Hello. I have a question regarding my site in the SERPS. I know that having keywords in your links text is important. Until most recently, I have been targeting the following keywords: professional genealogist (s) genealogy researcher (s) genealogist I was not getting much traffic. So, after the last Google update I decided to target an additional keyword: family history. According to Overture, this keyword is searched a lot more often each month. By getting backlinks with the link text "family history" I am reducing the overall percentage of links that use the old keywords. I just starting slipping in my rankings for the old keywords. They were steady until I started getting links with the new keyword "family history". Do you think these new "family history" links are responsible for my site dropping in the SERPS for my old keywords? For example previously: 45% of my text links were "professional genealogist" 45% of my text links were "genealogy researcher" 10% of my text links were "genealogist" Now, they may becoming: 30% family history 30% professional genealogist 30% genealogy researcher 10% genealogist So, is my site considered not as "strong" as it once was for the old keyword. Therefore, my site will drop in the SERPS for that keyword? Or, perhaps this makes no difference at all. Maybe, each keyword is taken into consideration individually to determine the SERPS. Can a site rank high in the SERPS for many keywords? What if the site ranks very well for the allinanchor for many keywords? In other words, by increasing one am I decreasing the other?
I've never heard of such a thing (an increase in one causing a decrease in others because the "%" of anchor text goes down). it's possible they have something like this, but very, very, very unlikely. besides, if this ewas the case, it'd be easy to make competitors drop like a rock.
Hi. Well, as a semi-newbie I just didn't want to do anything that might hurt my rankings for the keywords I'd already established. The theory is my own. I was thinking about how Google determines theme. What link text people link to your site with must be what you site is about. So, perhaps they look at your anchor text to help determine it's theme and also what keywords best represent your site. Sincerley, Melaniejk http://www.ancestralbranches.com
Melaniejk, I have a few sites that rank highly for multiple keywords/keyphrases. One ranks #1 for over 50 different keyphrases (albeit less sought after phrases for many of them). I don't believe for a minute that ranking highly for one will, by itself, decrease the value of another. Be careful of dilution on the page content (many thread postings in DP on this).
There is a natural tendency to see cause and effect in every thing that happens. We all fall into the trap. What I try to do is stand back and apply logic. Lets do that to this question. 1. Allinanchor: search looks for the number of links with a specific keyword or phrase used as anchor text. It is not a measurement of percentage mix of anchor text used with a given URL. It is an absolute measurement of the number of links with the specific anchor text used to seed the search. 2. Assuming -- and you never made this point clear -- that all the links using "family history" are new, or additional, links and not replacement of the other links, then they have had no effect on the allinanchor ranking of the old or existing links. 3. Allinanchor and SERP placement do appear to have some correlation, but allinanchor is certainly not the only only factor affecting SERP placement. 4. You do not reported what your allinanchor rankings for you old keywords where and what they are now. It is possible that while you have been busy placing links for "family history" your competitors have been placing more links for the old terms and as a result your allinanchor reanking has fallen for these terms. Not because of the percentage mix, but simply because your competitors now have more links using the old anchor text phrases. 5. You also mention page theme in relation to this situation. That is too big a subject to get into here but I can't see how "family history" would not be thematically related to your page.
It may sound like a bit of a job to do now that you have put in the work but if you are trying to rank one page (your homepage) on 4 different searches I would take a different approach. You want to rank on 4 searches? family history professional genealogist genealogy researcher genealogist I would focus your efforts on your high traffic search on the homepage (one phrase per page). And set up internal pages to target the other three phrases. You can still get links to your site on all four phrases, but you would ask for the links to be to the respective pages. I would focus moreso on theming pages, than theming sites as your 4 are all on the same topic / theme anyway. If you want to look more into theming sites I would look at throwing in some semantic content (search this forum for 'semantics' if you don't know what I mean.
Are you sure about this? I have yet to find a definitive answer to how the allinanchor search results are sorted. I did not think it was simply sorted based on the number of google indexed pages containing the anchor text being searched. I thought many other factors must come into play such as the pagerank of the linking pages, the density of the search term within the anchor text, and probably lots of other things as well.
I could never find any reason to the way they're sorted, myself. although yahoo tends to show similar ones to google towards the top.
First the name of the search tels what is included. It is a search for all the links with the particular keyword phrase in the anchor text. Whatelse can it be? Now does the order of the sites in the allinanchor SERP indicate the site with the greatest number is necessarily in first place? I wasn't at all sure of this until I started to work on the McDar experiment. But go and use McDar's tool. There a very strong correlation between the sites in the top ten for any given search term and the sites with the highest allinanchor ranking. So yes, I'm satisfied that the placement in the allinachor SERP is indicative of the site with the most links using that particular search term, or keyword phrase.
Google defines it this way: So it looks to me like the all part is just a modifier to insure that all the search words are in the URL. It may be a fine distinction, but it really does not say that it includes all the pages with the terms in the URL, just that the results will all have the search terms in them. IMO it may be that there are other terms that influence the positioning of the results of the allinachor search besides the number of links.
It could be any number of factors and/or combinations of them, but there seems to me to be no direct correlation between only the number of anchor text links that contain the terms and the positioning in the results of the allinanchor search. For instance, suppose you do an allinanchor search for the term search engine optimization and have two pages that both have ten links containing the term pointing at them, but one page simply has the link text search engine optimization pointing at it and the other has the link text expert search engine optimization pointing at it. Both sets of pages will be returned in the allinanchor SERP, but it seems logical that the one with the closest match and the highest density should be positioned higher. This factor alone would mean that there is no direct correlation between the number of links and the positioning. Certainly the page with the search engine optimization text only is the more relevant of the two pages. Note also that a page with any number of links with just the anchor text search engine in them will not be returned in the search engine optimization search no matter how many links there are, so lets add another page that has nine links with the anchor text search engine pointing at it. Now given our three pages what will an allinanchor search for the term search engine return? Will it return the page with the nine links containing search engine first or will it return one of the pages that have ten links that contain the term search engine but also other terms? And if one of the pages with ten links which one first. Next I suggest that we have to look at the intent of the allinanchor search operator. Is it just provided as a search operator to count the number of pages that contain the search term or is it a search to determine also the relevancy of those pages that contain the term? Google is silent on this as it is on many things that relate to its algorithm, but I tend to lean towards the relevancy side of the fence. Certainly search engines value relevancy above anything else, and that is enough to tip the scales that way for me. For instance why could the allinanchor search not simply select all those pages that contain the term in the anchor and then rank those pages with the regular Google algo? Evidence Bob? If you mean hard evidence, I have never seen any hard evidence with regard to search engines, just analytical evidence supported by examples. I have no ready examples but the above reasoning and a gut feel acquired from a few years of running such searches and looking at the results. If we want better evidence, then it might be a good idea to set up a site to test the results of an allinanchor search as theorized above.