Forum Moderators: Robert Charlton & goodroi
In reply to a question from Brett Tabke, Matt said that there wasn't a sandbox, but the algorithm might affect some sites, under some circumstances, in a way that a webmaster would perceive as being sandboxed.
So, for some sites, in effect there IS a sandbox.
I hope you achieve a cultural shift in the way people refer to the "Sandbox effect". If this problem is broken down into smaller pieces, obviously, it will be much more manageable in terms of reporting symptoms and analysing results.
The "Sand Trap" affects new sites only. I like that too.
BTW, thanks for sharing the list of things that worked for you. I will combine it with other hard data on the thread to help with launch of two sites this week!
I like "sand trap". Established sites gone astray due to a mis-hit or faulty execution due to the webmaster...but can't quite figure out what's wrong with his/her swing.
Thanks for your approval. But lets limit the use of Sand Trap to only new sites that aren't getting into the SERPs for even specialized and uncompetitive searches. Maybe the solution for the Sand Trap will also help a poorly ranked established site, like aspirin can fix headaches AND backaches. But lets use separate names for both ailments.
Wouldn't the act of constantly creating new web sites possibly be a big signal that these web sites are spam sites?
Uhhhh......NO.
Besides being insulting to me and other site builders, that is probably the lamest idea in this entire thread. And I only threw "probably" in there to be polite.
How about this far-fetched possibility--people love your work and you have a growing business that allows you to take on new sites as you please.
As I work on only one site at my day job, I build 4 or 5 a year on the side as a way to pick up extra cash. I bet there are people on this board who build way more than that. A few years ago I was with an agency full time and I must have been building 10-12 a year.
The majority of my work is for commercial concerns, mostly brochure sites and a few catalog sites as well. They don't use AdSense, affiliate programs, or steal content. Even if a client asked me to build a spammy site, I doubt I would do it--unless they made me an offer I couldn't refuse. There are more than enough honest people out there with money who are willing to spend what it takes to do things the right way.
People who build numerous commercial sites consider the sandbox another cost of doing business, but it's the client who pays.
One has absolutely nothing to do with the other.
The sandbox is a site based penalty, not a page based penalty. So look for site based factors. Its a penalty, not a devaluation, so it is an indicator that G has detected an unnatural linking pattern. A site can go into the sandbox at any time that it trips the filters. Therefore, the filter is probably relatively simple to calculate.
Perhaps something like the ratio of natural links to unnatural links, where unnatural links may be:
What brings a site out of the sandbox? Perhaps site age is not a factor at all, except that the new sites happen to have few links and thus the linking patterns look more unnatural to google.
As I noted on page 6 of this thread, I've avoided the sandy thing twice. I've avoided it once in what I'd call a semi-competitive commercial area. Google showed most of our "money queries" had 3,000,000 to 5,000,000 organic results. From Google, we were getting 400 to 900 referrals per day, within a month. Traffic lasted like that for about six months and dropped to about half of that a couple of months ago when our SERPs dropped.
BTW... I don't know how I missed it and I certainly don't claim I could do it again.
Just between the two of us, it was travel/lodging.
But let's keep this kind of quiet.
At London Pubcon someone said - "Wow if you to THAT - it gets rid of all the SPAM." - Matt took notes and investigated (after he left London)- this was the start of the sandbox as we know it I think as the "THAT" was incorparted into the algo. I have tried to figure out what the THAT was - which was said. There is the door just need to find the key. Anyone at PUCON remeber this being said - what was the THAT?
If it has the same circumstance:
Site ranks well in Yahoo and MSN but poorly with Google.
And it's the same infraction:
"Unnatural" site or link growth.
And it has the same penalty:
In the index but way down in the SERPs.
And it has the same cure:
Sit and wait until your site matches the growth model.
Then it should have the same name:
Sandbox.
It will be more common to new sites, but can certainly happen to old sites.
Once again if the remedy is to do nothing--old or new--you're in the box.
And if you make very illogical assumptions, you come up with poor conclusions.
And if you make very illogical assumptions, you come up with poor conclusions.
Please expound upon the "very illogical" reference.
Unless you know something about the subject, you might be better off lurking.
We have (sadly yet again) reached the stage where supporters of one particular theory are telling people who disagree with them to stay out of their threads.
And Sparkys_Dad, since you asked, assertions of fact with less than rigorous underpinnings:
And it's the same infraction:
"Unnatural" site or link growth.
What makes it an infraction? What do you consider "unnatural"? What do you consider Google considers "unnatural"? And with what evidence?
And it has the same penalty:
In the index but way down in the SERPs.
Why do you argue that this is an example of a penalty? What is the evidence for this being a penalty?
And it has the same cure:
Sit and wait until your site matches the growth model.
So what about those who agree with you who still have sites awaiting "release" from their punishment for over nine months? What evidence do you have that they will eventually rank where they think they should do?
Two people sit in a leaky shack with rain pouring through the roof. "Ah", says one, "you angered the spirits by your evil actions. Now all we can do is wait for them to have pity on us. I have always found that once we have been punished enough, the sun will return and we will be dry again."
And you know what? From their point of view, he is right.
This is why words and the way that we use them are important ("penalty", "infraction", "sandbox") and why an impartial definition of a problem is necessary before attempting any analysis.
[edited by: nzmatt at 8:36 am (utc) on Nov. 21, 2005]
Presumably you mean 'if you add 300 pages overnight to a site which has already been indexed by G'?
So what about those who agree with you who still have sites awaiting "release" from their punishment for over nine months? What evidence do you have that they will eventually rank where they think they should do?
I think he has the same evidence of the rest of us who have been with this from the start. As far as I know all of the sites I have launched during the last 18 months (13) have been sandboxed. (Two or three of the smaller sites I did not bother tracking after launch because the clients did not ask for SEO.)
All of those that I tracked apart from one have been released. This may not be the hard evidence you are seeking but it is certainly evidence of a trend. The one that has not been released was launched exactly one year ago. It is currently number one in both MSN and Yahoo for its main search phrase.
This is the longest I have known any of my sites to be in the sandbox and I (obviously!) hope to be able to report its release in the next week or two.
At least you got one right.
Now work on your urge to presume wild assumptions are a useful starting point, and that everyone else should presume what you presume.
I think he has the same evidence of the rest of us who have been with this from the start.
The theory (apologies if my understanding is incomplete):
There is a something which is automatically applied to every new site which results in a delay in it attaining its "proper ranking" in Google. The something is time-based and will eventually expire with or without webmaster input.
The evidence:
A few webmasters/SEOs report some sites not being affected by this.
A number of webmasters/SEOs report that some sites appear to show these symptoms followed by eventual proper rankings.
A few webmasters/SEOs report that some sites have not appeared where they would expect them, after many months of waiting.
Would that be a fair summary?
From previous experience I would expect Google to Page Rank my site within 3-4 months and BEGIN to start ranking (albeit erratically) from then forth.
There is definitely a delay of some sort for new websites.
The theory (apologies if my understanding is incomplete):
There is a something which is automatically applied to every new site which results in a delay in it attaining its "proper ranking" in Google. The something is time-based and will eventually expire with or without webmaster input.
Agreed, but we must remember its "proper ranking" is not necessarily number one in the SERPs ;) I like to think that my own expectations are realistic and that I am qualified to judge when I should be some "somewhere" as opposed to "nowhere".
The evidence:
A few webmasters/SEOs report some sites not being affected by this.
Agreed. Some people seem to have managed to dodge the filter inadvertently. It may even be that there are one or two others who know why they have dodged it but are understandably keeping it to themselves.
A number of webmasters/SEOs report that some sites appear to show these symptoms followed by eventual proper rankings.
Agreed. This is my experience.
A few webmasters/SEOs report that some sites have not appeared where they would expect them, after many months of waiting.
Agreed. My record is 12 months and counting.
Would that be a fair summary?
Yes.
>>Would that be a fair summary?
Yes.
Good, in that case:
Random Thoughts on Reading Sandbox Threads
(in no particular order)
1) If there is a something which is being automatically applied, then why are sites not going in or not escaping? Is this due to inefficiencies in automation or to things that they do or never do possess?
2) If there are differences in length of effect, then does this show any correlation with topical areas? If so, what relevance does this have to any theory?
3) If there are differences in length of effect, then does this show any correlation with types of sites? If so, what relevance does this have to any theory?
4) Why necessarily posit a theory based on sites? Why would a search-related change not have the same effect as a "site-related" change?
5) What are search engines aiming to provide? What value would an automatic "sandbox" add to or detract from that goal?
6) Is there any published research which indicates something along the lines of the sandbox theory?
7) Is there any published research which indicates approaches which could have similar effects to those described by the sandbox theory?
8) Are there any previous instances of search engine actions which could lead to similar effects to those described by the sandbox theory?
9) If there is a temporal effect in any changes why should these be site-related rather than index-related or internal to Google?
10) Why (and here I join a number of other people here and elswhere) in discussions do we concentrate on those sites which have failed to perform satisfactorily, instead of looking at those which are performing satisfactorily?
For competitive key word ("widget information") on 5 month old site, I'm top 10 in MSN (fluctuates, usually number 1), usually top 10 in Yahoo, don't make top 1000 in Google. So, I believe in the "sandbox effect".
No owner bias involved, there is no doubt that for this competitive keyword I definitely have the best site on the net. It's widely acknowledged in the niche, I've spoken with several competitors and many complementary sites and everyone agrees it's the best by quite a margin. So, Google is doing its searchers a disservice.
For niche search terms ("joe's blue widget"), I rank fairly well in Google, and the site is 100% indexed and is crawled regularly, PR5.
In googles defense, I may have been a bit overzealous about link building in the early days, so maybe I got "sandboxed" because of that.
BeeBeeDubbleU .. I would suggest those who have dodged it were operating in moderately competitive areas at best, for those sites in the real big money areas I'm not sure if it's possible?
I would love to see some examples of sites that have avoided the sandbox and the terms they are ranking for ... nobody seems willing to divulge this info, which I can understand as they are potentially sat on a pot of gold, even if they haven't worked out just yet why the site dodged the sandbox yet ...
Having spent a while convinced there is no sandbox I am now beginning to think otherwise. My most recent site is currently ranking in MSN and Y for terms that I hoped it would. Google has now pushed the very same pages back to position 800 or so. The site has been live now for 6 months. I hadn't noticed anything as it didnt rank so well in MSN and Y over that period of time ao I assumed 'no sandbox' and that my SEO was probably a bit crap.
Anyway, following a few updates from Y and MSN and my site is ranking nicely. Google still provides me with the same amount of traffic as before and the pages that rank on the other 2 se's DO NOT rank in google.
Wonder how long this is going to last? Mind you, traffic from the other 2 has rocketed so I'm not too bothered with what google thinks.
Ska
What makes it an infraction?
What do you consider "unnatural"?
What do you consider Google considers "unnatural"?
And with what evidence?
Why do you argue that this is an example of a penalty? What is the evidence for this being a penalty?
So what about those who agree with you who still have sites awaiting "release" from their punishment for over nine months? What evidence do you have that they will eventually rank where they think they should do?
Poor question, poorly phrased. I cannot know where people think they should rank, and even if I did, not knowing the particular circumstances of their site, I could not know if their expectations are realistic.
I will rephrase your question for you:
What evidence do you have that sandboxed sites will achieve Google rankings reasonably consistent with their Yahoo, MSN and Google's allinanchor rankings?
Excluding sites that violate Google, Yahoo or MSN's Webmaster Guidelines, all sites eventually attain SERPs reasonably consistent across MSN, Yahoo, Google's allinanchor and Google. This is my experience. If you know of a site for which this does not apply, please feel free to run it by me.
I'll be waiting right here.
Excluding sites that violate Google, Yahoo or MSN's Webmaster Guidelines, all sites eventually attain SERPs reasonably consistent across MSN, Yahoo, Google's allinanchor and Google. This is my experience.
That's not my experience. There are many sites with top tens on one engine and nowhere on the others.
That's not my experience. There are many sites with top tens on one engine and nowhere on the others.
There can be a fairly wide variation in rankings but if you are ranking high on Y and MSN then you should be somewhere in Google. I think you find that "nowhere" with respect to normal ranking is a bit different from "nowhere" in the sandbox. Typically, sandboxed sites cannot be found in the top 1000 results.
A non-sandboxed site that has any sort of "legal" optimisation will be found somewhere. It may be in position 50, 100 or 200 but certainly in the top 1000. Similarly it will not be outranked by sites that have no relevance to the search.
