dickbaker when did the rankings change?

It's been in the last couple of months. When I started reading the threads about the Caffeine update, I started to check, and was surprised at what I found.

Most of my rankings are intact, but a couple of important ones dropped. One went from #6 last summer or fall to #10 today.

Have you analyzed RECENT growth in backlinks, including the strength of the linking page? Sometimes that factor - a new spike of interest - can trump the mere historical number and strength of links.

No, Ted, I haven't tracked backlinks for the competing sites, if that's what you mean. Some of them came out of nowhere to be on the first page.

One phrase I lost position on was "[name of a state here] widget stores." I was #1 and #2 for that phrase for probably four years. Now I'm #3 and #4. At #2 is the site for a widget manufacturer, but they're not based in that state, and there's nothing on the home page of their site, which is what's shown in the results, for that particular state.

Are you sure you've lost the rankings on a global basis, or are you just going by what you see in the serps when you search from your location?

How are you checking your ranking from other geo loations?

My reason for bringing up recent backlinks is that Google does watch backlink trends. Long-time links can fade in their power under some situations. And even though you have more than the competition, having more links that are recent can make a difference, so getting some more BL activity still might help you. Link juice is not just a numbers game, or a PR game anymore.

I sure do see that your situation would be frustrating, especially if it goes on for a long while and the newly ranking sites are less relevant.

Just a guess here. There were several algo factors that Google mentioned in past patents (beginning with the historical factors patent in 2005) that sounded like good ideas, but rarely seemed to be in play. It seemed to me that these factors would require a lot of computing power to be practical. With Caffeine, Google may now have enough power to phase in some of those factors more prominently.

As I said, this is just a guess, but it is an idea that's been coming up for me more and more lately as I do SERP analysis.

Ken_b, I have remote access to servers in other parts of the country, so I can check search results there.

Tedster, it does seem like everything is almost purely a guess right now.

I do know that I'm getting new backlinks, although it's not usually by me trying. It's people linking to my site through forums or blogs on their own. Maybe the little site in the #1 spot is getting links faster.

Here's another instance that illustrates the unusual search results:

One of my pages was ranking anywhere from #3 to #7 for a particular phrase. It's now at #9 Above it in the #1 spot is the manufacturer, and in the #2 to #4 spots are big, well-known online stores that sell this product.

In the #6-#8 spots are the usual "about" sites, wiki, and a YouTube video. In the #5 spot, though, is a site that every indicator shows gets very little traffic. The site sells the product. However, right on the home page of the site, the owner says that he and his wife are retired and traveling the country all the time in their RV, so they can't always return phone calls right away. They also say that is means that they can only ship on Monday, Wednesday and Friday. Everything about this site screams "hobby." I can't imagine that anyone buys from it.

What's it doing at #5?

@dickbaker

How many unique root domains are linking to your site versus the site outranking you now?

My observation is that domain diversity trumps large number of links if they are coming from fewer sites..

@tedster - interesting point ! some food for thought..

OMZen, Yahoo shows that my site has 10,312 inbound links. There's a good number that come from the same sites, especially two forum sites. If I had to guess, I'd say that there's 5,000 to 7,000 unique links. These are from blogs, directories, links from forums, and one or two-way links from related sites.

The competing site I mentioned in my last post has 43 links from other sites (according to Yahoo). The website I mentioned earlier has 271 links, again according to Yahoo.

What's it doing at #5?

My observation of the searches I monitor is that the #5 slot is sometimes used as an out of the normal flow SERP position to promote sites from the second page on a temporary basis, rotating a different site in and out about once a month. My sense is that these are pages which based on on-page factors are as valid results for the query as the pages above them, but don't quite have as much off-page push.

One of my pages was ranking anywhere from #3 to #7 for a particular phrase. It's now at #9

Google seems to have lost some focus lately. By that I mean there seems to be a tendency for broader focused more generalized sites with content matching a query to be promoted at the expense of more narrow focused specialized sites with content more limited to the query. My suspicion is it may be a reaction to the "search overload" "decision engine" ads, as one of the results of the softer focus is more recognizable brand name sites at the top of the SERPs.

But another result is worse SERPs in less competitive categories. For example one search I monitor only has five sites targeting a query "examplecity exampleservice"; three local specialists, and two regional providers who offer this service under an umbrella with many other services. Everything else in the results is only random mention background noise.

The norm had been for the three local specialists to be interchangeably one through three, with the two regional providers down at the bottom of the first page, But now the results have shifted and the two regional providers are one two, while the local specialists are number five, twenty, and one is buried at number two hundred and something.

I'm seeing the same trends in my personal searching also. Things that used to be good indicators to me that I had reached the bottom of the useful results are no longer reliable.

It isn't the number of links that matter but the quality of links. Like a link from the USDA might count for more than a 1,000 blog links.

I have a site that usually ranks around number 8 - 7 in Google for a very competitive, high traffic (though unfortunately not high paying) phrase. My site has about 1,000 back links. The site at around position number 6 has back links in the low six figure range. The site at position one has back links in the mid 6 figures (500,000+). So sheer number of back links may help but they clearly aren't the only thing Google looks at.

The rules have somehow changed, but how do you compete when you don't know the rules?

The rules always change. They search engines update their algorithms constantly. They tinker with favoring authority sites verses niche sites and back again all of the time. I have one site that almost completely drops in and out of the Google index multiple times per month. It needs some good links to stay in permanently but I haven't had time to work on it.

The best way to deal with it is to get high quality links, have multiple sites, if you can, and to go after hundreds or maybe even thousands of long tail terms per site.

Tedster is right, link value does fade and the reason why is simple. A good majority of your backlinks will be on pages within other sites that are naturally sinking in importance. An article written 3 years ago with a link to your site may be buried back on page 488 of their archives offering very little pr type value.

Checking how recent links are is important but checking WHERE those links are is equally important. To keep it simple just start at the index page and count how many clicks it takes to reach that link. Shallow, or on a page linked from the index, is desirable.

go after hundreds or maybe even thousands of long tail terms per site.

Would such an enormous feat help improve the ranking for the main KW as well. If I'm looking at targeting 1000s of Long Tail Terms, I could very well commit to such a strategy but will the effort simply result in ranking for those Long Tail terms or will Authority for the main KW be gained as well?

Using semantic analysis, you will get a boost for related keywords.

However, when you look at the sum total traffic from long-tail, and the conversion rate, you will likely decide the "main" keyword is not really important.

I have some "vanity" keywords ranking top-3, but they convert terribly. I have loads of mid- and long-tail phrases each bringing a fraction of the traffic, adding up to less than the main one, but bringing in orders of magnitude more money.

We're an ecom, and want the "Buy [att-1] [att-2] widget" types, but also have "widget". "Widget" has thousands of times the traffic, but results in very few sales. Sometimes, we pick up the main keyword "by accident", just by concentrating on the surrounding money terms

Thanks Shaddows and you're absolutely right the 'sum total traffic' is what matters. Along the way I sometimes forget that as I erroneously attribute success with ranking the main/root KW.