As I was stumbling around the web today, building some slides out for my Pubcon sessions next week, I came across some search results I couldn't help but blog about. Have a peek:
Check out that fourth result from hugthecook.com. It's currently down, and has been that way since at least October 20th (when Google's cache shows the "invalid website" page). The last date I can confirm it was up and available is January of 2008, when Archive.org shows a functional site. What makes this more perplexing is that page's complete lack of inlinks, according to both Yahoo! and Linkscape.
The query "stand mixer review" might not be the most competitive phrase, but it's certainly not insignificant either, and there are a lot of what appear to be very competitive and robust sites and pages ranking in positions 5-20. So, in the spirit of my new TV addiction to medical mystery shows, I thought I'd ask you - what's making that URL rank so well at Google?
If there's interest, perhaps we can have an ongoing series with more SEO diagnoses in the future :-)
Dude, is everyone serious on not knowing why it ranks? Granted, it's broken - but who knows the time limit/error code combo Google uses in regards to that. The root domain has 3K links, including links from places like CNet and the Washington Post and is 4 1/2 years old. There should be no debate on why it ranks for such a non-competitive phrase. The only debate would be why when it's down... but again, who knows Google's threshold/combo there.
Oh, and as for them beating out sites like cooking.com etc, just look at the pages:
https://web.archive.org/web/20071228014409/https://www.hugthecook.com/tkq_standmixer.aspx
it's optimized for that EXACT target phrase. Meanwhile, cooking.com for instance:
https://www.cooking.com/products/toprated.asp?DeptNo=3000&ClassNo=0324
never has the solitary word "review" once on the page - NOT ONCE... it doesn't even have "reviews" as text - only links.
The cookography.com page is highly optimized, but the root domain isn't as strong.
Thus why when you pluralize the search to "reviews" hugthecook.com moves to second page.
So, assuming that in June/July, before the site went down, it had good on-page optimization, I still think it's fairly interesting that the "no links" thing hasn't killed it yet. Domain trust is one thing, but once there's been 0 links (internal or external) pointing to a page for multiple months, it seems pretty odd - Yahoo! & Live certainly aren't ranking it anymore.
And if we're talking about domain authority, only-cookware.com and epinions.com (which both appear to do pretty good on-page targeting as well) are also ranking behind it.
To me, it's more about why they'd leave it in the SERPs so long. With client projects we've worked on, it takes only a week of downtime (sometimes just a couple days) before we see the listing fall to page 2/3/4.
BTW - did you see my theory above about some sort of temporary 302 issue possibly being responsible for this?
epinions.com only has the word review in the breadcrumbs and heading - and there is an additional word with the target phrase with each. Additionally, "stand mixer review" appears at the front of hugthecook.com's title tags vs. the end of epinion's.
Only-cookware.com has the phrase in the middle of it's title tag and only has the word review on the page singular form in TEXT one time. Additionally, as with the epinions page, it has additional words in the heading phrase it appears in, lowering the keyword weight to much less than the 100% of the hugthecook.com heading.
Additionally, hugthecook.com, when live, had more unique text: https://www.google.com/search?hl=en&q=%22The%201000-watt%20motor%20hummed%20through%20our%20bread%20dough%20test%20like%20it%20was%20whipping%20cream%22&btnG=Google+Search than the epinions page which is formed from product description feeds: https://www.google.com/search?q=%22From+Viking+comes+this+Colander/Sieve+for+7-qt.+Mixer%22&hl=en&filter=0 and only-cookware.com has not only less text, but the partial phrase "stand mixer" doesn't appear in the page again in TEXT vs. the hugthecook.com having a much higher keyword density with near double digit appearances on the page.
You're quoting it as having no links according to yahoo, but Yahoo has already recrawled the whole website since it has been down (and interestingly enough, they have also not removed it) so therefore, all internal links have been wiped. But, since Google is still ranking the page, they may be referring to their oldest cache without an error of the entire site, which might help the page keep it's internal backlinks. Again, the only oddity is why it STILL ranks considering the error code, but that relates to Google's treatment of error codes, NOT why the site ranks in the first place.
I agree with you on the error code/still ranking, but WHY it ranks is SEO 101.
But we do know that Google has visited the page at least once (and almost certainly many more times) since it's been down, meaning the keywords they should have for it are "bad request invalid hostname." The lingering on-page factors thing, if accurate, would certainly be new to me - I'm accustomed to Google considering on-page changes as soon as they re-crawl and see them. Like I said, the client projects we've worked on with similar issues have shown much, much faster drops (and there have been some great examples in Q+A, too).
I'm not arguing with your assessment of keyword usage. I don't think anyone is. If the page were active (and if I didn't suspect that it has been down since June/July), I wouldn't have any trouble seeing it rank there, but that's not the case. I'm curious because of the long downtime and strangely maintained ranking, not because they didn't (many months ago) have good on-page SEO and a strong domain.
>>>meaning the keywords they should have for it are "bad request invalid hostname."
You're assuming that they don't have a command to their bot that says "find 400 error up to three times, keep original cache" or something - again, this is likely something to do with how Google is treating this error (and subsequently, yahoo again, hasn't delisted them either).
>>> what's making the URL rank so well at Google?
Actually you did ask why the site ranks so well... why it ranks so well is all in the SEO. Why Google seems to be ignoring the error is a diff issue, IMHO.
That's what I was really asking about - not why the original page would have ranked, but why, in this instance, as compared to others, the page is staying in the index and ranking well in the SERPs. The time limit on this one just seems to be super long (maybe a result of high trust in the domain? Although I've seen rankings fall much faster on domains I thought were were much bigger and more authoritative than this one).
Maybe it's Lupus?
it's never Lupus.
True. How about Autoimmune? Fits the symptoms...
EDIT: Doh. Lupus is autoimmune. sigh.
Trying to figure out an individual result in Google is like trying to figure out why a sand particle is where it is in the big wide ocean. Good luck
I'm finding these more and more over the past few weeks, and appear to be caused by "wormhole" links... which is the term we've been for old (very old), trusted backlinks. Unfortunately, as you have mentioned, the content on the target domain isn't always valid (or active).
It started about... hmmm... 6 months ago, and some of those errors still exist today for valuable keyword terms. Do a search for "yellow pages", and you will find mcp.com still floating around -- even though they are completely invalid. The domain used to be owned by the publisher of the Internet Yellow Pages... and that is enough to rank the domain today, even though DNS changed hands, and the site is completely invalid for the term.
Last week I found a geocities cite that also didn't resolve properly -- and ranking for a very nice term. This time there were only 10 backlinks reported by google -- but all very old, and from trusted sources.
Lesson learned... place your links now, they will start working 5 years from now. ;) Oh -- or buy those websites with wormhole links and 301 them to do your bidding.
Stevensearchgeeks.org
I have a feeling it is ranked on some previous history. According to whois a Lisa Popiel is the administrator. Hmmm. Popiel does that ring a bell?
Similarly:
If I run a Google search on "data entry services atlanta" Google's LOCAL BUSINESS RESULTS shows 2 sites: Natacha's Data Entry Services and DataPlus Data Entry Services. Natacha's has been DOA for a very long time and results in a "SORRY THE SITE YOU REQUESTED HAS BEEN DISABLED." This site has been number one in Google Maps listing in Atlanta for months and months. I even went into the Google Webmaster blogs and posted about it. No change. I don't care, I am DataPlus and thus end up being the only listing one sees without clicking through. but it seems a waste.
"If there's interest, perhaps we can have an ongoing series with more SEO diagnoses in the future :-)"
Absolutely
I really think this is an example of "jumping the shark" as opposed to a "differential diagnosis".
This is a site that has been down for less than 3 weeks and while that seems like a long time, you have to look at the flip side of the problem. How quickly should ranking sites be removed from the index if they are experiencing a technical issue? Immediately? 1 day? 1 week? 1 month? If your site was experiencing a technical issue that couldn't be resolved right away and youwere delisted, I'd bet that you would be emailing, twitting and IMing every Google person you know even the janitors.
If anything, this is less a SEO issue as it is a Google freshness issue IMO.
Actually, I think the site's been down/unavailable since June or July. Here's my thinking:
Just my guess - it fits most of the evidence, but there's a lot of other explanations. I just think it's very strange that with no content and so few links, the site would remain ranked so well. Obviously, searchers can't be particularly happy with that result, and I'd bet the site owners really wish Google would list hugthecook2.com instead (although, without the redirect, the engines can hardly be blamed).
@Rand.
Why did you say DMoz listing ‘updated’ to point to hugthecook2.com. Because from the Last Web Archive of the Cookware page (dated Feb 13, 2008), HugTheCook was not listed at all.
https://web.archive.org/web/20080213130029/https://www.dmoz.org/Shopping/Home_and_Garden/Kitchen_and_Dining/Cookware/
So it can be that the first entry of HugTheCook into DMoz was for domain hugthecook2.com. Or maybe I am missing something.Also, your guess about the redirects seems true.Seems like they made changes to hugthecook.com last time on June 4, 2008. See Google’s cache here:
https://64.233.183.104/search?q=cache:kTXN75m2jdoJ:www.hugthecook.com/CMSHelp/+site:hugthecook.com+-intitle:invalid&hl=en&ct=clnk&cd=3
The interesting thing however is that this page was cached on Oct 7 and from hugthebook.com.
This is Google's cache of https://www.hugthecook.com/CMSHelp/. It is a snapshot of the page as it appeared on 7 Oct 2008 08:47:27 GMT. The current page could have changed ....
So Google was able to access it that time. Now searching for site:hugthecook.com intitle:"weber q300".We can see the result and in the serp, the URL shown is hugthecook.com/…. However, if you go to the cache, the URL shown is hugthecook2.com/…
https://64.233.183.104/search?q=cache:HIvvGGkV8dQJ:hugthecook.com/detail.aspx%3FID%3D7746%26prod%3DWeber--Q300-Stationary-Gas-Grill%26code%3DW-426001+site:hugthecook.com+intitle:%22weber+q300%22&hl=en&ct=clnk&cd=1
This is Google's cache of https://www.hugthecook2.com/detail.aspx?ID=7746∏=Weber--Q300-Stationary-Gas-Grill&code=W-426001. It is a snapshot of the page as it appeared on 29 Oct 2008 22:47:06 GMT. The current page could have changed ...
This page was cached on Oct 29, just around 10 days back. So maybe the redirects were working until very recently and maybe 302 redirects like you say.
P.S. I don't have much experience so I may have missed some obvious things.
Yeah - with DMOZ, since the only listing in archive.org for that page is in February, we only know that since then hugthecook2.com has appeared, and it was only registered in July. No way to know for sure whether Hugthecook.com got into DMOZ sometime in between (but I think there's at least an equal chance that it did, since a submission in July would be a really fast turnaround time for DMOZ - they're usually 6-12 months behind or more on submissions).
The rest of the case you make sounds solid to me, but I do think that hugthecook.com has been down since at least July, and that's why the new hugthecook2.com exists - can't be certain, of course, but it makes sense.
From recent previous experience: I had a site and took it down to move it and change the focus of the site to a completely new topic. I left the site down for about a month and was quickly droped from the SERP's. I put the site back up and within 4 days the site was ranking again, although a little lower in the top 10 but still in the top 10. My page rank was returned a couple days after that.
All my pages stayed indexed for some time but just didn't show up in the SERP's.
If a site is experiencing issues the site should be removed from the SERP's as soon as it is discovered and when the site is back up it should be put back in. I know I hate clicking a link and going to an error page but seems a little worse when it happens from a SERP link.
Google must read SEOmoz.
The site is not listed in my search results today. ;-)
Interesting topic. I found similar several months ago with this search query.
Note URL www.guitarhero4.co.uk.
Several months back, it was live (August) with related content. The site went down for a while and now it has content back up, but has absolutely no relation to the search term "Guitar Hero 4".
Even after reporting this as a misguiding link to the big G, it is still there.
Hello just came acroos this site when i found something interest in here.....just this time some lack that google did ........superb guys
Hi Randfish, it is surprise to some thing like broken link in google results. I was wondering how can this happen since google has a state-of-art technology for crawling?
Same here, at least not on the first page for "stand mixer review"
In the spirit of 'why is it featuring?', I'm a bit perplexed at to why the top ranked site even appears for the search term "uk's best web sites".
Google has flagged it with a "This site may harm your computer." It has had this alert on it for months - you would have thought Google would drop it down the rankings.
I had quite a few pages that still appear in Google for some niche keywords, so I put the pages back up. The pages are html that were related to a project and I have a blog. Once when tracking back some visits I saw that lot of people where searching something related to IT governanace and finding (or not) the html pages and then they probably just erased the page to look at the domain.
Is there any SEO reason to leave a title tag open and stuff it with keywords? I ask because I was trying to figure out why a site ranks for a key term. The site or page in question is at https://www.boutiquekittens.com/Picture-Purr-fect.264.0.html and it has a bunch of spammed keywords and an equally spammed open title tag . Is this a mysterious black hat trick?
So much for Google relevancy....
I have seen redirected sites do this.
although i think the diagnosis of this oddity is fun, I'm kind of direct and would like to just ask someone at Google. the point you stand to make is that the search result is 4th and is not useful at all to a searcher. more info that could help them in being interested in giving any feedback is the popularity of the search term. call me naive but i would love to hear what they say even if it's non specific, as i would assume they'd have to be.
even many time I have come across results on first page which have link broken. As said in comments made above may be they were trusted domans before or something else.
But this fact can only be answered by google.I guess its a loop hole in algorithm from google's side.
Keep an eye on it, if it's not fixed, I say it will be gone 3 months, top. Any takers??
I haven't had much time to look at the site in question but I think this is an excellent idea for a series of blog posts - I think watching expert SEO's try and diagnose problems is one of the best ways to learn about seo.
The only concern I'd have is about 'outing' information about the site owners? Personally it doesn't bother me but I wonder if some people in the community might take issue with it?
I've got a few such strange search results myself I might pass your way :-)
I think anyone who monitors serps will see quite a few odd results. I posted one on Sphinn about how Aaron Wall's SEOBook is ranking in top 10 for 'lindsay lohan nude'. But I know he wouln't be competing for that keyword. Nor I think anyone coming from that keyword would be likely to subscribe to Aaron's blog because of that. In fact, it was a spammer who created a profile on seobook (as Aaron mentioned in comments).
I think if you open the door of outing once, then people would start blogging about their competitors or their clients' competitors. They would raise even small issues to make them look large.
Instead, I would recommend what Kim (from Cre8pc) said when Rand outed NP recently: Pick up a successful site that is ranking and deserves that ranking. From there, we can analyze why they are ranking. This will also teach us guys what makes a site rank and deserve that.
One of my favorite sessions at PubCon last year was a series of fun debates that was very much "What Not to Wear" of sites' SEO strategies. I know Rae was on the panel and... I can't remember who else. But that dialog below between Rand & Rae about what's what on the page is highly entertaining and should most definitely be a regular video podcast. I talked to Rae about it, but I guess she doesn't like crowds or cameras (sad! she did great!).
I'd love to see that, though. A regular online series of opinionated and talented SEOs who analyze a site or two a week and have an entertaining back-and-forth in the process. And you wouldn't have to worry about outing a site since you'd just call for submissions. It's not hard to get takers for free personalized SEO advice.
i just check and to me its working properly... i cannot see any reason why its not working for you or why its working for me but i can access exact page
Some of us also addressed a similar experience over at WMW, and i think that the general consensus was the domain trust and authority status given by Google, regardless of the content being displayed...still played a huge factor in ranking..almost to the point of if you are in good standings, its pretty hard not to get dinged.
https://www.webmasterworld.com/google/3702873.htm
Can't it be a certain trust was created previously and that would work as ranking factor?
I remember earing something from eric ward on that but not sure if it still works this way tought.
I like this - site diagnosis. Although not the same, I have run into an issue with a site that is clearly the content/relevancy/usablity winner, has more links, more link value, yet sits beneath some sites of somewhat 'lesser value' in regards to presentation, content and SEO techniques employed.
Although Google, we are told, loves fresh content, the number two site on Goolge hasn't had content updates in years -- and the site is younger than the one I speak of, in question.
This is one of those anomolies where what we hold as true and industry-standard gets tossed out of the proverbial window.
Posts like this that asks the great SEO community for input is a learning experience for all who read it - I'd like to see more of this type of questioning and suggestions.
~ Joe
Maybe there's another page on a different domain with inbound links that redirects to the broken page?
Wormhole Links.
All the way.
Link quality perhaps? I'm clueless, really. Gotta keep reading the comments for help...
296 URLs indexed in Google for hugthecook, all showing the same Page title and description.
"Invalid Website, The requested domain name is not configured for any web site..."
From looking at the results, looks like they have an expired licence key for their ecommerce solution...
Is it something to do with the fact that it's a 400 error and not actually a 404 error? I suppose if it's an old page and has quite a bit of authority, Google could be giving it the benefit of the doubt that it might come back.
I think you're on to something with this.
HTTP Status Code: HTTP/1.1 400 Bad RequestContent-Type: text/htmlDate: Sun, 09 Nov 2008 00:20:24 GMT
Response type 400 is defined
The request could not be understood by the server due to malformed syntax. The client SHOULD NOT repeat the request without modifications.
RAND: I'd love to see more diagnosis of oddball situations. I have a couple myself.
This happens in Google when the competing sites aren't strong in SEO. Google will rank the sites well just for being somewhat relevant. Like you said before the site doesn't have anything keeping it to be ranked well. But I think in 3-4 months time the value of that page will go down substantially and other more relevant pages will be in its place.
I suggest e-mailing those competing websites and offer optimization, and improve the quality of the sites for google.
I'm not sure - epinions, cookography, howstuffworks, cooking.com, cnet and consumer reports all rank below them. I think it might be a stretch to say the "competing sites aren't strong in SEO."
If the page is unavailable, how does that translate to its relevancy?
This matches what I've seen in the past as well. There are three factors at work here:
- Not many other pages are optimized for this text
- Before erroring out the original site had large root domain trust
- The site is now errored out, but this apparently hasn't had as much influence in this case
I think it's mostly a combination of weak keyword competition matched with the large trust in the root domain.
The more interesting question to me would be, is this page showing up for "strong" competitive keywords? The original site apparently had a lot of work put into it optimization and link-wise. Then I think we can start wondering at the utility of Google including errored out site. If this site shows up where people are actually trying it could be more worrying.