| 2:50 pm on Jul 13, 2004 (gmt 0)|
Make sure the site selling links capps the number they sell per page. Nothing looks spammier than a box of "sponsored links" with 30 or so unrelated links in it.
Try to be at least vaguely related to topic of the site.
|Watcher of the Skies|
| 3:49 pm on Jul 13, 2004 (gmt 0)|
So, is a nice text-inside-a-banner-ad with 20 related, different content links in it ok?
| 3:50 pm on Jul 13, 2004 (gmt 0)|
Write a small paragraph and incorporate your link within.
| 3:53 pm on Jul 13, 2004 (gmt 0)|
Mix up the anchor text.
| 4:26 pm on Jul 13, 2004 (gmt 0)|
>>>So, is a nice text-inside-a-banner-ad with 20 related, different content links in it ok?
Having 20 links defeats the intent, which is about receiving PR, and the benefit of anchor text.
On the other hand, you can disregard the PR altogether and just harvest anchor text.
Will sitewide links raise you above the radar? I'm not sure about that.
| 5:07 pm on Jul 13, 2004 (gmt 0)|
|Will sitewide links raise you above the radar? I'm not sure about that. |
Depends on how many pages, if it's in the hundreds or thousands it seems to trigger the "sandbox" or "quarantine" penalty. Don't have a good handle on where the number is yet.
| 8:01 pm on Jul 13, 2004 (gmt 0)|
|Write a small paragraph and incorporate your link within. |
Hmmm, what a gem that is. ;)
The benefit of anchor text and at the same time, the benefit of proximity. Makes for a winning combination, wouldn't you say so?
| 10:42 pm on Jul 13, 2004 (gmt 0)|
So, what kinds of things will put you over the radar?
Seems like a low profile pr 8 is better than a high profile PR 8, if it's PR you're after.
High profile website: Yes? No?
| 10:53 pm on Jul 13, 2004 (gmt 0)|
Martinibuster - that all depends on the reason it's "high profile" in the first place.
Over the radar
Links from sites that are known to be selling links for PR value or SEO.
Mass (same) anchor text inbounds from "homepages".
Under the radar
Deep links from authoritative on-topic resources.
Within radar range but known to be acceptable
Links from sites which are known to be selling links for advertising purposes, not blatantly selling "PR" or "SERPS manipulation".
[edited by: trillianjedi at 11:00 pm (utc) on July 13, 2004]
| 10:56 pm on Jul 13, 2004 (gmt 0)|
|So, what kinds of things will put you over the radar? |
Buying links from folks announcing the fact that they are selling links with the sole purpose of transfering PR.
| 11:14 pm on Jul 13, 2004 (gmt 0)|
Make sure you don't link from a PR9 to a PR2.
Eg, incrementally increase your PR. Start with PR5s and then move on to PR7/s, then 8s, then 9s .. etc.
Probably the best advice though is to avoid buying PR links based on PR.
If you buy them for their traffic, Google will probably even reward rather than penalize you.
In otherwords, you'll be flying under the radar if you act ethically (but smart).
The ethical part is the buying based on traffic. The smart part is paying a premium for high PR, good positioning, and custom anchor text.
| 12:02 am on Jul 14, 2004 (gmt 0)|
I have bought links with a varying amount of PR and have never had a problem with them linking to low PR sites.
The "radar" seems to come in to effect when sitewide inbounds occur (regardless of PR).
| 12:13 am on Jul 14, 2004 (gmt 0)|
|The "radar" seems to come in to effect when sitewide inbounds occur (regardless of PR). |
I still feel there a "magic number" involved. I have seen site wide links from a PR6, total of 6 pages take effect right away. However sitewide from PR4-5 with hundreds of pages get quarantined.
| 1:14 am on Jul 14, 2004 (gmt 0)|
There comes a point when no matter what you do, because of either a) the position of your site, b) the authoritative status of your site, or c) the market that your site is in, or all 3, it is almmost impossible to stay below the radar.
Are we talking about flying under the radar of bots or of reviews by competitors and editors?
| 8:53 am on Jul 14, 2004 (gmt 0)|
Blaze, Good point. The bought links should blend in with your natural links ie a site with 20 pr4 incoming and then one pr8 may attract attention.
agerhart, human review for link spam (from competitors) is an issue. I think deep linking your incomings would prevent them finding the 'suspect' links and prevent a spam report to google.
Some good points made in this thread, cheers
| 9:46 am on Jul 14, 2004 (gmt 0)|
Can we even assume there IS a radar until Google's position with text link advertising clarifies?
| 11:43 am on Jul 14, 2004 (gmt 0)|
To assume there is, even if there is not, would be the sensible option.
|Watcher of the Skies|
| 11:49 am on Jul 14, 2004 (gmt 0)|
Oh, what a tangled web we weave....
Seems to me that if Google could've kept the details of PR 100% confidential and under wraps (ok, impossible) that it might have been a good reflection of the web. As it is now, they have, in succession:
1.) Implicitly encouraged everyone to spend very valuable resources acquiring links...
2.) But penalized, progressively stronger, linking to vague "bad neighborhoods" ...
3.) Which has resulted, just for one example, in a surprisingly large number of websites simply pulling the plug on their outbound links either by decimating them, or having their link page(s) conveniently not load.
So a company that, philosophically, would profess a desire for "natural" linking has inadvertantly caused much of the commercial web to be linked in the most unnatural ways at both extremes. Now, they're in a bit of a conundrum where they don't want people to sell PR but "naturally" a text link is not only intuitively ok to all casual observers, they often look nicer. Certainly the retro, green, hip Google would prefer a simple text phrase to a flashing neon banner. Of course, if they want to offer Adwords, THOSE text links are ok. Hmmmm.....
Bottom line? I think they can only, sadly, encourage and not refute various rumors of penalties when, I suspect, far fewer exist than many of us seem to think. If it IS the case that a few text links out are bad - paid or not - then everyone's in trouble. Seems they're caught a bit in their own trap. The sooner they admit that the effect of PR and links are being minimized (save the one OVERRIDING factor of anchor text), the better, or they risk looking a bit foolish. Not to explain it to US, but to the underwriters. Of course, there's really no way to catch people selling text links if they're NOT in banners - damn, this is what they WANT, an integrated web.
Now, where's my glass of wine...?
| 12:16 pm on Jul 14, 2004 (gmt 0)|
*To assume there is, even if there is not, would be the sensible option.*
It would certainly be the safer option, but I've seen little to suggest the Big Two are now taking a jaundiced view of Text Link Advertising per se.
Of course if it's badly done, and gets pinged as a "links scheme designed...." that's a different story...
| 7:28 pm on Jul 14, 2004 (gmt 0)|
What we learned:
- don't do site-wide links > 200 links
- pages with 200+ links of the same anchor text will get problems, especially if they are "young" ones ... e.g. not very long on their PR levels
- site-wide links cause the "display URL-only" problem in both linking directions (linking and linked to sites get penalized!)
- if site-wide links are removed, the URL-only page dying seems to stop within days
| 9:23 pm on Jul 14, 2004 (gmt 0)|
I can tell you almost conclusively that buying links from sites which plaster your link on all of their pages (thousands) WILL trigger the sandbox/quarantine penalty. I have had it happen to two sites, both PR7s, both with similar inbounds. I know that some will say that seeing that on only two sites is not conclusive, and I agree, but I will also never buy those types of links again. On one site, we were ranking number one for most of our terms, I bought some of those High PR links (the newspaper type) to bolster us through the holidays, and bam! Gone like a rocket with a gray bar. The next site had great inbounds and was moving up, then BAM, two weeks before the published link update, that one was gone, too.
I also believe that there is a lot to be said about too many inbounds with identical anchors. Does it look like "natural" linking when you have a few thousand inbounds with the identical anchor? How about when the anchor is a highly competitive keyword? I think so! This would not be hard for Google to filter if you think about it. I also remember a post somewhere in here that had a breakdown by percentages, like never have NN% of you links with the same anchor, etc...
Another Note: We had gone from a published link count of 900 to 12,000 in one update: Could this also trigger a penalty? Most likely. Watch out for that!
| 8:24 pm on Jul 15, 2004 (gmt 0)|
So I guess the question that arises is whether or not it's now possible to torpedo a competitor's rankings by buying them a bunch of links.
| 8:46 pm on Jul 15, 2004 (gmt 0)|
That has been my question for a year! A lot of people say "No, Google wouldn't allow for that, it's too easy, etc...", but I believe that yes, to a certain extent, that this could be possible. Case in point: We all know that Free-For-All and guestbook links aren't looked upon nicely by Google, so how would they know if it was you who added the links or someone else? As far as their algorithim or filters are concerned, there isn't a difference.
I believe that this is one of many flaws with relying completely on the link "vote" system. There must be other filters aside from just the ones that are discussed here, but unfortunately you don't know until you get torpedoed!
I can see the day when webmasters send this to Google: "The inbound links referenced below were not acquired by us so please do not penalize us for them."
[edited by: martinibuster at 8:57 pm (utc) on July 15, 2004]
[edit reason] Consolidated two posts [/edit]
| 8:53 pm on Jul 15, 2004 (gmt 0)|
|>>>So I guess the question that arises is whether or not it's now possible to torpedo a competitor's rankings.. |
>>>I believe that this is one of many flaws with relying completely on the link "vote" system.
>>>The inbound links referenced below were not acquired by us so please do not penalize us for them...
Not really. In general, it's who you link to that matters- not who links to you.
| 9:01 pm on Jul 15, 2004 (gmt 0)|
|Not really. In general, it's who you link to that matters- not who links to you. |
Let's say that my competitor ranks for "red widgets". I pay two sites to link to him using the keyword text "red widgets". Each of these sites places the link in the same spot on each of their 100,000+ webpages.
Based on everyone's claims, this would have a negative impact on my competitor's website for the term "red widgets".
Unless I'm misinterpreting the situation.....
| 9:09 pm on Jul 15, 2004 (gmt 0)|
|Let's say that my competitor ranks for "red widgets". I pay two sites to link to him using the keyword text "red widgets". Each of these sites places the link in the same spot on each of their 100,000+ webpages. |
I'd say that would cause some problems for the competitor. My understanding is that there are many things you can do to harm a competitor. I'd rather not discuss them, but the above I believe is one of them.
And then again, it could backfire on you and your competitor could secure all the top spots because of your actions. ;)
| 9:13 pm on Jul 15, 2004 (gmt 0)|
I have sitewide links into several websites (not purchased), and have had no problems.
For the cases where sitewide links caused a penalty, I can see how several thousand sudden inbounds might raise eyebrows, but there can be other factors in play- having to do with things done months before.
For the above cases to be conclusive, the purchase of ROS would have to be the only iffy things done with the websites. Now, I'm not pointing fingers, honest, but where there's smoke there's ususally fire.
Certainly it's conceivable that a ROS could cause a penalty (and so are many other theories), but in my experience I haven't seen it happen.
| 9:17 pm on Jul 15, 2004 (gmt 0)|
|Sorry, but are you referring to the TWO posters who bought ROS links? |
| 9:25 pm on Jul 15, 2004 (gmt 0)|
I agree that it isn't the smoking gun when it comes to recent penalties, and other websites with links on those same ROS sites have seem to escaped what we experienced, but it seems so easy for Google to pick up on purchased links this way. I personallt believe that it would involve a few issues:
Your site shows 10,000 new backlinks from a single domain within one spidering of the publisher site. Google would now:
A)Look at the link text of the links in question: Are they all identical? Is the anchor written in natural language or is it just a competitive keyword?
B)Look at what surrounds the link. Other, unrelated links?
C)Look at other inbounds to your site to see if they also have this pattern.
If so, then maybe this could trip the filter. Maybe the others who seem unaffected by this just haven't been affected yet. Maybe it's just a crackpot theory. Either way, I still believe that this would be effective in reducing purchased inbounds, especially from large sites with thousands of pages. Google sees these ROS links as detrimental to thier ability to properly rank, and I am sure they know about them. How else would they deal with what they see a shortcut to high ranks?
| This 38 message thread spans 2 pages: 38 (  2 ) > > |