Forum Moderators: Robert Charlton & goodroi
1) I added a few hundred pages of content to try to monetize my site. Each page included an affiliate link with a rel='nofollow' tag. Perhaps google recognized the affiliate links and took away my trust rank? Perhaps I added too many pages? My site went from ~300 pages to ~900 pages.
2) All new pages were added in a sub-folder. Perhaps google thought the theme of my site had changed and penalized me? (although my index page stayed relatively the same)
3) Moved site to a dedicated server about 2 months previous.
Any thoughts?
(Well, If i had 300 pages with me at one shot, maybe I would have done the same thing and got into trouble! I didn't, and could get only 5-10 written every day).
I would wait a week or two in any case before doing anything - dont think the latest refresh is over yet from what they're saying in the other threads... and maybe you will be back with the next refresh.
I'm aware that Google reads scripted links. The script calls a PHP at my domain, processes the returned text, and sets a few innerHTMLs (the hidden submenus). I wouldn't think a call to the same domain would trip any spam sensors. All the links in the submenus are internal as well.
You're correct--by removing nav submenus containing about 100 links, I changed the PR distribution rather drastically. The 50,000 pages on my site now focus their links on only 5 pages plus adjacent pages instead of 100 pages plus adjacent. This was fine for about a month (my SERP actually improved), but on Sept 15 it suddenly went sour. I have the same number of pages, but the page content is reduced by perhaps 30-60% -- that is, the submenus were ~38K while other content of the pages is ~20-70K. You can see why I would want to 304 the submenus.
Might the reduction in content be as suspicious as monster88's change in page count?
Here's a more answerable question: What are the symptoms of being sandboxed? My PR remains at 5. I'm still crawled 2-3K pages/day. Are there telltales other than SERP change?
It's just clearer now, IMO anyway. Tripling the size of a site: I would never do that now...unless the site was very small.
Further, I'm not sure but it sound like you added not just new pages but a slew of potentially non-relevant pages. Affiliate pages. Yes, this plan was problematic I'm afraid.
IMO: Take those pages down. See what happens. All is not lost yet.
I'd respectfully disagree, a PR8 is no mean feat and could involve promotional methods that an informational site may get away with, but an affiliate site would not...
And either way I don't care. What I care about is what we learn and see for ourselves.
We've proved it to ourselves on ... ummm ... a number of different occasions. Let's say, more than five times in the last two years. Plus I've seen it twice with client sites I do SEO for.
It's a problem. The interesting thing is, the better the added pages, and/or the sooner the site gains additional high value inbound links, the sooner the site comes back. G knows what they're doing. One can take issue or disagree with why or how they do it... but they're doing it.
I added about 600 pages similar to this, ranging from just about anything that was offered by Commission Junction. shoes, clothes, electronics, laptops, bikes, sports equipment, flowers, books, etc. Most of the pages got lots of hits and most converted to sales.
You suddenly went from a site with a defined (well mapped theme) to a site with a highly diluted theme (especially with all those aff links pointing to CJ)..
You may need to take all that down and work to build out one or two profitable relationships with like themed sites...(for revenue)...
As far as when Google would relax the filter on your site...I don't think anyone can predict days, weeks or even months on this...but you will want to undo the damage (bad business decision to effectively SPAM Google this way)...
The reduction in content is actually a reduction in link text for the targets of those links.
I don't know if Google treats the link text as actual content as well.
What you are hinting at could be correct, but I'm disinclined to give it much weight.
You also completely changed the size of your template.
I have a strange feeling without seeing things that your site may just bounce back once all of the changes ripple through.
Yes I can understand why you wanted to change the size of your navagation structure. However that is a critical aspect of ranking and I'm certain your change wasn't gradual.
caveman,
How ya doing? Yes, I agree I think you have to watch out for exceeding your historical(or is that histerical) change rate. monster88's change rate went through the roof even though the absolute number of pages added wasn't many. It was also a major percentage size change as well.
glengara,
I would have thought by now folks would know that the amount of content to the number affiliate links must be several orders of magintude larger ;). How do we measure that? Is it usefulness of the content or is it one hundred words per length of the affiliate link in bytes?
monster88,
glengara has a point about affiliate links, the content on those pages is the key along with those pages being related to the rest of the site.
[edited by: theBear at 9:35 pm (utc) on Oct. 4, 2006]
Any one of these changes might have been sufficient to trigger a filter, but it's more likely the complete loss of trust was due to a combination of 2 or more of these changes:
1. % change in the number of pages
2. % change in the pace at which pages are being added.
3. Shift in number of relevant keywords from a small cluster of related keywords to a much larger number of keywords -- including unrelated ones.
4. Shift in site focus, from a narrowly focused, easily defined site to one that is confusingly or suspiciously diversified.
5. Addition of large number of affiliate links.
6. Unusually diverse mix of affiliate links.
I think it's the content, and the fact that he had an aff link had nothing to do with it. IMO, even if he linked to, say, netflix.com directly he would have had the same problem. On the other hand, if he had reviewed Netflix and in the middle of the 500 words he had an aff link, he'd be fine.
As I stated above, only google knows the truth, we're just guessing.
I don't think an affiliate link would hurt. But I wouldn't put an affiliate link to the same site on every page, especially if you have many affiliate links, all pointing to the same one merchant, or that all have the same affiliate tracking number. I don't think/know if Google can check this, but to me, it seems something that could easily get caught in the next filter tweak.
I add affiliate links, but I make sure they are from different merchants, have different affiliate tracking numbers, and I don't put too many on any single page, nor do I put them on all. But then again, I admit to be a tad paranoid.
After taking down the pages, I intend to email Google and ask for forgiveness. There's no guarantee that this would work, but it's worth a shot.
My site contains valuable information and is cited on many technical related sites and many .edu sites. This information is now lost because it doesn't show up in any searches. Although I still get traffic from MSN, it's still not worth the sacrifice of losing my Google traffic. The reason I put up the affiliate pages was to pay for the dedicated server and pay for the software development of the free software I offer on my site. I got a little carried away.
Does anyone know what is the best way to contact Google?
MC has the details to a re-inclusion request on his blog.
Does anyone know what is the best way to contact Google?
The best way is through WebMaster Central. If you start an account there and add your site, you don't have to add a sitemap (although I always do) and they'll give you various stats on how Google sees your site, and there are also links in there on how to request reinclusion. I don't remember from the beginning of this post if you are out of the index entirely or just dropped ranking. But if it's any consolation, I had a client who was out for over two years because of someone adding them to a link farm and a 'bad neighborhood' and once we cleaned all that out and asked for reinclusion, we got the whole site (200k plus pages) back in within a few weeeks. Didn't rank, but we were back in. Now, about nine months later, we're finally ranking on the first page for the phrases we want. So it *can* be fixed.
Have you made sure you have changed all "/index.htm" and "/default.htm" etc to "/"? This can potentially cause your site to trip on one of several duplicate content filters.
But there may be other filters [ and we haven't found them all yet either ]
I'm interested to see what patterns you went through and if we find anything that can help.
My bet is that you recently changed your pages from "/index.html" and "/default.htm" etc to "/" and Google hasn't had the time to re evaluate, index and restore your SE positioning.
If this is true, my concern is what happens next in the cycle [ we don't know as we are in the same position] and if the adding of new pages has compounded any problems.
Could you confirm your recent actions. This is critical to understanding how sites are restored after these changes are enacted.
[edited by: Whitey at 9:55 pm (utc) on Oct. 8, 2006]
It's very possible that this change may have caused my destruction in the SERPs. I still get a few referrals from Google now, but they are for very unique searches. Traffic from Google has dropped by over 99%.
What motivated you to invoke these changes? Were your pages not preforming or had you lost positioning? We need to know if you had a problem before or subsequent to your "fix"
Are the meta titles and meta descriptions unique?
I don't think the new pages you added are the problem, however they may have just added to the original problem by altering the severity of the number of pages considered duplicate.
Frankly, i don't know as you are only the 2nd person that i've seen who has reported this problem, but it would be good to get a better understanding of your predicament so that we can try to understand how the restoration process works [ or maybe doesn't! ].
I might start a new thread on the subject to assist.
[edited by: Whitey at 10:31 pm (utc) on Oct. 8, 2006]
All meta tags are unique.
After I made the change, my SERPS were still fine for about 2 months. Then every disappeared on the Oct 1 Google update.
Rankings lost after "/" fix [webmasterworld.com]
I'll cut and paste this to the new one.
What I'm seeing is this:
-Our big sites [ 80k pages ] were effected from at least BigDaddy [Feb06]
-Your small site [ 600 pages ] was "filtered" and/or deleted from the index when 30% extra pages were added [ i suspect some small sites escaped earlier "filters" ].
-You had a duplicate content problem anyway and then you fired strong links to add fuel to the fire, into each path to rank both sets of pages [ Ouch! - this is going to seriously effect you ]- Google's saying which path? - it's duplicate content.
- Google took 2 months to go back through your site and then decided to apply the filter [ with the extra 300 pages adding weight to it ]
That's how i see it.
You need to read this thread and look at g1smd's recommendations closely : [webmasterworld.com...]
Have you applied 301 redirects from the old sets of pages to the new ones?
Make sure any old links pointing to "bad" duplicate pages are redirected to the good "/" pages.
Now when you have fixed this part of it [ you may already have done this ] then you may find yourself at the same stage of restoration that we are in.
And this is where the other thread takes over, because i have no news of success stories once the work has been completed ie filters may still be applied - therefore i have no understanding of the restoration process and need help to understand what others are going through.
[edited by: Whitey at 12:39 am (utc) on Oct. 9, 2006]
Since we can't get into specifics it's really hard to compare the two situations and see what might be different, but I'll offer up this much:
- Every page was dynamic
- Same page name with varying parameters
- Same directory path
Maybe the fact that it was just a single page name made all the difference, hard to tell.
huh? you're not allowed to make money of your site now?
Spot on! GOOG must have their cut OR.
I had an entire site (6 years old site) disappear from the index two days after adding an affiliate link at the top of pages. "rel=nofolow" did not help much either.
The traffic Nazi gets very upset when it detects affiliate links.
The greed filter?
[edited by: Web_speed at 3:37 am (utc) on Oct. 9, 2006]
I had an entire site (6 years old site) disappear from the index two days after adding an affiliate link at the top of pages.
Nice conspiracy theory but I'm oozing in affiliate links and it's not an issue.
I would suspect something else is the problem.
My site seems to be viewed as an authority in our little niche and has constantly grown to about 450 pages during the past five years. Most of these 450 pages are entry pages for our shop system, ordered in a well chosen folder-structure, the names of the folders covering the semantical embracing terms of the product groups listed.
In early summer I received some very good data-material from one of our suppliers, which I inserted into our database. In one single day the number of pages doubled; this time I had chosen an alphabetical folder structure eg mydomain.com/orders-on-demand/w/widgets.html
A search for some zero-competitive search terms reveals, that googlebot has swallowed all these pages; however: my webmaster-central sitemaps page still lists this huge block as "page-rank not yet assigned" for more than three months now, which makes me a bit sceptical. There are no affiliate links on those pages. The content of those pages (mainly product-names) is of course closely related to our branch, although I doubt the semantic filters ever have a clue about this long tail.
I have a particular interest in this topic, because I am currently planning to add another 20.000 products (this time from affiliate partners) to my website, in order to offer my visitors an alternative in case my site-internal product-search-function shows no results. The topic of these products will be much less related semantically, I do not have a single affiliate link yet. With the other 450 pages yet waiting for pagerank: Is my site "under closer observation"? Should I better launch the new products slowly or better stop the whole project at all? What do you think?
I doubt it really makes sense for google to give too much weight to page-growth in times of databases and csv-files. That internet-shops grow rapidly from time to time, is only natural, and I believe google has well accounted for that.
Nice conspiracy theory but I'm oozing in affiliate links and it's not an issue.
javascript links by any chance?
In my case, ancore text link to a well known aff program (CB). The effect was almost immediate
[edited by: Web_speed at 9:29 am (utc) on Oct. 9, 2006]
