I'm now quite certain that the root of my problem with my smaller site is that it has been hijacked. I think it is the root for many of the sites that have been discussed here. It appears to me that in the case of hijacked site google has penalized the whole domain not just individual pages.
Saying that I need to point out that it is not just the fact of being hijacked. I find in that my larger site has not been hurt at all even though it was hijacked by the same domain as the smaller site was. I can see several possible factors why the hijack damaged one while leaving the other unscathed.
The larger site has a great many more good quality incoming links.
The larger site is, well, much larger.
The larger site has better internal linking and navigation.
The larger site uses CSS while the smaller one uses tables
The smaller site must have a greater percentage of incoming scraper links to it's quality related links. I've discovered it has at least a hundred scraper links from made for adsense sites. The big site probably has as many but it has so many more incoming links overall. These incoming links are completly out of our control but they are there and may be considered.
Any or all of these factors might have helped the larger site escape the duplicate penalty while the smaller one got it.
[edited by: annej at 7:00 pm (utc) on June 12, 2005]
|Billy, I sent you a PM earlier today on that tool you mentioned to check for Google penalties. (You may want to ask a mod if you can post that since it no doubt is a VERY VERY useful thing we can all use). |
Check your stickymail now, Clint. Sorry for the delay, but I was asleep. I'm in a way different timezone than the rest of you. :-)
Must've been the other Billy you were talking to... I'm Billy B. There is also a BillyS here.
Hijack - Canonical URL - same thing.
Hijack is just when Google thinks your Canonical url is the url from the hijacker.
Er - not actually sure thats true - might be though, not just a stab in the dark
"Steveb - got any thoughts on this - monitoring any sites which might be recovering from Canonical url issues?"
The one I watch the most has been slowly recovering, both before Bourbon and especially during Bourbon. It ranks #1 for its site name now, after having been in the hundreds (when hijacked) then moving to #9 or so, then moving to the top four, then now to #1. It has recovered somewhat for regular search terms, although it still ranks very poorly for its most competitive keyword (down 600 spots). It does rank #7 now though for its best keyword, where it previously ranked anywhere in the top ten from number one to number ten.
The_Contractor, non-www/www and absolute links is not the only canonical issues. Having duplicate articles all over the Internet might cause problems too, or linking to both "/" and "/index.php" can cause problems.
"I ran all the sites on the first few pages of G SERP's through the header check"
Clint you have mentioned that several times, and I'm afraid I have no idea what you think you are doing. More to the point, (and yes I say this all the time) don't get in the trap of thinking there is ONE BIG THING that is 100% consistent. Google deals sensibly with these canonical problems all the time, and they have for years. But SOMETIMES they don't. In the past year, every single update, a different batch of people end up with canonical problems. Don't take this literally, but lets just say 1% of websites had their canonical issues bungled by Google this time, while Google handled 99% sensibly. Google isn't perfect.
steveb: "don't get in the trap of thinking there is ONE BIG THING that is 100% consistent"
I agree, and whilst it is no particular comfort to people losing large amounts of traffic ( we are down 80% this week in revenue but last week we were 90% down ) I don't think there is one single element in this.
No-one I have talked to has any real clue - canonical issues MAY be part of it - as MAY in/out links/keyword density/crosslinking/new competition/old competition/file size/word count etc.
We are working on 3 "cunning plans" 1. do nothing as probably more changes on the way 2. Subtle page changes and monitor SERP changes 3. Forget the site and transfer to a spare URL
At this stage we are on "Cunning Plan" 1 and 2.
Jeeez, I hope it doesn't come down to cunning plan 3. What a drama.
>We are working on 3 "cunning plans" 1. do nothing as probably more changes on the way 2. Subtle page changes and monitor SERP changes 3. Forget the site and transfer to a spare URL<
Sounds very interesting. Would you be kind to elaborate more, if possible, about "cunning plan #2".
I hope not but I just spent 2 weeks on and off writing code very late at night for "Cunning Plan" 3 just in case. We have people to pay!
|Hijack is just when Google thinks your Canonical url is the url from the hijacker. |
That's little solace for webmasters with sites that are hardly seen by humans anymore.
There is a big difference between Google being a little confused in attributing PR to your www. vs. your non-www pages, and Google creating phantom webpages out of script links, and attributing your PR to its phantom page rather than your own legitimate one.
Reseller sure no problem on "cunning plan" number 2.
Assumptions: there may or may not be changes to come - I think there will be - GG estimate of a finish to changes about 21st June- I think he said this somewhere. GG will not give a "do this" answer.
So lets try to find out what the factors are. We made a huge spreadsheet of many of our page positions as at about May20/21.
We have many pages some will be ranked high some low - thats life.
Take some of the lower ranked pages and make some changes little risk - different changes to different pages. So far -
1. increase KY count density slightly in text
2. remove some internal links
3. remove some external links
4. Do some real high keyword density stuff
5. Increase file size
6. Add external adverts - we don't use adsense but have some doubleclick adverts
7. Add keywords at beginning of page and try some at bottom
8. set up the www/non-www in htaccess - thanks to all in forum and one guy especially by the way
9. Add keywords outside the nested tables design
10. Refine header tags description/title etc
11. Look at people above you in the SERPS
Sit back - Wait for the Googlebot
Preliminary Conclusions CURRENTLY as may change, Pr makes no difference, titles and description changes makes no difference, 75% of pages where we increase keyword density jumped up about 20 positions, but not all. We have a policy not to do massive keyword stuffing on pages - but trying it one minor page.
Pages we have in SERPS page one we didn't alter these have maintained positions +/- a few most days - including today.
Very heavily crawled last night so I will wait and see what happens in a few days.
Sorry so long! Don't all rush out and do the above as all industry sectors are different and what works for one may not work for another! Reader Beware!
Very generous reply. Much appreciated.
No problem reseller
We haven't really reached a real conclusion!
If Google think I am going to rewrite content just for them rather than the reader - well I'm not!
Interesting full page article in our Sunday Times Business section today - Google share price expected to rise BUT very vunerable if one of the other SE's get their act together.
I'm lining up stuff to put our version of Cunning Plan 3 into place... it's not something I really want to do, as we have a lot of clout with our current brand, but if things don't sort themselves out rebranding may be necessary.
What a drama.
I'm holding out for Cunning plan #1 then #2 as well....Also,
<<Interesting full page article in our Sunday Times Business section today - Google share price expected to rise BUT very vunerable if one of the other SE's get their act together. >>
Yah, I'd read something of the sort recently as well. That is, how Google's Stock is expected to soar yet how they should be a bit fearful of MSN in particular considering how much time Mr. Gates is investing in MSN.
Sorry all, I know this was OT.
In regards to cunning plan #2:
Recently, after seeing so many pages that merely linked to me, showing up in the serps, and my site having disappeared for the last few months (since the March update), i tried one simple thing:
I changed my title from "My Title Name" to "Welcome to My Title Name".
I know...sounds like a silly thing to even try. But I wanted Google to understand that this is the REAL site, which is why I am welcoming the users to it. This is not some OTHER site with simply a link to mine.
I certainly didn't think it would make a difference, but after a while, you go crazy and will try anything.
Three days later - I was back. Coincidence? Probably. But who knows? Maybe it unconfused Google.
So that was my cunning plan #2 that I'll say worked, just cuz I CAN say it. :)
Don't think it off topic given the title of the thread - perhaps a new one "Google dealing with the consequences of the Bourbon update"
I tried to do the very same thing... no effect.
But inconsistency is the catch-cry of this update - your plan may well work for hundreds of sites. Just not mine.
It's in human nature to try and fix things when they seem broken. I can't help but tinker with something that seemed wonderfully sucessful for five years. I hope the tinkering won't do more harm than good.
dazzlindonna: Interesting - perhaps I will wait a bit longer on the title changes. perhaps Google "lost" the connection between your site and the links in.
By coincidence "Welcome" is what I took OUT of our test pages text!
However I do think this a multi-factor change - we haven't been affected by any Google changes for years and years...I never really checked SERPS until May 20th.
It's also possible we have to wait until sites above us have been recached and refiltered.
"I changed my title from 'My Title Name' to 'Welcome to My Title Name'."
Thanks for the tip.
However, in my case, I refuse to make changes to my sites just to target a key money phrase that only represents 3% to 5% of my overall traffic.
I know my sites are built properly using Brett's recommendations. Those recommendations helped me reach pretty good positions in MSN.
If Google "thinks" my sites should be displayed in positions 100 plus for key money phrases, Google loses.
When the user goes to MSN and finds my sites because he-she didn't find valuable content in Google, Google loses.
Yes that is one of the things that worked in the past DD. You could also flip flop a few words back and fourth and that would start bringing the site back. I think they’re aiming now at a variation of what Yahoo is doing. If you can decisively beat their algo on a keyword you may tumble for that particular word.
If I was introducing an algo tweak I'd mask it by simultaneously implementing penalties I'd been holding off on.
The resulting confusion among Webmasters looking for a clear-cut "commonality" would be phenomenal....
I think it's a coincidence. I'm back on all but these 4 DCs:
I hope these are replaced soon though, it has gone on long enough.
|When the user goes to MSN and finds my sites because he-she didn't find valuable content in Google, Google loses. |
It's all about market share. I just spoke to someone outside of the web industry and he was aware of troubles with searcg results from Google.
|If I was introducing an algo tweak I'd mask it by simultaneously implementing penalties I'd been holding off on. |
The resulting confusion among Webmasters looking for a clear-cut "commonality" would be phenomenal....
That's a very "us vs. them" attitude. Google should be encouraging webmasters to make better websites, not to try and figure out ways to figure out, evade, and circumvent their rules of play.
I'm just putting myself in their shoes Billy, I'd have no qualms in trying to mask my latest algo tweak from the enemy, particularly if it only involved juggling a bit with the timing of penalties.
Some time ago, the attitude to SEOs from one of the founders was described as a mother Grizzly watching a hunter poke her cubs with a stick.
Sort of comment you don't forget... ;-)
[edited by: glengara at 10:29 pm (utc) on June 12, 2005]
"No-one I have talked to has any real clue..."
People can read the past year or more of posts on webmasterworld or not. These canonical problems have been happening for a long time, and to be hopelessly blunt about it, every time an update happens like this a whole pile of "new users" post but don't read the threads about the very same thing from February, December, September, etc.
Do what Google Guy suggested (plus hunt down and remove 302s that seem like they could be a problem). There is no excuse for not doing it, and it is just plain weird that people choose not to.
At the same time, every update also is just that, an update. Things go up and down. Some changes are attributable to factors only effecting one domain, maybe it was offline when the bot came by twice, maybe it lost a great link, maybe the great link site was offline when the bot came by.
There isn't some 100% answer that applies to everyone, and expecting that is silly. But if you don't learn from the many, many people who have been effected before you in a very similar way, then you are probably making a mistake.
Walkman, I really do think it was a coincidence. But saying it isn't makes me feel like i had some sort of control over my site. LOL.
Because of the consequences of the Bourbon Update on the sites I manage, I'm compelled to explain to owners of those sites why Google is ranking them in positions 100 plus.
It's difficult to deal with Google's myth that it does everythink OK. It's difficult to explain to technical and non-technical people Google sometimes messes up.
GoogleGuy wrote a few days ago his insight on Google's improvements on infrastructure and other stuff.
Did GoogleGuy meant to say "... summer is a good time to MESS UP THE work on ways to rank/score pages."
Hey GooglePlex, that's no good business.
[edited by: zafile at 10:45 pm (utc) on June 12, 2005]
steveb: "At the same time, every update also is just that, an update"
Sure - agree - pages come and pages go - everything moves on - but we address every issue seriously and always keen to learn, succh as canonical problems now dealt with as far as we can.
Yes I do read/learn as much as I can. But we have been up and running since the mid '90's and just concentrated on what we do - good clean content - add/update content answering the phone to enquiries - newsletters - competitions - good base number of users - repeat sales - gross sales in the 1-2 millions - great people working for me.
Then suddenly 90% drop of referals.
And with that - said Zebedee - time for bed! Tomorrow is the beginning of the rest of my life.
Steveb, ( a few of these points apply to your post)
I read them old posts... No solution there either so what's the point.
Track down and remove 302's - are you out of your mind?
There is nothing that any of us should do other than write decent white-hat sites for our visitors.
Move my site to another host who has a clue how to implement www to non-www consolidation? you must be kidding.
They only thing we, who are affected, can do is to publicize our plight and attempt to get this issue visible in the media and to the market analysts.
There's gotta be somebody here who knows someone or works for a company that knows someone! Isn't there? :(((