Out of the Sandbox Options

[NorthStar]

FYI - I've got several sites that have been "sandboxed" for almost a year that are now showing up like they "should" based on optimization and links.


Once again it appears that patience and good SEO work.

[Black_Knight]

There have been comments elsewhere that Google have changed at least one of their algorithms in this past week, with quite a lot of speculation that Latent Semantic Indexing (AKA Latent Semantic Analysis) is at least a part of what is in play in the changes.

This does tie with what I'd been seeing, but unable to prove, regarding sites that were sandboxed.

The common feature I was noting with sandboxed websites is that while the amount of link-building varied from little to lots without changing much, those with more diversely worded links seemed the least affected. In fact, in some cases it almost seemed like artificial link-building was a contributing factor to a site becoming sandboxed.

My initial hypothesis (which I never had opportunity to test far enough to become a proper theory) was that LSI was being applied to the anchor-text attribution algorithm(s) in Google. If a site had an 'unnaturally' heavy proportion of links that all used the exact same text then it stood a higher chance of being sandboxed.

That is completely unproven, and in fact I didn't even have a single sandboxed site to test it on, so take it as nothing but a possible blind lead on what should be examined with your own site(s) to see if that might support or dismiss the hypothesis.

[NorthStar]

I believe that the sites I was watching may have had identical link text in many inbound links. However, when you are dealing with a major brand name product that is something that is hard to get around. It is natural to have the link say "Brand Name" unless you are trying to improve rankings by making the link text "popular keyword". I'll check into if I have the time.

The majority of the links to these sites are generated by placing content on other sites. Compared to most, these were high quality valid links.

So far i'm seeing radical increases in rankings for the 3 sites that I had that were exhibiting behavior that appeared to be the result of the "sandbox". All sites are optimized using white hat techniques and have generated valid high quality links (no reciprocal links). They have all been around from 8-12 months.

[NorthStar]

All sites now back in the sandbox.

[Charlie_D]

:mad: :mad: :evil:


:twisted: :onfire: :doh:




:laptop: :crash: :dumb:





:agrue: :crazyeyes: :stress:

[NorthStar]

Not to worry. It appears this is just some good old fashioned "dancing". I'm sure the new results will stick around for good in the next few days.

[Charlie_D]

Hey NorthStar

Just saw the same. This is one wild roller-coaster ride. I can't take the suspence.

[callback]

I am out of "sandbox." My high rankings that have disappeared for a year now came back (from #91 to #5). In the meantime, my PR dropped from 4 to 3.
So, why should I care PR3 or 4, as long as I am ranked high.

[NorthStar]

All my sites are out of the sandbox that were in it. That is the good news.

However, I've got some other sites that use to rank well on google that are now gone. They had relied mostly on reciprocal linking for their link pop (these weren't client sites). This has worked very well in the past few years. Anyone else seeing a demotion of reciprocal linked sites? I also haven't done anything to update site content in over a year or so on them. That might also be the cause. Thankfully I don't care about the sites anymore but if this had happened a year ago, i'd be sweating this one out.


Charlie D - Patience is the key with SEO. Keep applying the basic principles of SEO and over time you will see great results. But there may be some real ups and downs on the way.

[Charlie_D]

NorthStar

Thanks for the word of encouragement. If the serps stay this way through this dance. I will get about 500% increase in google traffic. The results are very favorable for my site. It's just not knowing if or how long that it will last that shakes my cage. But I will keep the chin up and like you say practice sound techniques along the ride.

I also haven't done anything to update site content in over a year or so on them

I have read quite a few times that SE's like to see frequently updated page content. I don't know how much weight is placed on this principle of updating pages, but maybe someone in the forums could shed some more light on that subject.


Charlie

[bragadocchio]

I'm not completely convinced that just changing content will help.

But, I will say that a blog that is regularly updated, and includes material that is timely and topical tends to do well in search engines. If your content reflects what people are presently searching for, or will be searching for, then you can get some nice results.

For instance, if you were to anticipate some events happening in New Jersey on your FunNewJersey site, in a blog for instance, and include a nice amount of text on those topics, you might be able to capture some nice traffic.

For instance, a week of blog posts from February 1st through Febraury 6th on different things to do for Valentines Day in New Jersey might get you a lot of traffic the next week.

I posted a couple of blog posts on a legal blog a couple of months ago about a long trial involving Disney in Delaware courts. There was at least one story about the trial in the larger newspapers in the country every day for the weeks that the trial went on. Those blog posts attracted a fair amount of search traffic.

So that's one way that updating content can bring you some visitors.

[Charlie_D]

Thanks for that input bragadocchio.

I have a section on my homepage where i do list some time sensitive upcoming events and always wondered what is a safe time-frame window to make sure the SE's pick it up.

For instance today is FEB 8th and I list some upcoming events on Feb 20th. Will the SE's pick this up in time and snapshot it for me?

Whats a good ballbark as a time frame as you never really know when the SE's will do updates. Would you say that 30 to 60 days is a fair window to be safe in? You mentioned in your post that a week in advance might do well in a blog for valentines day. Is that a safe window for just blogs or any type of site?


Usually I try to go about 60 days in advance if I was going to list say Things happening for halloween.


But your right - I've noticed that I do get a good amount of traffic on my time sensitve upcoming attractions. Google seems to crawl my site quite often and sometimes I can find results within a day or two.

[bragadocchio]

That's a good question.

I've noticed that if I make a blog post or two everyday on a regular basis, Googlebot will come back everyday and grab it. If I update less frequently, Google's indexing program doesn't come back as often.

One risk of changing content on a regular basis, is that if you remove something that is indexed, and people arrive looking for it, they may get frustrated and leave.

The nice thing about blog posts is that they stick around for a while, and people can see them when they come and visit, or visit the archives and look for them.

60 days might be a good time window for halloween on a site that isn't visited by Googlebot as frequently. It's possible that a shorter window of time may work for you, too. It's worth testing and investigating.

Any big annual events coming up that people might search for that you could write 400 or so words about for a new page? Might be worth putting a page, just for that event, a month or two ahead of time to see how it does, and linking to it from your main page.

[NorthStar]

The common feature I was noting with sandboxed websites is that while the amount of link-building varied from little to lots without changing much, those with more diversely worded links seemed the least affected. In fact, in some cases it almost seemed like artificial link-building was a contributing factor to a site becoming sandboxed. 

My initial hypothesis (which I never had opportunity to test far enough to become a proper theory) was that LSI was being applied to the anchor-text attribution algorithm(s) in Google. If a site had an 'unnaturally' heavy proportion of links that all used the exact same text then it stood a higher chance of being sandboxed.

I've checked this on one of the sites that was sandboxed and found that at least for this site the locations of the links were varied (eg. at the bottom of an article, on related site's links/helpful resource pages, directories, etc) and also that the inbound link text for the links was relatively diverse. So, in this case at least it would seem that it wasn't that these were "artificial links" or that the link text was identical or even relatively similar. In 25% of the cases the link was simply the URL.

It certainly seems like it was more a case of just not getting credit for good quality links for the better part of a year. However, now all my sandboxed sites are "out" (I've been avoiding working on any new sites for the past 6 months and stuck to big established sites so I don't have any sites currently experiencing the sandbox). What I'd like to know is this phenomena a thing of the past or will all new sites be subject to this "waiting" period for the forseeable future.

[Black_Knight]

Well, I can say for certain that not all new sites have been subject to the 'sandbox' since it first became an issue. However, there seems little hard-and-fast common ground to those that were sandboxed, or to those that were not - at least, not from what has been reported to me.

[NorthStar]

Black Knight

Care to share the info on the sites that weren't sandboxed? I didn't end up getting the work but I recently met with a large publishing company about working on their sites. They had 10 major sites for their biggest magazines (eg. 50,000 pages, 30,000 inbound links). They had just launched some for their smaller magazines. They were spending a lot of money in search ads. They seemed to be sandbox proof.


So, I agree, not all new sites were affected. But I certainly wouldn't feel confident talking to someone launching a smaller site about their chances of obtaining rankings quickly on Google.

[callback]

My site is not a new site at all, and it's been over a few years. Why did I get "sandboxed" and just got back to where it deserves? I had suffered once from Florida and then went back to normal high rankings at #1 or #2, and then last Spring it went down to #20+ or last month #91, but now got back to #5, which has made me more than happy.

[NorthStar]

callback

In my opinion, your site was affected by something other than the sandbox. Losing rankings or getting lower rankings like you describe has been something that sites have experienced since the Florida update and before. Sites that were in the sandbox wouldn't rank for their own name even if the name was unique. For any keywords they weren't in the top 1000 rankings while at the same time (at least my sites) ranked in Top 10 positions for their keywords on all the other search engines. The Sandbox was not a demotion of rankings, it was an abscence of ANY rankings and.....traffic.

I have other sites that have experienced what you are talking about. I don't think those were in the "sandbox". Something else is affecting them.

[NorthStar]

Looks like they are still dancing. All my sites have gone back into the sand...

[NorthStar]

back out with better rankings.