Welcome to WebmasterWorld Guest from 188.8.131.52
Forum Moderators: open
P.S. Anchor text is not a four letter word and is still a major part of the algo. Just be use variations and creat natural looking linking patterns.
As in "gold"?
Anchor text is still the most important part of the algorithm. Removing releveant anchor text is slitting your own wrists, and also an example of the pointless seo that is not google-friendly. Link to you red widgets page with the words "red widgets". Your users will like it and Google will like it.
Took it down and phrase seems to be coming back. Can't say if it was algo tweek or removing it. The effected site has limited backlinks so that may make a difference also.
Take a look at the sites out ranking you. Look for the size of the site, the sources of their links being on-topic, and links off to other sites (including sub-domains as other sites) Just a place to start.
Anchor text is the first thing you think of, but as you delve deeper, you'll see that it is not the problem.
Why should multiple votes for yourself count for anything?
Because other people have given me the right to vote for myself.
There are enough people linking to my site to make it a high PR6 on the home page. There are about 8000 pages. Not every page on every subject has external links coming to it, but there are so many people that think that the site in general is good that they are in fact voting to a certain extent for *all the pages* on the site.
Another site that is fairly similar in that regard is webmasterworld.com. The threads are full of great information, yet very few of the incoming links are to the thread level, yet the internal link text does give you a good idea about the topic of the thread.
I hope that helps answer your question.
Why not start by looking at the sites that are ranking well? Find common characteristics and look into changing your site accordingly.
For our market the answer would be:
- Completely not related to the search term
- Sites that have the words blue and widgets in them but have absolutely nothing to do with blue widgets.
- Directories that contain links to our site or have the words blue widget in them
- .org and .edu that have blue and widget in them that may or may not have anything to do with blue widgets
- sites in all Japaneese (doing a US search on wwww)
- site that was removed many months ago - dead link
Guess I should make the site about blue apples and then mention the word widget once but don't think I would sell anything.
The site is a two keyword url, spam fulled keyword density, spam fulled title
Spam fulled header.
In actual fact the site is so spammy "optimized" people here would comment on it.
Today page 1 number #1, from 2.7 million results
Probably won't last, I don't care, just a $500 bet I could do it within 2 months, lol..5 days was enough.
G I luv G
I have a 12,000 page site with 3,000 concerning one product. Before Austin, for a year the 45 money pages have ranked 1, 2 or 3 for the 2 keyword search most use. Often I was at 1 and 2. With Austin they were below the top 100.
I compared with the newly ranked 1 through 4 sites. Often it was two pages each from the same two competitors. One has over 300 backlinks and the other 50 or more with no links outside their domain. Mine has 12, so there doesn't appear to be a penalty. First I noticed no keywords in many of the their meta tags or titles. Then a low concentration of the 2 keywords in their heading and text. I didn't consider anchor text at the time. I wanted to make some small changes to see if they would make a difference.
I cleaned up my keyword meta tag, probably 30 words. Reduced to 10 including including the money keywords. Left the title as is and made sure the keywords in the text were consecutive and reduced their number.
In 2 days, Google indexed 6 money pages and all moved to the first page, usually followed by another related page from my site. Looks like I had been penalized for my keyword concentration. Now comes the interesting part. One of my competitors was consistently ranked just above my pages.
Now I had something to compare my site with, so I started looked at keywords in anchor text. I tried specific NON-money keyword pairs that might be in both sites until I came up with a pair where he was one spot above me. I used related non-money keyword pairs thinking it might narrow the search for a page that would indicate what Google likes. My site is on topic and his is concerning a completely different product that before Austin would have been on page 1000, if at all.
My site had both keywords, one in the title, keywords, description, heading and text and the other in the text. Only one was in anchor text. His had both keywords only once, but each was in anchor text. One was a single and the other was part of a anchor phrase. Neither was anywhere else on the page or in his code.
It appears that keywords in anchor text ranks very high with Google. Maybe there was something else in his site that I missed, but it's a simple html page with java script generating an option list and some text links.
I found this situation in several other instances, so I'm uploading my site with the money keywords included in anchor text. I probably should have before now, but why make changes with the results I had before. I expect Google to index the site completely within the next week, so we'll see what happens.
Excuse the long post, I'll get better at this.
Obviously they should. Large domains have hundreds or thousands of pages.
They aren't all equal, but they will often be on the same theme, so internal links tell the bots which is more important.
The other clear example is I have some three page articles on a site. Obviously I have dozens of links to page one of the article and only two to page two and page three. The bot needs to know that *I* consider page one the more important place to start with the topic content.
Lastly, it is plainly clearly that what all search engines need to do now is completely ignore whether a link is "external" or "internal". Domains can be bought for a few bucks. It makes no sense at all to value external links from the same family of sites over internal links. That would be goofball search engineering. It's just one big Internet of pages.
Internal links show depth, especially when they are going between related pages. If Google can see related themes internal links can help them find large sites on one general topic.
Also internal links help visitors find their way around. After all sites are supposed to be for the visitors not just Google.
IMO, internal links and anchor text will soon be completely discounted.
Why should multiple votes for yourself count for anything?
Because the internal links and anchor text are likely to be descriptive.
If I have an article on Shelbyville ob der Tauber, wouldn't it be reasonable for my Elbonia articles index to list that article as "Shelbyville ob der Tauber"? And for Google to think "Gee, that article is probably about Shelbyville ob der Tauber"? Especially if Google's textual analysis of the article suggests that it may indeed be about that fine Elbonian city?
Of course, if Google's textual analysis yields a result of "Hmmm...the keyword distribution in this article seems artificial," or "The presence of 50 keywords in the title and a triple-hyphenated domain name look suspicious," or "The terms 'Viagra' and 'debt consolidation' appear more often on this page than the term 'Shelbyville ob der Tauber' does," then maybe the benefit of the internal links and anchor text will be diminished. :-)
In my experience, using descriptive anchor text for internal links won't hurt (and will usually help) Google rankings--at least if your pages can pass muster with whatever SEO "sniff test" or filter that Google may be using.
OK, I'm a Topic-Sensitiveist ;-)
PR is/was a "value" rank based on the inter-connectivity of a page, one link, one vote, with no attempt made to vet the voters.
So we had internal inks, off-topic links, FFA links, and whatever-you're having- yourself links, all counted equally to the single on-topic link from a relevant outside source.
Very democratic, very idealistic, but with no effort made to check voter identity, it resulted in the massive electoral fraud that is PR today.
Well not any more.
G is switching from Universal Suffrage to a qualified voting system based on Topicality, and the only internal links likely to count would be from those pages returned for a particular search query.
Even then, "affiliated" status could well mean those votes get excluded.
All strictly IMO of course ;-)