2007]
[edit reason] Removed specifics. [/edit]
I'm broadly with Crush on this, though I'm not too sure about his numbers.
|I simply reciprocate with other sites that are relevant to the client's realm of interest....The SERPs derive from that. |
That the SERPs "dreive from that" is something you can prove (apart from quoting Matt)? It would be fair to require the same level of proof for assertions others make as you provide for the claims you put forward.
It's been a while since Google started apparently devaluing links from links pages. In fact, even links.htm pages that were around since before Google - and since before citation based ranking - seem to be completely devalued. And that seems to happen even when the links really are relevant, go to quality content and edu type sites, and haven't changed in the last 10 years. I believe Google's gone overboard with their link exchange paranoia and with protecting against manipulated links.
Which means that it makes even less sense now for reciprocals. The naive use "three way reciprocals" that they believe can't be detected algorithmically, LOL. It is generally accepted that you need lots of links to make a difference in SERPs for any meaningful term. The more links you've got the easier it is to spot three way linking. If you are my competitors please continue chasing these recip and three ways. ;)
Oh, and BTW everybody, I'm in pedantic mode. A reciprocal link is a link that's exchanged. "Reciprocal link exchange" is tautologous, annoying, and even more guaranteed to get your email straight into my delete folder.
Oh, c'mon, somebody say something about "more guaranteed" :)
In our experience the "purchased" links worked for about one month then fell from serps. I experimented with it on a couple of under performing pages... things went well, links from very on topic sites.. it worked very well for a while .. even causing those pages to "rank" higher than our home page according to webmaster tools... then poof... less traffic ... and drop in rank. Took about 30 days.
I think google is intelligent enough to know who is selling links and it probably just takes a googlebot cycle to figure it out.
| This 44 message thread spans 2 pages: 44 (  2 ) > > |