Forum Moderators: Robert Charlton & goodroi

Message Too Old, No Replies

Acceptable ratios for anchor text distribution?

         

smithaa02

7:28 pm on Sep 23, 2011 (gmt 0)

10+ Year Member



A site from a guy I know dropped about 45 spots which appears to be a -50 penalty. He's not saying for certain what did it, but I suspect anchor text distribution as most of his backlink anchor text links were just for the best keywords (it didn't look natural).

Did some more research and some blogs are suggesting a strong correlation between lack of anchor text distribution and the -50 penalty.

My first question is what everybody else opinion is...can a -50 penalty result from an unnatural distribution of anchor text? Also, if so what is the correct ratio?

The most detailed idea I found suggested certain percentages for different types of anchor text, such as exact keyword match, keyword variations, exact brand, brand variations, exact url, url variations (with and without www), and "useless" anchor text such as "click here".

Good advice? Opinions?

[edited by: tedster at 3:41 pm (utc) on Sep 26, 2011]

tedster

3:54 pm on Sep 26, 2011 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member



I can understand why some would find strong correlation between "unnatural" anchor text and a penalty. I don't know that it would be the exact cause, but it's clearly a signal that the backlink profile is being controlled, rather than showing a strong presence of "freely given, in-content-area, editorial" links.

The idea of chasing exact ratios is kind of like saying "how can I intentionally create the appearance of naturalness"? The answer is that you can't. With all the data that Google has, the unnatural "tells" will be there - in the types of sites that give backlinks, the relevance of those pages, etc.

The answer lies in good quality content and content marketing. The idea of controlling the anchor text is not helpful today - natural will actually BE natural.

linkbuildr

4:01 pm on Sep 26, 2011 (gmt 0)

10+ Year Member



if you look at his top anchor text, are the target keywords the to? From what I've seen, preach and practice is to have their site/brand name always the 1st or 2nd most linked anchor text.

Controlled anchor text still dominates most of the SERPs I see, you just have to definitely have a better ratio.

deadsea

4:39 pm on Sep 26, 2011 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member



Our link builders have taken the approach of letting the person doing the linking to us choose the anchor text most of the time. Some of the time we make suggestions so that we have some (20-30%) of our built links with exact desirable anchor text for competitive keywords.

To avoid appearing spammy, our link builders try to contact the type of webmasters that link to us naturally and nudge them in the correct direction, or make them aware of our offerings.

FranticFish

6:42 pm on Sep 26, 2011 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member Top Contributors Of The Month



Controlled anchor text still dominates most of the SERPs I see

Me too. The devil is in the details, but as far as I can see Panda has really just swapped one set of sites gaming Google for another set of sites gaming Google. Every time I really crunch through data for a set of search terms I still see lots of unnatural anchor text for the top sites.

tedster

7:15 pm on Sep 26, 2011 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member



Agreed - and that's why I said I think there might be some correlation, but it's not a cause-and-effect thing at all. The penalty gets applied for severely unnatural backlink profiles, but anchor text itself is not the measure. Unnatural anchor text may trigger a penalty on that keyword, but not a site-wide penalty.

deadsea

11:51 am on Sep 27, 2011 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member



It wouldn't surprise me if it actually were a penalty trigger. It would be an easy analysis for Google to perform. 99% of the time too much keyword anchor text is caused by link building. Google regularly applies penalties that have greater than 1% false positive rate.

The correlation isn't just some one off observation either. Here is an article that notes it as an oddity from SEOMoz 2011 ranking reports:

Linking Root domains with partial anchor text are more highly correlated to positive results than exact match anchor text. Very weird to find this information out and SEOMoz couldn’t explain it. Maybe Google is beginning to notice brands setting out to over-optimize for a set list of anchor text? It may be worth it to try and vary your link building keywords every now and then with partial anchor text.


[marketingpilgrim.com...]

deadsea

11:59 am on Sep 27, 2011 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member



Here is a penalty that we fell into on a perfectly legit page. Several years ago, a Yahoo representative told us that our product listing page wasn't ranking because an algorithm was detecting it as scraped search results. The page was a listing of 10 products with the title of the product and a snippet about each product. We thought that the algo must be pretty broken because the links for each product were internal links, not external links like you would find with search results. In any case, we had a perfectly legit page that ranked well in Google, but it was a false positive under a penalty at Yahoo.