Some Light Shining on this New Update

After a lot of research I've come to a couple of conclusions and I'd love to hear some responses on this.

First off, has anyone else noticed that the allinanchor:keyword is matching almost 100% with the keyword result?

It's as though google is still working hugely on anchor text, but if a link is deemed "crap" by them it won't count for your allinanchor. I sooo hope this is true!!

get your allinanchor up, and you'll rank number one. But, and I mean big BUT...

Getting your allinanchor up is not going to be as easy as getting sitewide links and link exchanges. I'll bet that link exchanges count for very little towards your allinanchor but will still count... I have one site that still ranks okay on it's subpages but not as well it is 100% link exchanges from the last year.

When I went through 200 link pages (meaning the page where our link partner links to us from) I found that the average number of links is over 50. Which means to compete with a site that has 5 article exchanges (meaning their link is the only outbound link), you'll need 250 link exchanges just to keep up.

Now here's the bigger problem, in order to get those 250 link exchanges within a month, you're going to set off the dreaded sand box. Bye Bye rankings. It can't be done anymore, a site that gets 5 article exchanges where their link is the only outbound link is going to be miles and miles ahead of a site that has 250 link exchanges.

If they get ten articles written about their site you'll need 500 link exchanges just to keep up.. and if you're supposedly only allowed to get about 5-10 links a week to not set off the sand box... you're going to have to wait about 100 weeks for the same effect as having 10 unique articles written about your site.... this is starting to explain why people are sandboxed for over a year.

One way links that are in articles where your link is at the top of the page will help your allinanchor... a one way link at the bottom of a relevant page won't help at all.

The sites that I manage that have dropped are link exchange sites, link-vault sites and one way link exchange page sites. The last one is meant as three way linking.

Ones that are succeeding have articles written about them and that's it.

Get links from pages that have only your link on them, reciped or not...brand new pages help a lot once they're cached and your rankings will greatly improve.. as to how to do that.. that's your job not mine.

So why were the dmoz links dropped in value?

All the wonderful links you get from dmoz are no different then links you get on someon's links page. They're a whole bunch of links so the value gets devided, just like PR does. If there are 50 links on a dmoz page, the value of that link gets devided by 50, the sites that have more pages with only their link on it, will rise above. If you've been a stickler for pages that have only 10 -20 links on it, you will have done better in this update than others.

If dmoz wrote you individual articles for each site that gives you a link and your link was the only one on the page, yeah, you'd be high up again.

Two hours ago I hated Google, I thought they were stupid, but now I think they've finally done the right thing. No longer in the google rankings will there be sites that just have more links than other sites.

okay let the flames begin. If anyone has counter explanations that make more sense I will gladly admit that I'm wrong.

This is truew for many keywords for me, however not for "mortgage broker bonds", I am #1 allinanchor, but 170-190 in the DCs. Hopefully this is still working itself out.

The same page has #1 for allinanchor and in the serps for "mortgage broker bond". This is something that has perplexed me, as the IBL and internal anchor text varies almost evenly. The only thing I can think of is that the plural version IBLs are crap, where the singular IBLs have some "valid" links. I guess I'll have to wait to see when all the DCs match and where I'm at.

This seems to be true with my observations as well. I was hit by the update, but i am still ok with it. It seems G is finally doing what they always claimed they would.

May or may not be entirely true, see above.

Good observations, too many links on a page dilutes the value for all, makes sense and is something we all always suspected. Now it seems these links are diluted more than ever.

While I believe there is a difference in value, saying that a link at the bottom of a relevant page won't count at all is a just wrong.


Wow you just described my site!

I was at the top of the industry prior to the update. I do not think I was penalized (which I was thinking yesterday before a good nights sleep). I think that many of my IBLs were devalued bringing my site down from where it previously was.


Good to know...It's funny I was racking my brain on how to obtain links by purely white hat means last night. I didn't realize how much I was leaning towards black hat until I couldt think of many good white hat techniques!

Still all I have been able to come up with are

-Articles
-RSS Feeds
-THEMED recips or three ways will work for some, but not for my industry as it is too small/competitive in nature (good thinking on being one of the few links on the page)

Agreed.

Last night I was furious, howver now I have cooled off and taken a closer look at my site. I am back where I was prior to the superbowl update, but I'll climb back. I dont THINK i am penalized, just my IBLs don't count the same as before, i can live with that.


Anyone who flames you for this is a moron, you get all my rep for the day. This is the first post I have seen that is very close to actually seeing what this update is about.

Thanks for sharing your thoughts with us all...

This is the start of a good theory - it hits you in the gut as being correct, or at least very close to the truth.

However, there is one glaringly odd dichotomy in the logic. I don't mean this as a flame in any way, but I do feel it's important to note. Like I said, I think this is very close.

You begin by stating:



but the rest of the argument expounds upon which links are better, like so:



Now, I haven't verified any of the statements you made, but doesn't it follow that since the allinancor: search simply considers the volume of links to a site containing that phrase, and the allinanchor: searches match the regular SERPS, that an exchanged link would count just as much as a one-way?

In other words, this theory cannot be correct unless a) the allinanchor: search does not match the regular SERPS, b) the allinanchor: search is not based purely on volume, but is instead weighted as you suggest or c) there is no weighting of any particular type of link in the regular SERPS.

Now, if c) then the entire theory is shot to hell.

If b) then the theory becomes pure speculation, since there is no correlation between the presumed known constant and the delta.

If a) then the theory gains some real merit.

Again, not trying to shoot you down - just my early-morning thoughts. Overall, I think you're close to something and I suspect we'll find that the allinanchor: searches really don't match the SERPS.

Let me take this a step further.

The top SERPs also depend on the creation date of the whois profile and the date the pages were first indexed.

Have a look at your top serps for your keyword. I can't find any sites registered after 2003. Most of the sites rank from late 90s to 2002.

It's almost like Google has gone and sandboxed all tough terms past 2003 again.

Obviously the terms I'm looking at are pretty competitive but all the sites I have, my pre 2003s rank fine, and my post 2003s seem to have taken a few hits.

Anyone else seeing that date is a massive factor now?

Hexed

I noticed the same thing immediately, see: http://forums.seochat.com/t53827/s.html


however, I am not sure if it is a coincidence or if there is a sandbox scale now rather than being just in or out. You can be deeper in the box the newer your site is.

I plan on doing a lot of rss feed submissions and articles to see if I can crawl my way past my competitors again. I'll keep you posted if I make it to where i was pre-update.

SO what do you think take down LV and the one ways and hope for the best...OR sit back and wait and see what happens

also if this is the case about links look at number 1 and 2 site for

Dect Consolidation expplain that. Links are bought 1 way links and link exchanges

Leave them up for now...They won't hurt (I don't think) provided you don't have outgoing LV links as well.

It appears they simply count a fraction of what they use to. They do appear to have some value though, as some LV members have sites ranking with LV links only.

so maybe I should take down all bought links that are on bottom of pages as SPonsored links?

I am still seeing a few highly competitive phrases that the allin is not matching the actual rankings. Actually, come to think of it, our site is ranking a bit lower for our phrases (not as competitive) than our allin as well. Neither of which is due to the sandbox.

I think a lot of times people come up with the cause or most important factor, before looking at the sites or numbers and then find "evidence" to support their ideas. The other thing is that the differences in the niches might lead to different conclusions.

I think taythin has this right on. Ive spent the better part of the last two days looking over sites, mine, my competitors, seochat members and gathering information, trying to figure out a pattern. This is what ive come up with so far:

1. A few of the domains we bought, we moved to a new server/ip and made changes right away. these sites had a small drop in traffic. The ones we changed server/ip left alone for a while then changed around seemed fine. The filter for site changes seems to have tightened. Dont add too many new pages all at one time or make drastic changes. That might send up a flag.

2. We hand write some of our content and some is dynamically generated. the dynamically generated content pages seems to have taken a hit. this is obviously probably to take down scraper sites. unique hand written content seems to be even more neccessary.

3. Links. This update seems to have a lot to do with links. so this one will be broken up.

a. recips - i think recips are now worth less (not worthless). The staple for recip links is a page called links.html that has 50 outgoing text links. these are now less valueable. themed recips inside some type of content can still have value. i dont suggest you quit reciping, i suggest you only go for recips that are of some value to your users.

b. article links - i think these have gained value. links inside content that are one of only a few outgoing on a page. the content is relevant to the link anchor text and the title of the page its pointing to then you have a valueable link.

c. rss links - if you have an rss feed and people are using it then this falls in the same category as the articles. themed links inside content pointing to a themed pages. value added.

d. sitewides - sitewide links IMO are not much better than 1 link. i believe G may only count 1 link from 1 domain as having value. I could be wrong.

e. internal links - i think that internal links have been slightly devalued too. if the internal links are very relative then they help, but if its a "blue widgets" page and your internally linking to "red thingamabobs" that might not do much to improve the pages power.

f. any links - i dont think that any link can hurt you. they can just have little to no value. googles goal is to make their results unmanipulateable. they do this by making factors that we cant change(yet) like link age, site age and whatnot.

This is all of course what ive determined based on research over the past two days

Theres alot more that i theorize about links but id be typing for about another hour if i explained it all and would probably confuse other people or myself. My best advice?

Get links inside content paragraphs with anchor text relevant to the title of the page you linking to. surround that anchor text with relevant content and words that produce a high c-index. i believe google not only analyzes the anchor text but also the text around the anchor text to determine relevancy. Good luck with the new update and remember not to do anything drastic. come up with a new plan and execute. weve all survived past updates and im sure well get through this one.

if you dont know about keyword relevance and c-indicies as explained by orion i think you should:

http://forums.searchenginewatch.com/showthread.php?threadid=48

ok, ive typed too much, sorry if its a long read.

http://www.google.co.uk/search?hl=en&q=number+plates&btnG=Google+Search&meta=

If you look at the above link, our SERPs are all over the place. 2 of our competitors are seeming to have taken the top 4 pages with subpages.

How can this be an improvement?

Especially the 2nd result which is a spam page.