Those of you posting more regularly will already know our site was hit by some kind of Google penalty about 4 months ago. From one day to the next, Google traffic was more than halved. I think we have tried just about everything now, but there has been ZERO improvement. I'll list a few recent things we tried and I'd be keen to hear what else we can do.....
* redesigned the site and made all navigation text links to avoid a possible red flag with images as navigation. Cut down somewhat on load time due to this. * already had no followed a lot of external links, but have now encrypted them and stuck them behind a robots.txt file. * given google an incredibly strict robots.txt file to limit what it has to crawl. * made some redirects which were 302 (to our login page) into a 301...
Emails to both Matt Cutts and Adam Lasnik went unanswered (of course).
The site is large, has 20 000 back links being reported in webmaster tools, an active community of users and only unique content. It should be easy to crawl and is well structured from a semantics point of view. Everything is on topic and there are no bad neighboorhood links.
We expect to rank top ten for terms like 'travel blogs', 'travel forums', 'wiki travel guide', 'travel photography' and related terms, all without quotes. In Yahoo and MSN we do. Actually we ranked top ten for 'travel forums' until a few days ago but that seems to have dropped now too (I didn't think it could get any worse, but it obviously might?!).
The site is still being crawled and some pages still rank well, but terms like 'wiki travel guide', for which there are only 4 global sites that exist do not see us in the top 50 (I give up after that).
Aside from of course suing Google for dropping a site that complies with their terms (I would seriously consider this if I thought it would give me the answer as to what has caused this) what else can I do? I'm worried some hack/proxy issue/negative seo has caused this to happen but without Google sharing what they see as the problem, it seems impossible to fix since it can't seem to be found by anyone :(
I always thought Google wanted to list the best results per search term, but have seriously lost faith in their ability to do this based on our own experiences.
Have you allowed enough time for these "recent" things to take full effect?
> * made some redirects which were 302 (to our login page) ....
Can you be more specific about this? Do you mean that if a request for PageA is made by someone who has not logged in they are 301'd to the login page? If so, it would to follow that the page that you are 301 redirecting FROM would be dropped from the indexes. If you have other links to PageA, that would only add to the confusion! Hopefully this is NOT what you mean.
> The site is large, has 20 000 back links...
Yahoo reckons that less than a thousand of these are unique URLs.
> We expect to rank top ten for terms like 'travel blogs', 'travel forums', 'wiki travel guide', 'travel photography' and related terms, all without quotes.
On what basis do you "expect"; it is obviously a desire/ambition but is there objective evidence that other less-good sites are treated more favourably?
> The site is still being crawled and some pages still rank well ...
Yes, you and Google have done a good job with your sitemap.xml. I picked a page that your sitemap says is updated hourly and has a priority of 0.9, and found sure enough Google had indexed it only 2 hours ago. Pretty good eh? I'm not sure they will keep up that frequency though as I notice the most recent update appeared to be nearly a week old.
Finally, have you analysed what happens with your SERPs when British/ US spellings are used? If I google for [traveller blog] and for [traveler blog] (no quotes but notice the single/double L. Of the top ten SITES returned for those two keyphrases, only TWO sites appeared in both lists - both near the bottom of the first list and at the top of the second list. So I think it is clear that although [traveller] and [traveler] might for most practical purposes be seen by everyone as the same or equivalent, Google clearly indexes them very differently, with hardly any correlation in the little test I did above.
It might be worth analysing your site content with an eye on keywords that have different US/British English spellings - they certainly affect SERPs for many keywords but I don't how much that impacts your site.
> Have you allowed enough time for these "recent" things to take full > effect?
Thanks for the reply Robbo! I know it has most people stumped, so I'm keen to hear any ideas that are still left! That's a tricky question. We are assuming this is a penalty, for the simple reason that we used to rank top ten for these terms and were dropped to the last page with no warning. Overnight. So, considering the Google team is up to speed on the reinclusion requests, fixing whatever would have caused the penalty should also give the reverse effect. That's been pretty well documented. So in essence, yes, although I doubt any of these things would have caused the problem to start with. They are just attempts we're making at trying to unravel the mystery of what has transpired.
> > * made some redirects which were 302 (to our login page) ....
> Can you be more specific about this? Do you mean that if a request > for PageA is made by someone who has not logged in they are 301'd to > the login page? If so, it would to follow that the page that you are > 301 redirecting FROM would be dropped from the indexes. If you have > other links to PageA, that would only add to the confusion! > Hopefully this is NOT what you mean.
Some pages are only for logged in users. Naturally we don't expect these to rank at all. Things like a control panel or the blog/map management area. On some pages though it makes sense to link to those pages as a 'teaser', ie. click here to update your blog. If a member clicks there and they aren't logged in they get redirected to the login page, if they are logged in they are obviously taken straight to the correct page. We just had a standard coldfusion redirect in place and doing some tests I found that this gives a 302 header so I changed it to a 301. I'm not actually wanting any of those pages to rank, although I assume you didn't mean that the page that the link was on would be dropped for having a link to a page like this right?
> > The site is large, has 20 000 back links...
> Yahoo reckons that less than a thousand of these are unique URLs.
That's the number from webmaster tools. I'm not sure how you do the unique url search on yahoo, but I'm seeing Inlinks (59,567) for www.travellerspoint.com on yahoo for the 'except to this subdomain' and 'entire site' selected. Regardless of the number of back links, yahoo seems to be indexing and ranking the site basically as Google always has.
> > We expect to rank top ten for terms like 'travel blogs', 'travel forums', 'wiki travel guide', 'travel photography' and related terms, all without quotes.
> On what basis do you "expect"; it is obviously a desire/ambition but > is there objective evidence that other less-good sites are treated > more favourably?
I knew that question would come :) On the basis of a few things really.
a. that this is where we were always ranked by Google prior to May 1st for starters. I realize 'results from the past are no guarantee for the future' etc., but there is no logical reason I can think of for a site dropping off the face of search results for these terms when we certainly haven't been doing anything against their terms and conditions. The only logic countering this would be that they tweaked something indicating that the results of the past 5 years when it comes to our site must have been irrelevant. Our 110 000 members might beg differently, as would the tens of millions of visitors we have happily served in that period.
b. we still rank there on Yahoo and MSN (MSN even puts us at number 1 for travel guide which is total bs of course). Either both of these sites find that our site is relevant for these terms or they are both very very wrong. I'd actually be quite happy with anything at this stage actually, but when you put a result on the last page you are indicating that we are least relevant out of all the other pages you found for a certain search term.
and that brings me to c. which although mentioned last is certainly not intended as such. To point out the obvious, a search engine wants to rank the most relevant sites highest, right?! Well then I think based on our relevancy compared to our competitors we should rank at least together with them and with together I'm happy to mean below the ones that are our competitors, as long as it isn't 90 pages behind them and everything in between being comprised of blog posts that mention those words once or twice.
Wiki travel guide is probably the best example. There are 4 global wiki travel guides. We have one, there is world wikia and there is world66 and wikitravel.org. The best one is definitely wikitravel.org, second being either world66 or world wikia and ours would be 4th in terms of the amount of content (since it's only a couple of months old). Both world wikia and Travellerspoint can't be found on the first page for 'wiki travel guide' (no quotes), but in world wikia's site that's partly because they don't have their site well tuned to certain on page factors. If you search with quotes you get the really top notch results in the top ten, like this page: http://www.kichips.com/canada-travel/ No offense, but ??? Surely our site is a better reference than that with quotes? And surely our site has a few more backlinks and *just a little* more authority on the travel topic? I realize we don't run an MFA website, but does that mean we don't get to rank anymore?
> > The site is still being crawled and some pages still rank well ...
> Yes, you and Google have done a good job with your sitemap.xml. I > picked a page that your sitemap says is updated hourly and has a > priority of 0.9, and found sure enough Google had indexed it only 2 > hours ago. Pretty good eh? I'm not sure they will keep up that > frequency though as I notice the most recent update appeared to be > nearly a week old.
Hmm, I can't find that page; please would you be so kind as to let me know where you found it? My check revealed all pages with the highest priority are in fact the pages that should be highest priority (top level navigation). I tried to set the crawl rate to correspond with how often they are updated and also to coincide with caching on these pages.
Here are the only ones I'm finding: * http://www.travellerspoint.com/forum.cfm 0.9 daily (fresh posts every 5 minutes or so, so I don't think daily should be a problem?) * http://www.travellerspoint.com/blogs.cfm 0.9 hourly (several fresh blog entries every hour, is that a problem?) * http://www.travellerspoint.com/photography.cfm 0.9 hourly (about 500-600 uploaded per day, so that's at least a few per hour updating that page) * http://www.travellerspoint.com/guide/ 0.9 daily (relatively new area of the site so perhaps only 25 updates per day, although these do tend to be long, content rich updates. Again, I don't think I'm pushing it with daily right?) * http://www.travellerspoint.com/accommodation.cfm 0.9 monthly (reasonably static page so I put a month on it as we do have changes planned - and if I wanted to be really stupid I could just point to the random picture that appears which does mean the content changes every time cache clears, but we're not that kind of site).
I'm also not complaining about the crawl rate. When this whole thing started off and I logged into webmaster tools there was a friendly note saying Googlebot would like to be set to Faster, so I did so for it's pleasure. One of the 'fixes' we attempted was to add a sitemap, although quite frankly Google was having a fine time getting to every single page we needed it to get to before that.
> Finally, have you analysed what happens with your SERPs when British/ > US spellings are used? If I google for [traveller blog] and for > [traveler blog] (no quotes but notice the single/double L. Of the > top ten SITES returned for those two keyphrases, only TWO sites > appeared in both lists - both near the bottom of the first list and at > the top of the second list. So I think it is clear that although > [traveller] and [traveler] might for most practical purposes be seen > by everyone as the same or equivalent, Google clearly indexes them > very differently, with hardly any correlation in the little test I did > above.
> It might be worth analysing your site content with an eye on keywords > that have different US/British English spellings - they certainly > affect SERPs for many keywords but I don't how much that impacts your > site.
I wish it was something like this. But we've always focussed on the word travel, which is of course spelled the same in both 'languages' ;) The main reason for this is that we already knew this could be a problem and so purchased both domain names and redirected the one 'l'd version to the double ll version. Most sites link to the correct version, but the occasional site gets it wrong. We're not in contact with 50%+ of those that link to us, so there's no way I could correct that and it is relatively rare (plus the redirect is a 301 - although this was a 302 for a while because Godaddy's 301 redirect upon testing turns out to actually be a 302 redirect even though they claim the opposite?! argh). We do rank number one for 'travellers point' as well as 'travelers point' and of course the single word variety with both 1 and 2 l's.
It's really not an important point, but the page I was referring to is:
Travel Blogs - Travellerspoint Travel Blogs by Travellerspoint members. Start your own free, highly customizable travel blog or read up on what other Travellerspoint bloggers are writing ... www.travellerspoint.com/blogs.cfm - 31k - 1 hour ago - Cached - Similar pages - Note this
This latest contribution on that page is dated 22 August.
Another thought (read: "idle speculation" :-) ):
Re-reading your comments, many refer to the site as a whole - its "collective" value arising from various factors such as the number of pages, the range and depth of content, the freshness with new contributions all the time - all good points.
But I am not sure which specific PAGE should score extremely highly for any given search term.
Also I notice that if you google for site:travellerspoint.com the first item in the SERPs in your Home page BUT WITH https: prefix. If anyone clicks on that link, goes to the page and then clicks on other links they mostly (all that I checked) link to the other pages using https: prefix.
Do you know why your Homepage is indexed with the https: secure prefix? Have you noticed any other (non-secure) pages with the https prefix? Have you considered using 301 to force http: instead?
I had a quick look at several other sites that perform high when googling for [travel blogs] (travelblog.org, travelpod.com, and lonelyplanet.com). Am I right in thinking that your site makes extensive use of subdomains and the other sites do not? When I look down a list of blogs for Mexico (just an example) I see that most of the links take the visitor away from the current (www) subddomain and to various different subdomains whereas the other sites seem to be using structured hierarchy of folders. (That was my impression; I may be wrong/over-simplying/being irrelavant!!!)
It is also noticeable that the server response on your site is very slow compared with the others. THis may be due to a period particularly heavy demand for your pages, but it would be worth checking from different request points. (There is a site that has a free tool for checking server response time from several different timezones but I cannot remember its URL.)
> It's really not an important point, but the page I was referring to > is:
> Travel Blogs - Travellerspoint > Travel Blogs by Travellerspoint members. Start your own free, highly > customizable travel blog or read up on what other Travellerspoint > bloggers are writing ...www.travellerspoint.com/blogs.cfm- 31k - 1 hour ago - Cached - > Similar pages - Note this
> This latest contribution on that page is dated 22 August.
Although I agree it's not super important, I think it goes to credibility for why I've created the sitemap as it is that I point out that this is not the case. The center part of that page are manually reviewed blogs that have been hand picked because they have great content. These will get the top spot for a few days and then drop down the page as newer ones take over. The entire right hand side of the page is automatically updated (think it's cached for one hour) with new entries. The top one showing there now was posted about 20 minutes ago.
If the speed of posting on the blogs slowed down quite a lot I'd make the updating less often.
> Another thought (read: "idle speculation" :-) ):
> Re-reading your comments, many refer to the site as a whole - its > "collective" value arising from various factors such as the number of > pages, the range and depth of content, the freshness with new > contributions all the time - all good points.
> But I am not sure which specific PAGE should score extremely highly > for any given search term.
I'd expect something like this:
blogs page ranks for travel blogs forums page ranks for travel forums photography page ranks for travel photography page with blog entries from norway ranks for norway travel blog entries (or something similar).
etc.
Having done this for many years, we don't over optimize and stick with one or two phrases that we think are really relevant for a page. So our titles are really short and concise + very relevant. If you drill down really deeply you'd find variations of those terms on relevant pages, like the norway travel blogs one.
> Also I notice that if you google for site:travellerspoint.com the > first item in the SERPs in your Home page BUT WITH https: prefix. If > anyone clicks on that link, goes to the page and then clicks on other > links they mostly (all that I checked) link to the other pages using > https: prefix.
> Do you know why your Homepage is indexed with the https: secure > prefix? Have you noticed any other (non-secure) pages with the https > prefix? Have you considered using 301 to force http: instead?
That's a good point and one we've noticed. I have absolutely NO idea why they chose to index the https version. We have that for our own usage and don't link to it anywhere ourselves (hmmm, on double checking I see there is a link inside a pdf pointing at a secure admin area - I'll robots.txt the entire pdf folder out). Once you are on it, the relative url's all become https automatically and there's little to do about that. It actually seems to be the only page they've indexed with https aside from the above pdf. The cache on that is also from July 17th (seen by bigger file size and buggered up styles among others) whereas they have fresh cache from August 26 for the 'regular' home page. Interestingly enough, a lot of the results showing in site: search are actually the ones that we are no longer ranking for; could that be related?!
> I had a quick look at several other sites that perform high when > googling for [travel blogs] (travelblog.org, travelpod.com, and > lonelyplanet.com). Am I right in thinking that your site makes > extensive use of subdomains and the other sites do not? When I look > down a list of blogs for Mexico (just an example) I see that most of > the links take the visitor away from the current (www) subddomain and > to various different subdomains whereas the other sites seem to be > using structured hierarchy of folders. (That was my impression; I may > be wrong/over-simplying/being irrelavant!!!)
lonelyplanet only has one travel blog and is really irrelevant to the search, but that's an off topic rant on sites that are high on authority ranking for terms that aren't really related.
What you've noticed is basically my biggest concern though. I originally noted this when posting my first reinclusion request but of course I have no guarantee that this was looked at. Considering how desperate we are, it is not beyond us to change this, however we're talking around 6000 blogs and I don't particularly relish the thought of breaking this to our users if it's not the problem. That would be a very user unfriendly to do and that's generally not how we like to come across. Of course if this is the problem, we'd change the structure, but we used this structure since it's the default for blogging (a la blogspot, owned by google). Unlike the other travel blogging sites, we actually really tried to develop a 'blogging' system rather than a travel diary which is how I would personally class most of the other systems out there.
Also to clarify one thing. We did rel=nofollow all blog links on the blogs-per-country pages (like the Mexico one you mentioned) about 2 or 3 months ago in case this was the case. So we are not saying to Google that we trust all those blogs to be high quality. Although in our defense we do a good job at deleting advertising/spam ones because there are at least 3 editors subscribed to the blog entries feed + they are linked to from profiles and users that spam the blogs seem to be that daft that they also add a post in the forum, usually leading to a outright removal of their profile + blog. There's also a spam reporting tool developed but not live yet, in the unlikely event we miss an entry or two!
Should we noindex/nofollow all the blogs? Or should we try robots.txt'ing out the entire 6000 or so blogs (scary! Google, it would be nice if you could give just a teenie bit of guidance as to whether this might be the possible path to follow!!)?
> It is also noticeable that the server response on your site is very > slow compared with the others. THis may be due to a period > particularly heavy demand for your pages, but it would be worth > checking from different request points. (There is a site that has a > free tool for checking server response time from several different > timezones but I cannot remember its URL.)
Yes, this seems to be a bit of a problem the last few weeks. We're looking into this but so far no luck as to what exactly is causing this although it might have to do with our Facebook app. A server restart seems to fix it temporarily, but it soon slows again. I don't think this has much to do with the ranking issues though it is absolutely high priority! If you could remember that tool name I'd be interested in trying it out; sounds very useful!
> > It's really not an important point, but the page I was referring to > > is:
> > Travel Blogs - Travellerspoint > > Travel Blogs by Travellerspoint members. Start your own free, highly > > customizable travel blog or read up on what other Travellerspoint > > bloggers are writing ...www.travellerspoint.com/blogs.cfm-31k - 1 hour ago - Cached - > > Similar pages - Note this
> > This latest contribution on that page is dated 22 August.
> Although I agree it's not super important, I think it goes to > credibility for why I've created the sitemap as it is that I point out > that this is not the case. The center part of that page are manually > reviewed blogs that have been hand picked because they have great > content. These will get the top spot for a few days and then drop down > the page as newer ones take over. The entire right hand side of the > page is automatically updated (think it's cached for one hour) with > new entries. The top one showing there now was posted about 20 minutes > ago.
> If the speed of posting on the blogs slowed down quite a lot I'd make > the updating less often.
> > Another thought (read: "idle speculation" :-) ):
> > Re-reading your comments, many refer to the site as a whole - its > > "collective" value arising from various factors such as the number of > > pages, the range and depth of content, the freshness with new > > contributions all the time - all good points.
> > But I am not sure which specific PAGE should score extremely highly > > for any given search term.
> I'd expect something like this:
> blogs page ranks for travel blogs > forums page ranks for travel forums > photography page ranks for travel photography > page with blog entries from norway ranks for norway travel blog > entries (or something similar).
> etc.
> Having done this for many years, we don't over optimize and stick with > one or two phrases that we think are really relevant for a page. So > our titles are really short and concise + very relevant. If you drill > down really deeply you'd find variations of those terms on relevant > pages, like the norway travel blogs one.
> > Also I notice that if you google for site:travellerspoint.com the > > first item in the SERPs in your Home page BUT WITH https: prefix. If > > anyone clicks on that link, goes to the page and then clicks on other > > links they mostly (all that I checked) link to the other pages using > > https: prefix.
> > Do you know why your Homepage is indexed with the https: secure > > prefix? Have you noticed any other (non-secure) pages with the https > > prefix? Have you considered using 301 to force http: instead?
> That's a good point and one we've noticed. I have absolutely NO idea > why they chose to index the https version. We have that for our own > usage and don't link to it anywhere ourselves (hmmm, on double > checking I see there is a link inside a pdf pointing at a secure admin > area - I'll robots.txt the entire pdf folder out). Once you are on it, > the relative url's all become https automatically and there's little > to do about that. It actually seems to be the only page they've > indexed with https aside from the above pdf. The cache on that is also > from July 17th (seen by bigger file size and buggered up styles among > others) whereas they have fresh cache from August 26 for the 'regular' > home page. Interestingly enough, a lot of the results showing in site: > search are actually the ones that we are no longer ranking for; could > that be related?!
> > I had a quick look at several other sites that perform high when > > googling for [travel blogs] (travelblog.org, travelpod.com, and > > lonelyplanet.com). Am I right in thinking that your site makes > > extensive use of subdomains and the other sites do not? When I look > > down a list of blogs for Mexico (just an example) I see that most of > > the links take the visitor away from the current (www) subddomain and > > to various different subdomains whereas the other sites seem to be > > using structured hierarchy of folders. (That was my impression; I may > > be wrong/over-simplying/being irrelavant!!!)
> lonelyplanet only has one travel blog and is really irrelevant to the > search, but that's an off topic rant on sites that are high on > authority ranking for terms that aren't really related.
> What you've noticed is basically my biggest concern though. I > originally noted this when posting my first reinclusion request but of > course I have no guarantee that this was looked at. Considering how > desperate we are, it is not beyond us to change this, however we're > talking around 6000 blogs and I don't particularly relish the thought > of breaking this to our users if it's not the problem. That would be a > very user unfriendly to do and that's generally not how we like to > come across. Of course if this is the problem, we'd change the > structure, but we used this structure since it's the default for > blogging (a la blogspot, owned by google). Unlike the other travel > blogging sites, we actually really tried to develop a 'blogging' > system rather than a travel diary which is how I would personally > class most of the other systems out there.
> Also to clarify one thing. We did rel=nofollow all blog links on the > blogs-per-country pages (like the Mexico one you mentioned) about 2 or > 3 months ago in case this was the case. So we are not saying to Google > that we trust all those blogs to be high quality. Although in our > defense we do a good job at deleting advertising/spam ones because > there are at least 3 editors subscribed to the blog entries feed + > they are linked to from profiles and users that spam the blogs seem to > be that daft that they also add a post in the forum, usually leading > to a outright removal of their profile + blog. There's also a spam > reporting tool developed but not live yet, in the unlikely event we > miss an entry or two!
> Should we noindex/nofollow all the blogs? Or should we try > robots.txt'ing out the entire 6000 or so blogs (scary! Google, it > would be nice if you could give just a teenie bit of guidance as to > whether this might be the possible path to follow!!)?
> > It is also noticeable that the server response on your site is very > > slow compared with the others. THis may be due to a period > > particularly heavy demand for your pages, but it would be worth > > checking from different request points. (There is a site that has a > > free tool for checking server response time from several different > > timezones but I cannot remember its URL.)
> Yes, this seems to be a bit of a problem the last few weeks. We're > looking into this but so far no luck as to what exactly is causing > this although it might have to do with our Facebook app. A server > restart seems to fix it temporarily, but it soon slows again. I don't > think this has much to do with the ranking issues though it is > absolutely high priority! If you could remember that tool name I'd be > interested in trying it out; sounds very useful!
Apologies for the multiple posts in a row. While adding the above directory to robots.txt, I ended up on the back links for the travel blogs page. Interestingly enough it was noting this url: http://太化.com/url.php?q=cambogia-riel&u=nwzoravfllfrsptcdonezithnzzgufsomtu sfs-co-it-35nezmohl
I'd tried clicking it before, but it wouldn't work, so now i copied and pasted it into safari. Sure enough, it goes straight to page of ours (not even the travel blogs page...) and isn't a backlink at all. Strip away all the extra and just enter the domain name and you'll see a clear spam site. Could this be negative SEO at work with a 302 redirect?
> As an aside, the /pdf/ folder is now blocked through robots.txt and > I'm requesting removal of the folder through webmaster tools.
> On Aug 28, 12:17 am, Sam I Am wrote:
> > > It's really not an important point, but the page I was referring to > > > is:
> > > Travel Blogs - Travellerspoint > > > Travel Blogs by Travellerspoint members. Start your own free, highly > > > customizable travel blog or read up on what other Travellerspoint > > > bloggers are writing ...www.travellerspoint.com/blogs.cfm-31k- 1 hour ago - Cached - > > > Similar pages - Note this
> > > This latest contribution on that page is dated 22 August.
> > Although I agree it's not super important, I think it goes to > > credibility for why I've created the sitemap as it is that I point out > > that this is not the case. The center part of that page are manually > > reviewed blogs that have been hand picked because they have great > > content. These will get the top spot for a few days and then drop down > > the page as newer ones take over. The entire right hand side of the > > page is automatically updated (think it's cached for one hour) with > > new entries. The top one showing there now was posted about 20 minutes > > ago.
> > If the speed of posting on the blogs slowed down quite a lot I'd make > > the updating less often.
> > > Another thought (read: "idle speculation" :-) ):
> > > Re-reading your comments, many refer to the site as a whole - its > > > "collective" value arising from various factors such as the number of > > > pages, the range and depth of content, the freshness with new > > > contributions all the time - all good points.
> > > But I am not sure which specific PAGE should score extremely highly > > > for any given search term.
> > I'd expect something like this:
> > blogs page ranks for travel blogs > > forums page ranks for travel forums > > photography page ranks for travel photography > > page with blog entries from norway ranks for norway travel blog > > entries (or something similar).
> > etc.
> > Having done this for many years, we don't over optimize and stick with > > one or two phrases that we think are really relevant for a page. So > > our titles are really short and concise + very relevant. If you drill > > down really deeply you'd find variations of those terms on relevant > > pages, like the norway travel blogs one.
> > > Also I notice that if you google for site:travellerspoint.com the > > > first item in the SERPs in your Home page BUT WITH https: prefix. If > > > anyone clicks on that link, goes to the page and then clicks on other > > > links they mostly (all that I checked) link to the other pages using > > > https: prefix.
> > > Do you know why your Homepage is indexed with the https: secure > > > prefix? Have you noticed any other (non-secure) pages with the https > > > prefix? Have you considered using 301 to force http: instead?
> > That's a good point and one we've noticed. I have absolutely NO idea > > why they chose to index the https version. We have that for our own > > usage and don't link to it anywhere ourselves (hmmm, on double > > checking I see there is a link inside a pdf pointing at a secure admin > > area - I'll robots.txt the entire pdf folder out). Once you are on it, > > the relative url's all become https automatically and there's little > > to do about that. It actually seems to be the only page they've > > indexed with https aside from the above pdf. The cache on that is also > > from July 17th (seen by bigger file size and buggered up styles among > > others) whereas they have fresh cache from August 26 for the 'regular' > > home page. Interestingly enough, a lot of the results showing in site: > > search are actually the ones that we are no longer ranking for; could > > that be related?!
> > > I had a quick look at several other sites that perform high when > > > googling for [travel blogs] (travelblog.org, travelpod.com, and > > > lonelyplanet.com). Am I right in thinking that your site makes > > > extensive use of subdomains and the other sites do not? When I look > > > down a list of blogs for Mexico (just an example) I see that most of > > > the links take the visitor away from the current (www) subddomain and > > > to various different subdomains whereas the other sites seem to be > > > using structured hierarchy of folders. (That was my impression; I may > > > be wrong/over-simplying/being irrelavant!!!)
> > lonelyplanet only has one travel blog and is really irrelevant to the > > search, but that's an off topic rant on sites that are high on > > authority ranking for terms that aren't really related.
> > What you've noticed is basically my biggest concern though. I > > originally noted this when posting my first reinclusion request but of > > course I have no guarantee that this was looked at. Considering how > > desperate we are, it is not beyond us to change this, however we're > > talking around 6000 blogs and I don't particularly relish the thought > > of breaking this to our users if it's not the problem. That would be a > > very user unfriendly to do and that's generally not how we like to > > come across. Of course if this is the problem, we'd change the > > structure, but we used this structure since it's the default for > > blogging (a la blogspot, owned by google). Unlike the other travel > > blogging sites, we actually really tried to develop a 'blogging' > > system rather than a travel diary which is how I would personally > > class most of the other systems out there.
> > Also to clarify one thing. We did rel=nofollow all blog links on the > > blogs-per-country pages (like the Mexico one you mentioned) about 2 or > > 3 months ago in case this was the case. So we are not saying to Google > > that we trust all those blogs to be high quality. Although in our > > defense we do a good job at deleting advertising/spam ones because > > there are at least 3 editors subscribed to the blog entries feed + > > they are linked to from profiles and users that spam the blogs seem to > > be that daft that they also add a post in the forum, usually leading > > to a outright removal of their profile + blog. There's also a spam > > reporting tool developed but not live yet, in the unlikely event we > > miss an entry or two!
> > Should we noindex/nofollow all the blogs? Or should we try > > robots.txt'ing out the entire 6000 or so blogs (scary! Google, it > > would be nice if you could give just a teenie bit of guidance as to > > whether this might be the possible path to follow!!)?
> > > It is also noticeable that the server response on your site is very > > > slow compared with the others. THis may be due to a period > > > particularly heavy demand for your pages, but it would be worth > > > checking from different request points. (There is a site that has a > > > free tool for checking server response time from several different > > > timezones but I cannot remember its URL.)
> > Yes, this seems to be a bit of a problem the last few weeks. We're > > looking into this but so far no luck as to what exactly is causing > > this although it might have to do with our Facebook app. A server > > restart seems to fix it temporarily, but it soon slows again. I don't > > think this has much to do with the ranking issues though it is > > absolutely high priority! If you could remember that tool name I'd be > > interested in trying it out; sounds very useful!
It looks like you are hyperlinking an image (ideally it should be text) and then I think you have a error in your css that will make it appear that you are hiding text. I'm so sleepy right now I can't think. Maybe someone else will check this out further...it's bedtime for me. Kerry
> Those of you posting more regularly will already know our site was hit > by some kind of Google penalty about 4 months ago. From one day to the > next, Google traffic was more than halved. I think we have tried just > about everything now, but there has been ZERO improvement. I'll list a > few recent things we tried and I'd be keen to hear what else we can > do.....
> * redesigned the site and made all navigation text links to avoid a > possible red flag with images as navigation. Cut down somewhat on load > time due to this. > * already had no followed a lot of external links, but have now > encrypted them and stuck them behind a robots.txt file. > * given google an incredibly strict robots.txt file to limit what it > has to crawl. > * made some redirects which were 302 (to our login page) into a > 301...
> Emails to both Matt Cutts and Adam Lasnik went unanswered (of > course).
> The site is large, has 20 000 back links being reported in webmaster > tools, an active community of users and only unique content. It should > be easy to crawl and is well structured from a semantics point of > view. Everything is on topic and there are no bad neighboorhood > links.
> We expect to rank top ten for terms like 'travel blogs', 'travel > forums', 'wiki travel guide', 'travel photography' and related terms, > all without quotes. In Yahoo and MSN we do. Actually we ranked top ten > for 'travel forums' until a few days ago but that seems to have > dropped now too (I didn't think it could get any worse, but it > obviously might?!).
> The site is still being crawled and some pages still rank well, but > terms like 'wiki travel guide', for which there are only 4 global > sites that exist do not see us in the top 50 (I give up after that).
> Aside from of course suing Google for dropping a site that complies > with their terms (I would seriously consider this if I thought it > would give me the answer as to what has caused this) what else can I > do? I'm worried some hack/proxy issue/negative seo has caused this to > happen but without Google sharing what they see as the problem, it > seems impossible to fix since it can't seem to be found by anyone :(
> I always thought Google wanted to list the best results per search > term, but have seriously lost faith in their ability to do this based > on our own experiences.
Yes, that css is indeed hiding the word Travellerspoint when the page is viewed in screen mode. If you are on a mobile browser the logo is hidden and only the text shows. This is done for maximum usability benefits and Google confirmed on an earlier posting I made asking about this that this kind of usage was okay. Someone from Google actually looked into it and confirmed it.
Note that the text says EXACTLY what the text on the logo also says so the same info is being served to both the users and the search engines. We don't benefit in any way, except that the text looks better :)
Any other theories? I'm really keen to tap the wisdom of all the really smart people here as if no one here can figure it out, it must really be a G bug....
> It looks like you are hyperlinking an image (ideally it should be > text) and then I think you have a error in your css that will make it > appear that you are hiding text. I'm so sleepy right now I can't > think. Maybe someone else will check this out further...it's bedtime > for me. Kerry
> On Aug 27, 8:04 am, Sam I Am wrote:
> > Hi everyone,
> > Those of you posting more regularly will already know our site was hit > > by some kind of Google penalty about 4 months ago. From one day to the > > next, Google traffic was more than halved. I think we have tried just > > about everything now, but there has been ZERO improvement. I'll list a > > few recent things we tried and I'd be keen to hear what else we can > > do.....
> > * redesigned the site and made all navigation text links to avoid a > > possible red flag with images as navigation. Cut down somewhat on load > > time due to this. > > * already had no followed a lot of external links, but have now > > encrypted them and stuck them behind a robots.txt file. > > * given google an incredibly strict robots.txt file to limit what it > > has to crawl. > > * made some redirects which were 302 (to our login page) into a > > 301...
> > Emails to both Matt Cutts and Adam Lasnik went unanswered (of > > course).
> > The site is large, has 20 000 back links being reported in webmaster > > tools, an active community of users and only unique content. It should > > be easy to crawl and is well structured from a semantics point of > > view. Everything is on topic and there are no bad neighboorhood > > links.
> > We expect to rank top ten for terms like 'travel blogs', 'travel > > forums', 'wiki travel guide', 'travel photography' and related terms, > > all without quotes. In Yahoo and MSN we do. Actually we ranked top ten > > for 'travel forums' until a few days ago but that seems to have > > dropped now too (I didn't think it could get any worse, but it > > obviously might?!).
> > The site is still being crawled and some pages still rank well, but > > terms like 'wiki travel guide', for which there are only 4 global > > sites that exist do not see us in the top 50 (I give up after that).
> > Aside from of course suing Google for dropping a site that complies > > with their terms (I would seriously consider this if I thought it > > would give me the answer as to what has caused this) what else can I > > do? I'm worried some hack/proxy issue/negative seo has caused this to > > happen but without Google sharing what they see as the problem, it > > seems impossible to fix since it can't seem to be found by anyone :(
> > I always thought Google wanted to list the best results per search > > term, but have seriously lost faith in their ability to do this based > > on our own experiences.
I don't know about noindex'ing and nofollow'ing all the blogs but maybe just nofollow'ing at least as an experiment.
The reason I suggest it is that I look at other blog systems and it seems that none of them link to any of their internal member blogs.
It would seem a shame to have to do that but it might be worth a try temporarily.
Doing that though, you'd want to make sure there was some way that each blog operator had some way, if they don't already, of setting up their own analytics, webmaster tools etc.
As for the strange backlink, I wouldn't think it could hurt you but I'm not that experienced with the 302 hijacking that use to be prevalent in the past.
A totally off the wall idea, there has been a lot of talk recently about proxies causing problems, have you been following any of those discussions and if so, do you think there might be something to look into there?
Craig
p.s. How about this idea, anyone whose problems can't be identified here in Google's Webmaster Tools Help forum becomes eligible for a "Get out of the shit-house free" card from Google?
> Yes, that css is indeed hiding the word Travellerspoint when the page > is viewed in screen mode. If you are on a mobile browser the logo is > hidden and only the text shows. This is done for maximum usability > benefits and Google confirmed on an earlier posting I made asking > about this that this kind of usage was okay. Someone from Google > actually looked into it and confirmed it.
> Note that the text says EXACTLY what the text on the logo also says so > the same info is being served to both the users and the search > engines. We don't benefit in any way, except that the text looks > better :)
> Any other theories? I'm really keen to tap the wisdom of all the > really smart people here as if no one here can figure it out, it must > really be a G bug....
> Sam
> On Aug 28, 10:03 am, kklynnt wrote:
> > Sam, I'm going to post something on here for others to check out....I > > just found it and maybe someone can take a quick peek.
> > It looks like you are hyperlinking an image (ideally it should be > > text) and then I think you have a error in your css that will make it > > appear that you are hiding text. I'm so sleepy right now I can't > > think. Maybe someone else will check this out further...it's bedtime > > for me. Kerry
> > On Aug 27, 8:04 am, Sam I Am wrote:
> > > Hi everyone,
> > > Those of you posting more regularly will already know our site was hit > > > by some kind of Google penalty about 4 months ago. From one day to the > > > next, Google traffic was more than halved. I think we have tried just > > > about everything now, but there has been ZERO improvement. I'll list a > > > few recent things we tried and I'd be keen to hear what else we can > > > do.....
> > > * redesigned the site and made all navigation text links to avoid a > > > possible red flag with images as navigation. Cut down somewhat on load > > > time due to this. > > > * already had no followed a lot of external links, but have now > > > encrypted them and stuck them behind a robots.txt file. > > > * given google an incredibly strict robots.txt file to limit what it > > > has to crawl. > > > * made some redirects which were 302 (to our login page) into a > > > 301...
> > > Emails to both Matt Cutts and Adam Lasnik went unanswered (of > > > course).
> > > The site is large, has 20 000 back links being reported in webmaster > > > tools, an active community of users and only unique content. It should > > > be easy to crawl and is well structured from a semantics point of > > > view. Everything is on topic and there are no bad neighboorhood > > > links.
> > > We expect to rank top ten for terms like 'travel blogs', 'travel > > > forums', 'wiki travel guide', 'travel photography' and related terms, > > > all without quotes. In Yahoo and MSN we do. Actually we ranked top ten > > > for 'travel forums' until a few days ago but that seems to have > > > dropped now too (I didn't think it could get any worse, but it > > > obviously might?!).
> > > The site is still being crawled and some pages still rank well, but > > > terms like 'wiki travel guide', for which there are only 4 global > > > sites that exist do not see us in the top 50 (I give up after that).
> > > Aside from of course suing Google for dropping a site that complies > > > with their terms (I would seriously consider this if I thought it > > > would give me the answer as to what has caused this) what else can I > > > do? I'm worried some hack/proxy issue/negative seo has caused this to > > > happen but without Google sharing what they see as the problem, it > > > seems impossible to fix since it can't seem to be found by anyone :(
> > > I always thought Google wanted to list the best results per search > > > term, but have seriously lost faith in their ability to do this based > > > on our own experiences.
The links within the blog to external sites are no followed (it's UGC and originally what the tag was meant for after all), there's just a few links to Travellerspoint stuff but perhaps no following these or just removing them all together is indeed the way to go. My annoyance with this is that it also blocks the good bots like Yahoo and MSN... but will definitely consider this!
> The reason I suggest it is that I look at other blog systems and it > seems that none of them link to any of their internal member blogs.
I guess you mean regular blogs, not travel blogs? Most 'travel blogs' sites link heavily to their users blogs and then re-distribute that content and duplicate it across other sites like Sitestep who pay them for the content to help increase their search engine rankings. It seems to work too, just not for us who aren't doing this :(
Also just to clarify; we do no follow all links to the blogs right now, at least from the directory layout, but right now not from the main blogs page which will list the last 50 or so entries roughly. Did you mean no follow all links from our site to the blogs or from the blogs back to our site? Might look into the first actually to see if there's anywhere else we might have missed. It stands to reason that any blog still getting a ranking after that must be being linked to from elsewhere so perhaps that would help.
> Doing that though, you'd want to make sure there was some way that > each blog operator had some way, if they don't already, of setting up > their own analytics, webmaster tools etc.
Analytics has been done but is not pushed live yet (took some time to figure out how to run two analytics code sets on the page!) although all users have access to some basic stats. 99.99% of our users wouldn't care about any of the webmaster tools, so we could live without those. Side note; do you know of any way we can add all blogs automatically to our webmaster tools area and pull it in under the existing travellerspoint account there?
> As for the strange backlink, I wouldn't think it could hurt you but > I'm not that experienced with the 302 hijacking that use to be > prevalent in the past.
I too thought this wasn't possible anymore, but it sure looks like an attempt at the exploit.
> A totally off the wall idea, there has been a lot of talk recently > about proxies causing problems, have you been following any of those > discussions and if so, do you think there might be something to look > into there?
I've looked into the proxies, but can't find any trace of any proxies that might have our site covered like this. I've tried a few unique searches that I can think of but no luck there so far. That isn't to say it's not the case, just impossible so far for me to find. I'm thinking of the reverse cloaking, but it just seems so utterly wrong to get involved in cloaking, even though it is to fix a google bug.
> p.s. How about this idea, anyone whose problems can't be identified > here in Google's Webmaster Tools Help forum becomes eligible for a > "Get out of the shit-house free" card from Google?
> > Yes, that css is indeed hiding the word Travellerspoint when the page > > is viewed in screen mode. If you are on a mobile browser the logo is > > hidden and only the text shows. This is done for maximum usability > > benefits and Google confirmed on an earlier posting I made asking > > about this that this kind of usage was okay. Someone from Google > > actually looked into it and confirmed it.
> > Note that the text says EXACTLY what the text on the logo also says so > > the same info is being served to both the users and the search > > engines. We don't benefit in any way, except that the text looks > > better :)
> > Any other theories? I'm really keen to tap the wisdom of all the > > really smart people here as if no one here can figure it out, it must > > really be a G bug....
> > Sam
> > On Aug 28, 10:03 am, kklynnt wrote:
> > > Sam, I'm going to post something on here for others to check out....I > > > just found it and maybe someone can take a quick peek.
> > > It looks like you are hyperlinking an image (ideally it should be > > > text) and then I think you have a error in your css that will make it > > > appear that you are hiding text. I'm so sleepy right now I can't > > > think. Maybe someone else will check this out further...it's bedtime > > > for me. Kerry
> > > On Aug 27, 8:04 am, Sam I Am wrote:
> > > > Hi everyone,
> > > > Those of you posting more regularly will already know our site was hit > > > > by some kind of Google penalty about 4 months ago. From one day to the > > > > next, Google traffic was more than halved. I think we have tried just > > > > about everything now, but there has been ZERO improvement. I'll list a > > > > few recent things we tried and I'd be keen to hear what else we can > > > > do.....
> > > > * redesigned the site and made all navigation text links to avoid a > > > > possible red flag with images as navigation. Cut down somewhat on load > > > > time due to this. > > > > * already had no followed a lot of external links, but have now > > > > encrypted them and stuck them behind a robots.txt file. > > > > * given google an incredibly strict robots.txt file to limit what it > > > > has to crawl. > > > > * made some redirects which were 302 (to our login page) into a > > > > 301...
> > > > Emails to both Matt Cutts and Adam Lasnik went unanswered (of > > > > course).
> > > > The site is large, has 20 000 back links being reported in webmaster > > > > tools, an active community of users and only unique content. It should > > > > be easy to crawl and is well structured from a semantics point of > > > > view. Everything is on topic and there are no bad neighboorhood > > > > links.
> > > > We expect to rank top ten for terms like 'travel blogs', 'travel > > > > forums', 'wiki travel guide', 'travel photography' and related terms, > > > > all without quotes. In Yahoo and MSN we do. Actually we ranked top ten > > > > for 'travel forums' until a few days ago but that seems to have > > > > dropped now too (I didn't think it could get any worse, but it > > > > obviously might?!).
> > > > The site is still being crawled and some pages still rank well, but > > > > terms like 'wiki travel guide', for which there are only 4 global > > > > sites that exist do not see us in the top 50 (I give up after that).
> > > > Aside from of course suing Google for dropping a site that complies > > > > with their terms (I would seriously consider this if I thought it > > > > would give me the answer as to what has caused this) what else can I > > > > do? I'm worried some hack/proxy issue/negative seo has caused this to > > > > happen but without Google sharing what they see as the problem, it > > > > seems impossible to fix since it can't seem to be found by anyone :(
> > > > I always thought Google wanted to list the best results per search > > > > term, but have seriously lost faith in their ability to do this based > > > > on our own experiences.
Sam, this is really the only thing I can see. Can you provide the earlier post. I'd like to see what they said and in what context. Having your site's data on a mobile is an awesome capability and of course I see exactly what you mean in terms of styling to suite the mobile platform's functionality.
I think there may be a conflict here with how this is being implemented. Hiding the image in the mobile style sheet and trying to add text to that area in the browser's stylesheet might cause problems for the Googlebot.
Have you considering using the mobile CSS to apply text in place of the image? -- basically i believe you would want to control everything related to the desired mobile styling with the mobile CSS exclusively. I think you could safely hide the image here and apply text in it's place within this style sheet with no problems.
As it stands now it looks like the browsers are relying on the order of the cascade to render your desired effects. I'm not exactly sure how Google is going to treat each individual style sheet when it is used in this capacity.
I'd just like to double check here so we can get you back up and running. :)
> Yes, that css is indeed hiding the word Travellerspoint when the page > is viewed in screen mode. If you are on a mobile browser the logo is > hidden and only the text shows. This is done for maximum usability > benefits and Google confirmed on an earlier posting I made asking > about this that this kind of usage was okay. Someone from Google > actually looked into it and confirmed it.
> Note that the text says EXACTLY what the text on the logo also says so > the same info is being served to both the users and the search > engines. We don't benefit in any way, except that the text looks > better :)
> Any other theories? I'm really keen to tap the wisdom of all the > really smart people here as if no one here can figure it out, it must > really be a G bug....
> Sam
> On Aug 28, 10:03 am, kklynnt wrote:
> > Sam, I'm going to post something on here for others to check out....I > > just found it and maybe someone can take a quick peek.
> > It looks like you are hyperlinking an image (ideally it should be > > text) and then I think you have a error in your css that will make it > > appear that you are hiding text. I'm so sleepy right now I can't > > think. Maybe someone else will check this out further...it's bedtime > > for me. Kerry
> > On Aug 27, 8:04 am, Sam I Am wrote:
> > > Hi everyone,
> > > Those of you posting more regularly will already know our site was hit > > > by some kind of Google penalty about 4 months ago. From one day to the > > > next, Google traffic was more than halved. I think we have tried just > > > about everything now, but there has been ZERO improvement. I'll list a > > > few recent things we tried and I'd be keen to hear what else we can > > > do.....
> > > * redesigned the site and made all navigation text links to avoid a > > > possible red flag with images as navigation. Cut down somewhat on load > > > time due to this. > > > * already had no followed a lot of external links, but have now > > > encrypted them and stuck them behind a robots.txt file. > > > * given google an incredibly strict robots.txt file to limit what it > > > has to crawl. > > > * made some redirects which were 302 (to our login page) into a > > > 301...
> > > Emails to both Matt Cutts and Adam Lasnik went unanswered (of > > > course).
> > > The site is large, has 20 000 back links being reported in webmaster > > > tools, an active community of users and only unique content. It should > > > be easy to crawl and is well structured from a semantics point of > > > view. Everything is on topic and there are no bad neighboorhood > > > links.
> > > We expect to rank top ten for terms like 'travel blogs', 'travel > > > forums', 'wiki travel guide', 'travel photography' and related terms, > > > all without quotes. In Yahoo and MSN we do. Actually we ranked top ten > > > for 'travel forums' until a few days ago but that seems to have > > > dropped now too (I didn't think it could get any worse, but it > > > obviously might?!).
> > > The site is still being crawled and some pages still rank well, but > > > terms like 'wiki travel guide', for which there are only 4 global > > > sites that exist do not see us in the top 50 (I give up after that).
> > > Aside from of course suing Google for dropping a site that complies > > > with their terms (I would seriously consider this if I thought it > > > would give me the answer as to what has caused this) what else can I > > > do? I'm worried some hack/proxy issue/negative seo has caused this to > > > happen but without Google sharing what they see as the problem, it > > > seems impossible to fix since it can't seem to be found by anyone :(
> > > I always thought Google wanted to list the best results per search > > > term, but have seriously lost faith in their ability to do this based > > > on our own experiences.
Don't worry Kerry, you won't be the first or last not to find anything else 'wrong' :) I think that's why some of the experts aren't chiming in, although I wish more people would chime in and say they can not find anything wrong as it might send a signal to Google that they really need to look at some of those cases.
Sorry, but you'll have to dig through some nasty comments back and forth by a few to the second page where Susan comes in and answers based on discussing it with Matt Cutts. My original question was based on our entire navigation being like this, something we have since changed.
I should also point out that this technique stretches far and wide on the internet among web designers and developers alike as it is the only way right now to get good looking text/graphics but still have a fall back option. So from a usability point of view, some things like this are absolutely necessary (for those users browsing on browsers that don't show images, you'd still like them to see what the image that is there says). My point being that it is not only for mobile users where the handheld stylesheet kicks in, but is indeed largely for usability, something that I also point out in that other thread is HIGHLY recommended by Google. And as you can see in this case, the text on the logo actually matches the text in the span to the letter, so there is no benefit one way or the other for us (plus of course we do rank for Travellerspoint already). I know there was a case where a known seo'er was so called 'caught' for using a similar technique, even though his case was perfectly excusable and blown way out of proportion by some fanatics, the thing that made it stand out for those that felt really strongly about it was that the text did not mirror the image.
> Have you considering using the mobile CSS to apply text in place of > the image? -- basically i believe you would want to control everything > related to the desired mobile styling with the mobile CSS exclusively. > I think you could safely hide the image here and apply text in it's > place within this style sheet with no problems.
Do you mean use a display:none on the mobile style and then add a content:before/after? That second command is not accepted by all browsers and would probably cause some problems. In essence this method is actually worse than what we are doing as well because you are physically adding content to the page that doesn't exist there to start with. If that isn't dodgier than clearly showing the word(s) in your source, then that would be beyond absurd.
On a side note; from what I understand Googlebot doesn't crawl the stylesheets at all just yet, so there should be no way Googlebot could even get this wrong. But I might have missed the announcement to that effect!
> Sam, this is really the only thing I can see. Can you provide the > earlier post. I'd like to see what they said and in what context. > Having your site's data on a mobile is an awesome capability and of > course I see exactly what you mean in terms of styling to suite the > mobile platform's functionality.
> I think there may be a conflict here with how this is being > implemented. Hiding the image in the mobile style sheet and trying to > add text to that area in the browser's stylesheet might cause problems > for the Googlebot.
> Have you considering using the mobile CSS to apply text in place of > the image? -- basically i believe you would want to control everything > related to the desired mobile styling with the mobile CSS exclusively. > I think you could safely hide the image here and apply text in it's > place within this style sheet with no problems.
> As it stands now it looks like the browsers are relying on the order > of the cascade to render your desired effects. I'm not exactly sure > how Google is going to treat each individual style sheet when it is > used in this capacity.
> I'd just like to double check here so we can get you back up and > running. :)
> Kerry
> On Aug 28, 3:24 am, Sam I Am wrote:
> > Hi kklynnt,
> > Yes, that css is indeed hiding the word Travellerspoint when the page > > is viewed in screen mode. If you are on a mobile browser the logo is > > hidden and only the text shows. This is done for maximum usability > > benefits and Google confirmed on an earlier posting I made asking > > about this that this kind of usage was okay. Someone from Google > > actually looked into it and confirmed it.
> > Note that the text says EXACTLY what the text on the logo also says so > > the same info is being served to both the users and the search > > engines. We don't benefit in any way, except that the text looks > > better :)
> > Any other theories? I'm really keen to tap the wisdom of all the > > really smart people here as if no one here can figure it out, it must > > really be a G bug....
> > Sam
> > On Aug 28, 10:03 am, kklynnt wrote:
> > > Sam, I'm going to post something on here for others to check out....I > > > just found it and maybe someone can take a quick peek.
> > > It looks like you are hyperlinking an image (ideally it should be > > > text) and then I think you have a error in your css that will make it > > > appear that you are hiding text. I'm so sleepy right now I can't > > > think. Maybe someone else will check this out further...it's bedtime > > > for me. Kerry
> > > On Aug 27, 8:04 am, Sam I Am wrote:
> > > > Hi everyone,
> > > > Those of you posting more regularly will already know our site was hit > > > > by some kind of Google penalty about 4 months ago. From one day to the > > > > next, Google traffic was more than halved. I think we have tried just > > > > about everything now, but there has been ZERO improvement. I'll list a > > > > few recent things we tried and I'd be keen to hear what else we can > > > > do.....
> > > > * redesigned the site and made all navigation text links to avoid a > > > > possible red flag with images as navigation. Cut down somewhat on load > > > > time due to this. > > > > * already had no followed a lot of external links, but have now > > > > encrypted them and stuck them behind a robots.txt file. > > > > * given google an incredibly strict robots.txt file to limit what it > > > > has to crawl. > > > > * made some redirects which were 302 (to our login page) into a > > > > 301...
> > > > Emails to both Matt Cutts and Adam Lasnik went unanswered (of > > > > course).
> > > > The site is large, has 20 000 back links being reported in webmaster > > > > tools, an active community of users and only unique content. It should > > > > be easy to crawl and is well structured from a semantics point of > > > > view. Everything is on topic and there are no bad neighboorhood > > > > links.
> > > > We expect to rank top ten for terms like 'travel blogs', 'travel > > > > forums', 'wiki travel guide', 'travel photography' and related terms, > > > > all without quotes. In Yahoo and MSN we do. Actually we ranked top ten > > > > for 'travel forums' until a few days ago but that seems to have > > > > dropped now too (I didn't think it could get any worse, but it > > > > obviously might?!).
> > > > The site is still being crawled and some pages still rank well, but > > > > terms like 'wiki travel guide', for which there are only 4 global > > > > sites that exist do not see us in the top 50 (I give up after that).
> > > > Aside from of course suing Google for dropping a site that complies > > > > with their terms (I would seriously consider this if I thought it > > > > would give me the answer as to what has caused this) what else can I > > > > do? I'm worried some hack/proxy issue/negative seo has caused this to > > > > happen but without Google sharing what they see as the problem, it > > > > seems impossible to fix since it can't seem to be found by anyone :(
> > > > I always thought Google wanted to list the best results per search > > > > term, but have seriously lost faith in their ability to do this based > > > > on our own experiences.
Also: http://host-tracker.com/ will give you the individual and average response times of your site from 30 centers across the world. It gave me averages of 9 seconds for your site and 0.6 sec for two of my sites that I just used for comparison (not like with like as mine are MUCH smaller than yours).
"Buying links in order to improve a site's ranking is in violation of Google's webmaster guidelines and can negatively impact a site's ranking in search results."
> Also:http://host-tracker.com/ > will give you the individual and average response times of your site > from 30 centers across the world. It gave me averages of 9 seconds > for your site and 0.6 sec for two of my sites that I just used for > comparison (not like with like as mine are MUCH smaller than yours).
Robbo, that's always been like that. We've always been an authority site, which is why we have the nice extended listing. But the search for the home page on the word 'travellerspoint' has never been the problem; it's all the other top level pages. Not that ranking for Travellerspoint is bad as it still gives us a few thousand uniques a week, but those people are looking for us because of our reputation and already know about us. It'd be nice if we got some traffic on our product pages too, like the blogs/photography/guide etc.
Yeah, the response time blows at the moment. It must be a cfm thing, but it's tough to figure out what is causing it exactly.
> Also:http://host-tracker.com/ > will give you the individual and average response times of your site > from 30 centers across the world. It gave me averages of 9 seconds > for your site and 0.6 sec for two of my sites that I just used for > comparison (not like with like as mine are MUCH smaller than yours).
Poor Donna, she doesn't deserve to be copied that much! She's one of those great webmasters that puts tons of time into her site (currently coded in straight html for example) and has been running it for over a decade but I don't think follows much of what goes on in the discussions over paid links, like probably 99% of the other webmasters out there. The link is clearly marked 'SPONSORED' and that's what google wants (it's the only form both humans and robots can see). In any case, that ad is being pulled (thought it actually had already been pulled) based on totally unrelated reasons. There is of course no way that google knows if I paid for that link or perhaps got it in return for a favour done (interestingly enough they've said paid reviews in blog posts are not considered paid for example?!). I'm personally not of the impression that sites buying links are being penalized simply because it means we could all go out and get our competitors banned but it's food for thought I guess. Google is yet to mention one case of this happening and the best Matt Cutts could do last week at SES was a case involving cloaking so that doesn't bode well...
I see Donna has a few serious proxy issues though so perhaps she's been penalized due to that Google proxy bug and in return all sites linked to from her site are also getting hit. Hmmm, interesting.
> Perhaps you ran into the paid links buzz saw. Sure they say they'll > penalize the sites that sell links, but the webmaster guidelines say > the opposite.
> "Buying links in order to improve a site's ranking is in violation of > Google's webmaster guidelines and can negatively impact a site's > ranking in search results."
> On Aug 28, 4:03 pm, Robbo wrote:
> > Sam
> > I see your homepage is back in the indexes and with a very nice entry > > too!
> > Also:http://host-tracker.com/ > > will give you the individual and average response times of your site > > from 30 centers across the world. It gave me averages of 9 seconds > > for your site and 0.6 sec for two of my sites that I just used for > > comparison (not like with like as mine are MUCH smaller than yours).
This kind of stuff annoys me beyond belief, but I'm still right in thinking this kind of stealing/harvesting of our content on sites paid for by adsense can't harm us right?
> Poor Donna, she doesn't deserve to be copied that much! She's one of > those great webmasters that puts tons of time into her site (currently > coded in straight html for example) and has been running it for over a > decade but I don't think follows much of what goes on in the > discussions over paid links, like probably 99% of the other webmasters > out there. The link is clearly marked 'SPONSORED' and that's what > google wants (it's the only form both humans and robots can see). In > any case, that ad is being pulled (thought it actually had already > been pulled) based on totally unrelated reasons. There is of course no > way that google knows if I paid for that link or perhaps got it in > return for a favour done (interestingly enough they've said paid > reviews in blog posts are not considered paid for example?!). I'm > personally not of the impression that sites buying links are being > penalized simply because it means we could all go out and get our > competitors banned but it's food for thought I guess. Google is yet to > mention one case of this happening and the best Matt Cutts could do > last week at SES was a case involving cloaking so that doesn't bode > well...
> I see Donna has a few serious proxy issues though so perhaps she's > been penalized due to that Google proxy bug and in return all sites > linked to from her site are also getting hit. Hmmm, interesting.
> On Aug 28, 11:31 pm, JLH wrote:
> > Not sure if this is an issue but a site that's been copied all over > > the web has a listing for you as a sponsored link:
> > Perhaps you ran into the paid links buzz saw. Sure they say they'll > > penalize the sites that sell links, but the webmaster guidelines say > > the opposite.
> > "Buying links in order to improve a site's ranking is in violation of > > Google's webmaster guidelines and can negatively impact a site's > > ranking in search results."
> > On Aug 28, 4:03 pm, Robbo wrote:
> > > Sam
> > > I see your homepage is back in the indexes and with a very nice entry > > > too!
> > > Also:http://host-tracker.com/ > > > will give you the individual and average response times of your site > > > from 30 centers across the world. It gave me averages of 9 seconds > > > for your site and 0.6 sec for two of my sites that I just used for > > > comparison (not like with like as mine are MUCH smaller than yours).
I'd like some time to read through and study this. This has caught my eye because as people do try to build functional sites for mobile platforms,etc. then CSS and how Google will treat all this will come into play.
With your particular site, I can see that accessing it via a mobile device would be a huge perk for the user. It would be a real shame if Google's system dinged someone for trying to build something that gives them the needed effects. (Not sure that is what is happening)
Thanks for the link...I definitely want to study this concept some more. Kerry
> Don't worry Kerry, you won't be the first or last not to find anything > else 'wrong' :) I think that's why some of the experts aren't chiming > in, although I wish more people would chime in and say they can not > find anything wrong as it might send a signal to Google that they > really need to look at some of those cases.
> Sorry, but you'll have to dig through some nasty comments back and > forth by a few to the second page where Susan comes in and answers > based on discussing it with Matt Cutts. My original question was based > on our entire navigation being like this, something we have since > changed.
> I should also point out that this technique stretches far and wide on > the internet among web designers and developers alike as it is the > only way right now to get good looking text/graphics but still have a > fall back option. So from a usability point of view, some things like > this are absolutely necessary (for those users browsing on browsers > that don't show images, you'd still like them to see what the image > that is there says). My point being that it is not only for mobile > users where the handheld stylesheet kicks in, but is indeed largely > for usability, something that I also point out in that other thread is > HIGHLY recommended by Google. And as you can see in this case, the > text on the logo actually matches the text in the span to the letter, > so there is no benefit one way or the other for us (plus of course we > do rank for Travellerspoint already). I know there was a case where a > known seo'er was so called 'caught' for using a similar technique, > even though his case was perfectly excusable and blown way out of > proportion by some fanatics, the thing that made it stand out for > those that felt really strongly about it was that the text did not > mirror the image.
> > Have you considering using the mobile CSS to apply text in place of > > the image? -- basically i believe you would want to control everything > > related to the desired mobile styling with the mobile CSS exclusively. > > I think you could safely hide the image here and apply text in it's > > place within this style sheet with no problems.
> Do you mean use a display:none on the mobile style and then add a > content:before/after? That second command is not accepted by all > browsers and would probably cause some problems. In essence this > method is actually worse than what we are doing as well because you > are physically adding content to the page that doesn't exist there to > start with. If that isn't dodgier than clearly showing the word(s) in > your source, then that would be beyond absurd.
> On a side note; from what I understand Googlebot doesn't crawl the > stylesheets at all just yet, so there should be no way Googlebot could > even get this wrong. But I might have missed the announcement to that > effect!
> On Aug 28, 10:25 pm, kklynnt wrote:
> > Sam, this is really the only thing I can see. Can you provide the > > earlier post. I'd like to see what they said and in what context. > > Having your site's data on a mobile is an awesome capability and of > > course I see exactly what you mean in terms of styling to suite the > > mobile platform's functionality.
> > I think there may be a conflict here with how this is being > > implemented. Hiding the image in the mobile style sheet and trying to > > add text to that area in the browser's stylesheet might cause problems > > for the Googlebot.
> > Have you considering using the mobile CSS to apply text in place of > > the image? -- basically i believe you would want to control everything > > related to the desired mobile styling with the mobile CSS exclusively. > > I think you could safely hide the image here and apply text in it's > > place within this style sheet with no problems.
> > As it stands now it looks like the browsers are relying on the order > > of the cascade to render your desired effects. I'm not exactly sure > > how Google is going to treat each individual style sheet when it is > > used in this capacity.
> > I'd just like to double check here so we can get you back up and > > running. :)
> > Kerry
> > On Aug 28, 3:24 am, Sam I Am wrote:
> > > Hi kklynnt,
> > > Yes, that css is indeed hiding the word Travellerspoint when the page > > > is viewed in screen mode. If you are on a mobile browser the logo is > > > hidden and only the text shows. This is done for maximum usability > > > benefits and Google confirmed on an earlier posting I made asking > > > about this that this kind of usage was okay. Someone from Google > > > actually looked into it and confirmed it.
> > > Note that the text says EXACTLY what the text on the logo also says so > > > the same info is being served to both the users and the search > > > engines. We don't benefit in any way, except that the text looks > > > better :)
> > > Any other theories? I'm really keen to tap the wisdom of all the > > > really smart people here as if no one here can figure it out, it must > > > really be a G bug....
> > > Sam
> > > On Aug 28, 10:03 am, kklynnt wrote:
> > > > Sam, I'm going to post something on here for others to check out....I > > > > just found it and maybe someone can take a quick peek.
> > > > It looks like you are hyperlinking an image (ideally it should be > > > > text) and then I think you have a error in your css that will make it > > > > appear that you are hiding text. I'm so sleepy right now I can't > > > > think. Maybe someone else will check this out further...it's bedtime > > > > for me. Kerry
> > > > On Aug 27, 8:04 am, Sam I Am wrote:
> > > > > Hi everyone,
> > > > > Those of you posting more regularly will already know our site was hit > > > > > by some kind of Google penalty about 4 months ago. From one day to the > > > > > next, Google traffic was more than halved. I think we have tried just > > > > > about everything now, but there has been ZERO improvement. I'll list a > > > > > few recent things we tried and I'd be keen to hear what else we can > > > > > do.....
> > > > > * redesigned the site and made all navigation text links to avoid a > > > > > possible red flag with images as navigation. Cut down somewhat on load > > > > > time due to this. > > > > > * already had no followed a lot of external links, but have now > > > > > encrypted them and stuck them behind a robots.txt file. > > > > > * given google an incredibly strict robots.txt file to limit what it > > > > > has to crawl. > > > > > * made some redirects which were 302 (to our login page) into a > > > > > 301...
> > > > > Emails to both Matt Cutts and Adam Lasnik went unanswered (of > > > > > course).
> > > > > The site is large, has 20 000 back links being reported in webmaster > > > > > tools, an active community of users and only unique content. It should > > > > > be easy to crawl and is well structured from a semantics point of > > > > > view. Everything is on topic and there are no bad neighboorhood > > > > > links.
> > > > > We expect to rank top ten for terms like 'travel blogs', 'travel > > > > > forums', 'wiki travel guide', 'travel photography' and related terms, > > > > > all without quotes. In Yahoo and MSN we do. Actually we ranked top ten > > > > > for 'travel forums' until a few days ago but that seems to have > > > > > dropped now too (I didn't think it could get any worse, but it > > > > > obviously might?!).
> > > > > The site is still being crawled and some pages still rank well, but > > > > > terms like 'wiki travel guide', for which there are only 4 global > > > > > sites that exist do not see us in the top 50 (I give up after that).
> > > > > Aside from of course suing Google for dropping a site that complies > > > > > with their terms (I would seriously consider this if I thought it > > > > > would give me the answer as to what has caused this) what else can I > > > > > do? I'm worried some hack/proxy issue/negative seo has caused this to > > > > > happen but without Google sharing what they see as the problem, it > > > > > seems impossible to fix since it can't seem to be found by anyone :(
> > > > > I always thought Google wanted to list the best results per search > > > > > term, but have seriously lost faith in their ability to do this based > > > > > on our own experiences.
Hey, for the free inbound links, as long as Google doesn't discount their MFA site links, enjoy it while you can before they get slammed, if they ever do.
Or, you can do what I do, which is be anal and report them as a spam- dexing MFA site even though their inbound link may actually benefit my site. :-()
I figure I don't need nor want any "help" they may have to offer so the sooner they are gone, the sooner the "Net is just a little bit cleaner.
At least until the register a new domain name and start all over again. :-()
> This kind of stuff annoys me beyond belief, but I'm still right in > thinking this kind of stealing/harvesting of our content on sites paid > for by adsense can't harm us right?
> On Aug 29, 12:33 am, Sam I Am wrote:
> > > Not sure if this is an issue but a site that's been copied all over > > > the web has a listing for you as a sponsored link:
> > Poor Donna, she doesn't deserve to be copied that much! She's one of > > those great webmasters that puts tons of time into her site (currently > > coded in straight html for example) and has been running it for over a > > decade but I don't think follows much of what goes on in the > > discussions over paid links, like probably 99% of the other webmasters > > out there. The link is clearly marked 'SPONSORED' and that's what > > google wants (it's the only form both humans and robots can see). In > > any case, that ad is being pulled (thought it actually had already > > been pulled) based on totally unrelated reasons. There is of course no > > way that google knows if I paid for that link or perhaps got it in > > return for a favour done (interestingly enough they've said paid > > reviews in blog posts are not considered paid for example?!). I'm > > personally not of the impression that sites buying links are being > > penalized simply because it means we could all go out and get our > > competitors banned but it's food for thought I guess. Google is yet to > > mention one case of this happening and the best Matt Cutts could do > > last week at SES was a case involving cloaking so that doesn't bode > > well...
> > I see Donna has a few serious proxy issues though so perhaps she's > > been penalized due to that Google proxy bug and in return all sites > > linked to from her site are also getting hit. Hmmm, interesting.
> > On Aug 28, 11:31 pm, JLH wrote:
> > > Not sure if this is an issue but a site that's been copied all over > > > the web has a listing for you as a sponsored link:
> > > Perhaps you ran into the paid links buzz saw. Sure they say they'll > > > penalize the sites that sell links, but the webmaster guidelines say > > > the opposite.
> > > "Buying links in order to improve a site's ranking is in violation of > > > Google's webmaster guidelines and can negatively impact a site's > > > ranking in search results."
> > > On Aug 28, 4:03 pm, Robbo wrote:
> > > > Sam
> > > > I see your homepage is back in the indexes and with a very nice entry > > > > too!
> > > > Also:http://host-tracker.com/ > > > > will give you the individual and average response times of your site > > > > from 30 centers across the world. It gave me averages of 9 seconds > > > > for your site and 0.6 sec for two of my sites that I just used for > > > > comparison (not like with like as mine are MUCH smaller than yours).
Craig, the problem is that there are literally billions of these sites. I don't know where to start.... I've filed a few but when you flip a page and have to start all over again you just give up.
Of course the whole problem wouldn't exist if Google was stricter about their adsense policies, but they must be making millions off these sites so likely aren't going to do that until someone figures it out and gets the word out in big media. I'd expect there's at least a 100 000 of these linking to us under their different subdomains and what not; could that be a problem at that scale?
> Hey, for the free inbound links, as long as Google doesn't discount > their MFA site links, enjoy it while you can before they get slammed, > if they ever do.
> Or, you can do what I do, which is be anal and report them as a spam- > dexing MFA site even though their inbound link may actually benefit my > site. :-()
> I figure I don't need nor want any "help" they may have to offer so > the sooner they are gone, the sooner the "Net is just a little bit > cleaner.
> At least until the register a new domain name and start all over > again. :-()
> In short though, no, they can't hurt you. :-)
> Craig
> On Aug 29, 7:54 am, Sam I Am wrote:
> > Hmmm, should I be worried about caches like this (hover over the links > > to see the quality sites underneath... this is not our site!):
> > This kind of stuff annoys me beyond belief, but I'm still right in > > thinking this kind of stealing/harvesting of our content on sites paid > > for by adsense can't harm us right?
> > On Aug 29, 12:33 am, Sam I Am wrote:
> > > > Not sure if this is an issue but a site that's been copied all over > > > > the web has a listing for you as a sponsored link:
> > > Poor Donna, she doesn't deserve to be copied that much! She's one of > > > those great webmasters that puts tons of time into her site (currently > > > coded in straight html for example) and has been running it for over a > > > decade but I don't think follows much of what goes on in the > > > discussions over paid links, like probably 99% of the other webmasters > > > out there. The link is clearly marked 'SPONSORED' and that's what > > > google wants (it's the only form both humans and robots can see). In > > > any case, that ad is being pulled (thought it actually had already > > > been pulled) based on totally unrelated reasons. There is of course no > > > way that google knows if I paid for that link or perhaps got it in > > > return for a favour done (interestingly enough they've said paid > > > reviews in blog posts are not considered paid for example?!). I'm > > > personally not of the impression that sites buying links are being > > > penalized simply because it means we could all go out and get our > > > competitors banned but it's food for thought I guess. Google is yet to > > > mention one case of this happening and the best Matt Cutts could do > > > last week at SES was a case involving cloaking so that doesn't bode > > > well...
> > > I see Donna has a few serious proxy issues though so perhaps she's > > > been penalized due to that Google proxy bug and in return all sites > > > linked to from her site are also getting hit. Hmmm, interesting.
> > > On Aug 28, 11:31 pm, JLH wrote:
> > > > Not sure if this is an issue but a site that's been copied all over > > > > the web has a listing for you as a sponsored link:
> > > > Perhaps you ran into the paid links buzz saw. Sure they say they'll > > > > penalize the sites that sell links, but the webmaster guidelines say > > > > the opposite.
> > > > "Buying links in order to improve a site's ranking is in violation of > > > > Google's webmaster guidelines and can negatively impact a site's > > > > ranking in search results."
> > > > On Aug 28, 4:03 pm, Robbo wrote:
> > > > > Sam
> > > > > I see your homepage is back in the indexes and with a very nice entry > > > > > too!
> > > > > Exacept what happened to "Travel Blogs" ?
> > > > > I hope this cheers you up!
> > > > > Also:http://host-tracker.com/ > > > > > will give you the individual and average response times of your site > > > > > from 30 centers across the world. It gave me averages of 9 seconds > > > > > for your site and 0.6 sec for two of my sites that I just used for > > > > > comparison (not like with like as mine are MUCH smaller than yours).
> I'd like some time to read through and study this. This has caught my > eye because as people do try to build functional sites for mobile > platforms,etc. then CSS and how Google will treat all this will come > into play.
> With your particular site, I can see that accessing it via a mobile > device would be a huge perk for the user. It would be a real shame if > Google's system dinged someone for trying to build something that > gives them the needed effects. (Not sure that is what is happening)
> Thanks for the link...I definitely want to study this concept some > more. Kerry
> On Aug 28, 3:49 pm, Sam I Am wrote:
> > Don't worry Kerry, you won't be the first or last not to find anything > > else 'wrong' :) I think that's why some of the experts aren't chiming > > in, although I wish more people would chime in and say they can not > > find anything wrong as it might send a signal to Google that they > > really need to look at some of those cases.
> > Sorry, but you'll have to dig through some nasty comments back and > > forth by a few to the second page where Susan comes in and answers > > based on discussing it with Matt Cutts. My original question was based > > on our entire navigation being like this, something we have since > > changed.
> > I should also point out that this technique stretches far and wide on > > the internet among web designers and developers alike as it is the > > only way right now to get good looking text/graphics but still have a > > fall back option. So from a usability point of view, some things like > > this are absolutely necessary (for those users browsing on browsers > > that don't show images, you'd still like them to see what the image > > that is there says). My point being that it is not only for mobile > > users where the handheld stylesheet kicks in, but is indeed largely > > for usability, something that I also point out in that other thread is > > HIGHLY recommended by Google. And as you can see in this case, the > > text on the logo actually matches the text in the span to the letter, > > so there is no benefit one way or the other for us (plus of course we > > do rank for Travellerspoint already). I know there was a case where a > > known seo'er was so called 'caught' for using a similar technique, > > even though his case was perfectly excusable and blown way out of > > proportion by some fanatics, the thing that made it stand out for > > those that felt really strongly about it was that the text did not > > mirror the image.
> > > Have you considering using the mobile CSS to apply text in place of > > > the image? -- basically i believe you would want to control everything > > > related to the desired mobile styling with the mobile CSS exclusively. > > > I think you could safely hide the image here and apply text in it's > > > place within this style sheet with no problems.
> > Do you mean use a display:none on the mobile style and then add a > > content:before/after? That second command is not accepted by all > > browsers and would probably cause some problems. In essence this > > method is actually worse than what we are doing as well because you > > are physically adding content to the page that doesn't exist there to > > start with. If that isn't dodgier than clearly showing the word(s) in > > your source, then that would be beyond absurd.
> > On a side note; from what I understand Googlebot doesn't crawl the > > stylesheets at all just yet, so there should be no way Googlebot could > > even get this wrong. But I might have missed the announcement to that > > effect!
> > On Aug 28, 10:25 pm, kklynnt wrote:
> > > Sam, this is really the only thing I can see. Can you provide the > > > earlier post. I'd like to see what they said and in what context. > > > Having your site's data on a mobile is an awesome capability and of > > > course I see exactly what you mean in terms of styling to suite the > > > mobile platform's functionality.
> > > I think there may be a conflict here with how this is being > > > implemented. Hiding the image in the mobile style sheet and trying to > > > add text to that area in the browser's stylesheet might cause problems > > > for the Googlebot.
> > > Have you considering using the mobile CSS to apply text in place of > > > the image? -- basically i believe you would want to control everything > > > related to the desired mobile styling with the mobile CSS exclusively. > > > I think you could safely hide the image here and apply text in it's > > > place within this style sheet with no problems.
> > > As it stands now it looks like the browsers are relying on the order > > > of the cascade to render your desired effects. I'm not exactly sure > > > how Google is going to treat each individual style sheet when it is > > > used in this capacity.
> > > I'd just like to double check here so we can get you back up and > > > running. :)
> > > Kerry
> > > On Aug 28, 3:24 am, Sam I Am wrote:
> > > > Hi kklynnt,
> > > > Yes, that css is indeed hiding the word Travellerspoint when the page > > > > is viewed in screen mode. If you are on a mobile browser the logo is > > > > hidden and only the text shows. This is done for maximum usability > > > > benefits and Google confirmed on an earlier posting I made asking > > > > about this that this kind of usage was okay. Someone from Google > > > > actually looked into it and confirmed it.
> > > > Note that the text says EXACTLY what the text on the logo also says so > > > > the same info is being served to both the users and the search > > > > engines. We don't benefit in any way, except that the text looks > > > > better :)
> > > > Any other theories? I'm really keen to tap the wisdom of all the > > > > really smart people here as if no one here can figure it out, it must > > > > really be a G bug....
> > > > Sam
> > > > On Aug 28, 10:03 am, kklynnt wrote:
> > > > > Sam, I'm going to post something on here for others to check out....I > > > > > just found it and maybe someone can take a quick peek.
> > > > > It looks like you are hyperlinking an image (ideally it should be > > > > > text) and then I think you have a error in your css that will make it > > > > > appear that you are hiding text. I'm so sleepy right now I can't > > > > > think. Maybe someone else will check this out further...it's bedtime > > > > > for me. Kerry
> > > > > On Aug 27, 8:04 am, Sam I Am wrote:
> > > > > > Hi everyone,
> > > > > > Those of you posting more regularly will already know our site was hit > > > > > > by some kind of Google penalty about 4 months ago. From one day to the > > > > > > next, Google traffic was more than halved. I think we have tried just > > > > > > about everything now, but there has been ZERO improvement. I'll list a > > > > > > few recent things we tried and I'd be keen to hear what else we can > > > > > > do.....
> > > > > > * redesigned the site and made all navigation text links to avoid a > > > > > > possible red flag with images as navigation. Cut down somewhat on load > > > > > > time due to this. > > > > > > * already had no followed a lot of external links, but have now > > > > > > encrypted them and stuck them behind a robots.txt file. > > > > > > * given google an incredibly strict robots.txt file to limit what it > > > > > > has to crawl. > > > > > > * made some redirects which were 302 (to our login page) into a > > > > > > 301...
> > > > > > Emails to both Matt Cutts and Adam Lasnik went unanswered (of > > > > > > course).
> > > > > > The site is large, has 20 000 back links being reported in webmaster > > > > > > tools, an active community of users and only unique content. It should > > > > > > be easy to crawl and is well structured from a semantics point of > > > > > > view. Everything is on topic and there are no bad neighboorhood > > > > > > links.
> > > > > > We expect to rank top ten for terms like 'travel blogs', 'travel > > > > > > forums', 'wiki travel guide', 'travel photography' and related terms, > > > > > > all without quotes. In Yahoo and MSN we do. Actually we ranked top ten > > > > > > for 'travel forums' until a few days ago but that seems to have > > > > > > dropped now too (I didn't think it could get any worse, but it > > > > > > obviously might?!).
> > > > > > The site is still being crawled and some pages still rank well, but > > > > > > terms like 'wiki travel guide', for which there are only 4 global > > > > > > sites that exist do not see us in the top 50 (I give up after that).
> > > > > > Aside from of course suing Google for dropping a site that complies > > > > > > with their terms (I would seriously consider this if I thought it > > > > > > would give me the answer as to what has caused this) what else can I > > > > > > do? I'm worried some hack/proxy issue/negative seo has caused this to > > > > > > happen but without Google sharing what they see as the problem, it > > > > > > seems impossible to fix since it can't seem to be found by anyone :(
> > > > > > I always thought Google wanted to list the best results per search
> Craig, the problem is that there are literally billions of these > sites. I don't know where to start.... I've filed a few but when you > flip a page and have to start all over again you just give up.
The real problem is, for you and me, as webmasters, they are not that much of a direct threat. Some would say they dilute available PageRank which may or may not be the case and they obviously make it bad for searchers, but for you and me, just and annoying nuisance.
> Of course the whole problem wouldn't exist if Google was stricter > about their adsense policies, but they must be making millions off > these sites so likely aren't going to do that until someone figures it > out and gets the word out in big media. I'd expect there's at least a > 100 000 of these linking to us under their different subdomains and > what not; could that be a problem at that scale?
About the only way they could be more strict is by manually reviewing each site. I don't know that they don't already do that but I sort of doubt it.
A better idea would be to find some way to algorithmically knock them out of the indexes so the benefit of creating a MFA Spam-dexing site would go away totally. No one to bring "you " free traffic so no one to see "your" ads so no reason to have ads or even a spam-dexing site in the first place.
So two possibilities, police the advertising side which can lead to legal issues or remove the incentive to abuse the advertising.
This is all academic though as sites like those, for webmasters in general and your situation specifically is not really an issue.
> > Craig, the problem is that there are literally billions of these > > sites. I don't know where to start.... I've filed a few but when you > > flip a page and have to start all over again you just give up.
> The real problem is, for you and me, as webmasters, they are not that > much of a direct threat. Some would say they dilute available > PageRank which may or may not be the case and they obviously make it > bad for searchers, but for you and me, just and annoying nuisance.
I like to try and think as a searcher, not a webmaster, so I just see them as plain being bad.
Although google might be strict on getting the adsense account, they obviously do not check at all where the ads are being run. It should be a no-brainer to be able to filter out cases where ads all of a sudden appear on x million domains all with the same bit of adsense code. The ones where it appears on a few hundred are not likely to be a problem, so you wouldn't have to bother with those.
Or you tie the code in with the top level URL. If you want to add some code to a different url, it needs to be added first etc. All of this is relatively easy to fix by Google with their thousands of brilliant minds. Anyway, it's kind of off topic and I have just come to accept that as a content site we are going to be ripped off all the time.
Any more ideas pertaining to the site and things we can try? I'd be keen to get your thoughts on the blog issue you mentioned and my follow up. Ie. do I block all links from www.travellerspoint.com to the blogs or all links from the blogs back to www.travellerspoint.com?
> > Craig, the problem is that there are literally billions of these > > sites. I don't know where to start.... I've filed a few but when you > > flip a page and have to start all over again you just give up.
> The real problem is, for you and me, as webmasters, they are not that > much of a direct threat. Some would say they dilute available > PageRank which may or may not be the case and they obviously make it > bad for searchers, but for you and me, just and annoying nuisance.
> > Of course the whole problem wouldn't exist if Google was stricter > > about their adsense policies, but they must be making millions off > > these sites so likely aren't going to do that until someone figures it > > out and gets the word out in big media. I'd expect there's at least a > > 100 000 of these linking to us under their different subdomains and > > what not; could that be a problem at that scale?
> About the only way they could be more strict is by manually reviewing > each site. I don't know that they don't already do that but I sort of > doubt it.
> A better idea would be to find some way to algorithmically knock them > out of the indexes so the benefit of creating a MFA Spam-dexing site > would go away totally. No one to bring "you " free traffic so no one > to see "your" ads so no reason to have ads or even a spam-dexing site > in the first place.
> So two possibilities, police the advertising side which can lead to > legal issues or remove the incentive to abuse the advertising.
> This is all academic though as sites like those, for webmasters in > general and your situation specifically is not really an issue.
Let's talk about Spam-dexing MAF sites after we get you back to some level of decent positioning again, OK? You've got some interesting ideas in the area, although maybe not well fleshed out but we could go on and on about what is easily the scourge of search for a looooonnnnngggg time! ;-)
> Any more ideas pertaining to the site and things we can try? I'd be > keen to get your thoughts on the blog issue you mentioned and my > follow up. Ie. do I block all links from www.travellerspoint.com to > the blogs or all links from the blogs back to www.travellerspoint.com?
I really REALLY hate to see you have to do it but I think at this point, it is about the only thing left to try.
Whether you block them in a robots.txt or nofollow them I think is up to which ever is easiest both for the doing and hopefully later, the undoing.
I'm really sorry I can't come up with anything better. :-(
But, I would like to make it clear why I am suggesting what I am and why it should only need to be a temporary solution.
I also want to walk through the "logic" I am using to come up with the suggestion so that you can make sure it makes some sort of sense to you.
For the moment, forget about different people having different blogs you provide space for, lets say all of the blogs combined were the output of you and your staff.
So, you would have a website of a couple dozen thousand pages or there abouts on the topic of travel, right?
Were that the case, would a site be penalized for linking to other pages within the site almost no matter how they did it?
Of course not. A given method of linking may not be the most efficient in the world but to see a huge drop due to it? Not likely.
Inefficient linking would more than likely be enough of its own punishment that there would be no need for Google to do anything specific about it.
On the other hand, what would happen if even one of your site's couple dozen thousand pages had something on it Google didn't like? That could cause a problem for any site, right?
I think I remember you saying you have been through all of the blogs and haven't been able to find anything but from my way of thinking, it can't be the linking that is the problem but instead, it seems it would have to be one or few specific pages/blogs that are the problem.
Trying to find that one or a few pages, without knowing there was even anything to look for in the first place would be like looking for something in a haystack but not even knowing what that something was.
So, If you block the links to all of them, you should be able to get a quick measure of whether or not the problem is being caused by something linked to and then if so, try to narrow down what might be linked to that could be causing the problem.
It just wouldn't make any sort of sense for a blog provider site to suffer just for linking to contained blogs so the only other thing it could be, link related, is if one or more of the links to internal blogs lead to something Google doesn't like.
What ever is wrong, there obviously has to be a reason. Google may seem unfathomable at times but not being easily understood does not mean there isn't a method to the madness.
The only thing to do in many cases is try to narrow down the madness. :-()
Blocking access to the links would seem the quickest way to hopefully find a direction to try to pursue because without a method, all we are left with is madness.
|
