Sandboxed site study results

Preliminary

[Powdork]

I have been doing a study of sandboxed site statistics and here are some preliminary raw results. This is probably not as scientific as it could be but as the sample size grows it will become moreso. At this point i am not including data from one of the questions relating to directory listings (I used a radio button rather than a checkbox, so it could not be unchecked). The numbers represent the number of sites that fall within each parameter.

Month Domain Registered
* January 2004
* February
* March 2
* April 1
* May 2
* June
* July
* August 1
* September
* October
* November
* Prior to 2004 6

Month Uploaded
* January 2004 1
* February
* March 2
* April 1
* May 1
* June 3
* July 2
* August 1
* September
* October
* November
* Prior to 2004 1

Month Indexed
* January 2004 1
* February
* March 2
* April 1
* May 1
* June 3
* July 2
* August 1
* September
* October
* November
* Prior to 2004 1

Number of Pages
* 1-25 5
* 26-50 2
* 51-100 1
* 101-500 3
* 501-1,000 1
* 1,000-5,000 1
* >5,000 1

Number of Backlinks
* 1-25 2
* 26-50 2
* 51-100 1
* 101-500 4
* 501-1000 1
* >1000

Method of Backlinks
* None/Natural
* Passive 6
* Moderately Aggressive 5
* Very Aggressive 1

Anchor Text
* Natural 4
* Varied 4
* Tightly Focused 3

Adsense
* Yes 3
* No 9

Adwords
* Yes 3
* No 9

Content
* New 11
* Moved 1

Level of SEO
* Low 1
* Moderate 6
* High 4

Pages are
* Static 10
* Dynamic 2

There was also one site which was mature but added a large % of new pages.
Not all fields were required which is why there aren't always the same number for each field.
It is interesting that all the adsense users are also adwords advertisers.

[jcoronella]

Neat idea, but Just seeing the aggregate is confusing. Not being able to map a site specifically through each category yields no useful info.

My Site1:
Month Domain Registered
* Feb
Month Uploaded
* Feb
Month Indexed
* May
Number of Pages
* 101-500 3
Number of Backlinks
* >1000
Method of Backlinks
* Very Aggressive
Anchor Text
* Tightly Focused
Adsense
* No
Content
* New
Level Of SEO
* Moderate
Pages are
* Dynamic

[phantombookman]

I have a preliminary theory based on my own experience that should be confirmed one way or the other in the next week (already seen a big difference but just in case).

There is much talk of backlinks, new content etc but I believe these largely irrelevant (to being in the sandbox that is)

My tentative suggestion is don't think incoming links think outgoing!
i.e. don't have any

Regards
Rod

[Vec_One]

Thanks for doing this Powdork.

I'm surprised that more people haven't participated. Hey all you bums, hurry up. It just takes a minute. :)

[my3cents]

I only have one site:

Month Domain Registered
* August 2001

Month Uploaded
* August 2001

Month Indexed
* October 2001

Number of Pages
* 750+ (adding about 25-30 per month)

Number of Backlinks
* showing on g - 108
* showing everywhere else - 750+

Method of Backlinks
* Natural, Passive and Moderately Aggressive

Anchor Text
* Varied to home page and internal pages

Adsense
* No

Adwords
* Yes - only for a few secondary kw's

Content
* growing at about 25-30 pages per month

Level of SEO
* Moderate

Pages are
* Static

[airpal]

Great examples, but the most important piece of information is missing from all of them: Rank of site for its "money term" in Google, Yahoo, and MSN.

[jcoronella]

multiple money terms. Most top 10 (#1-#7)

What is Yahoo and MSN? ;)

[cbpayne]

my3cents, your site was up from 2001 -- its not sandboxed.

[glitterball]

You should add another paramater to your study:
DMOZ listing: Yes/No

[Powdork]

from the first post

At this point i am not including data from one of the questions relating to directory listings (I used a radio button rather than a checkbox, so it could not be unchecked).

So I don't know how the original stats for the directory listings may have been skewed. I will be updating the results tonight and the new results will include directory data from recent postigns and some older ones I have determined to be accurate. Suffice it to say that many had yahoo and dmoz listings some of which had migrated to the google directory. It didn't seem to come into play.

[Patient]

Hi Phantombookman

Are you saying that removing all outbound links is improving your rankings?

[phantombookman]

Hi Patient
no what I am saying is since the sandbox came into play all my new sites have been immediately dumped into it. My latest site has, thus far, not been sandboxed, the only difference is that this site has never had outgoing links at all.

It is too early to say but I wonder if this is lateral thinking by Google. The easiest and cheapest way to get links (or sell them) is to build a new site and plaster outgoing links all over it.

Could this be an extension of their policy of ..
you cannot be damaged by incoming links only by dubious outgoing links.

Still early days for me but watever the outcome this site is performing hugely different to my other recent efforts
Regards
Rod

[taxidriver]

Hi Guys
here is one for the fire.

I registered a domain on 20/07/04 (I think) and I added a holding page only.
I arranged a couple of my other (related) websites to link to it and a couple of other outside websites and guess what it has been in the first page of google out of nealy half a million for the search term <snip>. It was added into google within a couple of weeks with a pr4.
My question is that this site prob missed the sandbox but to confound it all I registered another site about the 11/08/04 and used the same algo and same link ideas and it has been given a pr4 but the pages dont come anywhere for any of the search terms that I use.

hmmmmmmmmm

Please tell me what u think of this then!

If you need any more info on the sites then gees a shout.

The taxidriver

[edited by: ciml at 12:20 pm (utc) on Dec. 3, 2004]
[edit reason] No specifics please. [/edit]

[hbirnbaum]

You're saying outbound links actually hurt your SERP? Site ranks better w/o outbounds? This is completely opposite of what some others said here recently... Anyone else find this to be the case?

[junai3]

I added a few more things, including allintext, allinanchor, allintitle and equivalent listings with Yahoo. Also my site comes up #1 for it's domain name (non competitive). Does this mean it's not sandboxed or is Google sandboxing sites for only certain terms?

My Site:
Month Domain Registered
* June
Month Uploaded
* June
Month Indexed
* July
Number of Pages
* 968 indexed by Google
* 13 indexed by Yahoo
Number of Backlinks
* 78 indexed by Google
* 185 indexed by Google
Method of Backlinks
* Moderate
Anchor Text
* Tightly Focused
Adsense
* No
Adwords
* Yes
Content
* New
Level Of SEO
* Moderate
Pages are
* Dynamic with static URLs
Allinanchor
* #9
Allintext
* #8
Allintitle
* #8
Rank
* 566 by Google
* 112 by Yahoo, new pages have not been indexed yet

[Patient]

Thanks phantombookman.

I have a site that launched beginning of June with no outbound links.

Pretty quickly had >20k pages indexed and rankings were OK (mainly page 2).

About 6 weeks ago I added 5 unrelated outbound links to all pages and since then my site has dissappeared from google.

At this point I believe these two events are unrelated but... who knows?

[taxidriver]

To Patient
I have still to check but I dont believe that once your site had missed the sandbox and had been given a good rank in google and could be found in a search, (Page 2 you say), that it would be added into the sandbox at a later date after having a good rank in the first instance.

I do think that, if not now but in the future, unrelated links to and from your site will have an affect on the rankins of sites.

Hope you all have a Merry Xmass and a Happy New Year.

Taxidriver

[phantombookman]

Hi Patient
did I understand your post correctly.
5 outgoing links added to every page?

You mention 20,000 pages which would be 100,000 links.
Once again it is only supposition but I could see that causing Google to wobble.

My apologies if I misinterpreted your post
Regards
Rod

[airpal]

TaxiDriver, your site was registered on:
12-Aug-2003. This gives it an enormous amount of time to avoid the sandbox.

How can I tell? Well your site is the only one optimized for the phrase you posted above.

[Powdork]

Thats funny airpal, i thought I was the only one foolish enough to go through the whois info from all the results on the first page for that search.

[airpal]

Powdork, I've done some research doing who-is lookups on the SERPS of a lot of fairly-competitive searches. And ofcourse, not a single time have I found a site registered or optimized after May 2004.

I get the feeling that the press hasn't reported this because it hasn't affected any of the big name SEOs who have big clients with old domain names/sites. I wonder how a young, eager reporter might feel about the opportunity to break a story on a magnitude of this scale. Maybe I should spend less time doing who-is lookups and more time looking up media members contact information...

[FourDegreez]

When people say "sandbox" do you mean that your site is not in Google at all or that it is in Google but ranked a hundred pages back? Because I keep hearing about this sandbox but I've yet to have a site fail to get into Google search results within 1 week of being up and linked to (from a single PR5/6 site), and I'm talking about the last few months. I'm not exactly ranking well, but I'm in.

[Powdork]

airpal you know the site studying the sandbox does have a small list of tech reporters that have reported on Google items in the past.

Four Degrrez,
the sandbox refers to ranking, not to being indexed.

[Strider]

here's my case Powdork:

domain name registered: june 2004
Month Uploaded: june 2004
Month Indexed: july 2004
Number of Pages: 1-25
number of backlinks: 26-50
Method of Backlinks: moderately aggressive
Anchor Text: Varied
adsense: no
adwords: no
content: new
level of SEO: high
pages: static
dmoz: no

allinanchor allintext allintitle - all #1
yahoo position #8 google position - #600 or smth :(

Hope it helps

[Hollywood]

Airpal and the rest?

=== Start quote ===

You said --> "I've done some research doing who-is lookups on the SERPS of a lot of fairly-competitive searches. And ofcourse, not a single time have I found a site registered or optimized after May 2004.
I get the feeling that the press hasn't reported this because it hasn't affected any of the big name SEOs who have big clients with old domain names/sites. I wonder how a young, eager reporter might feel about the opportunity to break a story on a magnitude of this scale. Maybe I should spend less time doing who-is lookups and more time looking up media members contact information..."

=== End quote ===

Can anyone prove this wrong above? I am very curious about this? (In short prove a site is coming up in SERPS on the first page for decent terms and has been registered after May 2004)

Hollywood

[junai3]

So is that the cutoff date, May 2004?

[airpal]

The reason I pick May 2004 is that I've seen sites registered in March 2004 that are ranking very high for strong keywords. However, I remember reading another thread where a lot of people said their new sites got reincluded into the main index after a May 2004 Update. After that, I have not seen anyone mention a site newer than 6 months ranking for a strong keyword. I actually hope I'm wrong on this one.

[sit2510]

I don't think there is anything to do with May 2004 about the cut-off date. Since the last monthly PR update was in June 2004, it is not surprising that new sites registered or started up in or after May 04 would have PR0 and have to undergo a long freezing period of 3 months until Oct 04 when PR updated again. During this time, the ability to get inbound links is very difficult because of low PR...so most of the new sites simply don't make it. Moreover, the delay of link leverage (sandbox) can take up to 90-180 days, so no doubt we can hardly see any new sites performing well for any competitive terms.

Now the strategy is to plan ahead for long...longer time. For example, start whatever you can do today in Dec 2004 and expect financial return to flow in between Aug-Oct 05, provided you are lucky enough.

In short, the formula for new decent sites under the patronage of good seo would be 3+3 or 3+6 months before you'll see good ranking.

[europeforvisitors]

During this time, the ability to get inbound links is very difficult because of low PR...so most of the new sites simply don't make it.

Natural linking will still occur (because natural links occur for reasons other than PR). If the result is to boost organic sites at the expense of SEOed sites, that's probably just fine with Google.

[Powdork]

Pages have had PR the entire time. It has not to do with PR or indexing. It has to do with ranking.