Give Google feedback on “noresults” pages

I recently posted asking what issues the Google webspam team should tackle in 2009. Getting this outside feedback is really handy, because it’s helpful to compare our internal perceptions against what annoys hundreds of people outside Google. After the first 150 or so comments I did a very rough tally of suggestions to see what issues are reported the most.

The #1 complaint (20+ comments) was “empty review” sites. Tons of people said something along the lines of “I hate when I search for [productname review] and then click on a result, only to land on a page that says ‘There are no reviews for this product.’ Grrr.” Many times such pages are not created to deceive users, but “no results found” or “empty review” pages can be annoying and contribute to a poor user experience. They can also fall under Google’s webmaster guidelines in a few ways:

Use robots.txt to prevent crawling of search results pages or other auto-generated pages that don’t add much value for users coming from search engines.
….
Don’t create multiple pages, subdomains, or domains with substantially duplicate content.
….
Avoid “doorway” pages created just for search engines, or other “cookie cutter” approaches…

If a site does add a lot of value otherwise, our typical policy response would be just to remove from our index individual low-value or auto-generated pages from our index, without removing the entire site.

Given the number of people who complained about this, I’d like to ask for your help to gather examples of such pages. Specifically, you can help by sending us concrete examples of “no results” or “empty review” pages. I want the actual url that annoys you. We will be taking a close look at the reports, so this is your chance to provide example “no result” pages directly to the webspam team. Here’s how to report a bad user experience.

1. Go to our authenticated spam report form. You’ll need a Google account to sign into our webmaster console. This form is available in dozens of languages, not just English.

2. In the “Additional details” section, make sure you include the word “noresults” (all one word, all lowercase). Feel free to fill in the other fields with info if you want.

3. Provide an actual “no results found” or “empty review” example url. For example, in the “Additional details” section, the text can be as short as this:

When I searched for [blue widget reviews] on Google, the url http://www.example.com/review/2008?q=blue+widget looks like it has reviews, but when you click through you see the message “No Comments | 0 Positive Reviews | 0 Negative Reviews. Overall Rating: No Ratings. Leave Your Ratings or Reviews here!” The page doesn’t actually have any information or reviews of the blue widget product.

That’s a perfectly fine report. The main data I want to gather are specific site urls that demonstrate the “No reviews found” issue. Again, don’t forget to include “noresults” as a keyword in the report so that we can extract all the specific feedback. If this is something that you feel passionately about (and it appears that several people do), thanks in advance for pointing out which specific pages give a bad user experience.

120 Responses to Give Google feedback on “noresults” pages (Leave a comment)

  1. Hi Matt,

    I also posted a comment in your post about the spam results but that comment did not appeared.

    Matt, nowadays I see so many videos full of spam in Google search results. I do not understand how such scammers and spammers rank so high in google search results.

    I think people are using You tube to spam google search results and it will be great if Google web spam team can do something about it.

    I have reported multiple videos via the spam form but so far, I still see those videos in Google search results.

    Regards

  2. Huh? Why so worried about pages not having reviews yet? Maybe a review will be made the next day. Maybe the review there previously was garbage and needed to be deleted?

    Why don’t you just include another report called” content doesn’t match keyword” That way you can get all the pages reported to you that suck and don’t match the keyword for which you ranked them for.

  3. angilina, my comments are pre-moderated, so it might just not be approved yet. I’ll keep reading over there, so please use that thread to suggestion 2009 goals for webspam.

    Jaan Kanellis, the fact is that a lot of people were annoyed by these types of pages, so we’re going to check it out. If 2009 is about more communication and transparency, and that’s what people are concerned about, then I’m happy to dig into it more.

  4. Matt,

    I don’t have any of those types of review pages but please don’t get overzealous about this.

    It’s inappropriate to flag these types of pages in robots.txt but I could see asking site operators to include a NOINDEX in the review pages that have no review.

    Some of these pages can have perfectly valid content, such as articles about a product, just nobody ever bought or reviewed the product yet.

    Alternatively, they can modify the keywords on the page so that it doesn’t attempt to rank for a review until some reviews are posted.

    However, with that said, I have seen a bunch of pure affiliate sites (crap) that attempt to rank for product reviews just to yank in active buyers.

    Even worse are all these domain review sites.

    Here’s an ironic example:
    prsitecheck.com/site/www.mattcutts.com

  5. Matt – I’m glad you’re opening up the communication between Google and Webmaster’s. The most frustrating thing about building good content is worrying that somehow my work will be taken out of the index because it falls into a certain category – even if it’s intention is not to be bad.

    I think it’s a smart idea to lower the ranking of a page that has no reviews.

    Also in reference to “no results found” pages. With some sites some items people may be seeking currently have no results at the moment but may later. How do you deal with that?

  6. Never thought I would find this in your source:

  7. The fact that those pages exist is not the problem. The problem is that they rank and draw traffic.

    It seems like you are hinting quality publishers will pull these URL’s. But Google needs to be more specific about how to handle these pages. ie What server code etc etc.

    We’ve been trying to get an answer to this for months in Google Groups with no answer.

    Here’s the situation:

    Basically we create a page for a product or review or whatever and then the content on that page needs to be pulled for terms of service violations or some other reason. Obviously the page is going to look empty because we as webmasters have removed the offending content. However the page/URL still exists. How should we handle this page? 404 301 302. Assume that the page is empty for months but eventually we might get legit content to fill it in the future so we don’t want the page/URL to go away for good.

  8. Hi Matt,

    http://www.google.com/search?&q=buy+CPQ175KB

    Any of the top results, basically.

    This is a minor tangent that includes reviews – but also purchases. It came to mind because I had recently wanted to buy a keyboard I liked a great deal, and spent hours looking at 0 reviews, but also at “0 in stock” buy pages, in nearly all cases of which “substitutes” are given. Not sure this “should” be something gone after, but it is becoming more common to find this sort of bait-switch as discontinued products age in the index. And it seems to fit the same basic litmus of not showing the user what was promised in the SERP summary.

  9. Thank goodness, some of us spend allot of time reviewing products and servicies only to be beaten in rankings by non review pages.

    Will be submitting my selection which includes “opinions” pages that have no content.

    Cheers 🙂

  10. @angilina I agree SERPs are full of scam videos. Black-hats calling it ‘entering the night club’… when you alone want to walk in the night club guard will stop you but if you hang out with “Bono Vox and Paris Hilton” one on one shoulder the other on other you jump into night club… Bono = YouTube or ‘famous trusted’ and ‘Paris = attraction collector soc. bookmarking’ where spammers insert links to their scam.

    I hope Matt will get an idea how to filter that scam. First thing I would do match usernames because they usually use the same account names across different services and those low frequency keywords (usernames) if appearing in clusters links targeted videos cut them off. I think there is term for such a method… used in signal intelligence theory.

    About empty review websites, what about empty how-to like those scummers from “ExpertsExchange” I click [X] every time I see them in SERPs on user search wiki.

    What about obscure RSS aggregators like boxxet.com? They land users on crappy pages full of adsense and they drive surfers around theoretically user needs to make 2 clicks to go to original source behind some #333333 link [read original] and those guys are 9,000 in Alexa (because they ‘add value’ come on!)

  11. Truthfully – I think this is kinda lame. Not finding a product on an ecommerce site like Walmart or even Amazon is not uncommon at all! Finding empty review fields on those same pages is just as plausible also. Have you yourself visited some of the ebay popular product pages lately? How about the footer ebay automagically plants on the bottom of their sellers stores? How about the fact that all of their stores are 301 redirected domains that pass pagerank to popular product pages?

    (Already mentioned and ignored in GWM Forums)

    Why not target some of the BIG players who pay Google hundreds of thousands in adwords revenue, instead of the little guys who will be put out to pasture with another manual review?

    This post is nothing but an attempt to get more people reporting different types of sites as possible spam, so the manual review team can eliminate more from the index based on individual values and opinion of the reviewer.

    Should JohnsMart have any less relevance than WalMart? No it should not… but a manual reviewer may think that since WalMart is so popular and JohnsMart is not… well no-result pages from WalMart are OK, but JohnsMart goes out with the bathwater!

    Google has always stated that they wanted nothing done to please the search engines. Then came nofollow… now comes noresults. I would venture to estimate there are “many thousands” of small-time companies, who cannot afford a full time webmaster and are using an out of the box ecommerce application, that could and obviously WILL be pushed out of your index due to another “don’t do it for search engines” idea from Google webspam team!

    It would actually be refreshing to see Google “Be a Search Engine” WITHOUT manual intervention! We ALL know how well the ODP has turned out, I’m afraid Googles new methods of being the Internet Police is going to bite back at some point Matt.

  12. Thank you VERY much Matt.

    Once I get in to work tomorrow, I shall be going over old ‘noresults’ on my break as I have a few dozen to hand.

    Still… adding actual reviews from Google Product Search in to G Universal Search as persay x3 products can be shown at top:
    http://www.google.co.uk/products/catalog?q=ipod&oe=utf-8&cid=7546580999858004957#ps-reviews
    That, would be splendid! 😀 Nom, nom, nom.

  13. Morris Rosenthal

    Matt,

    In addition to review sites, you might look at questionable start-up Wiki and supposed social networking Q and A help sites. I noticed that with auto repair, there are huges sites seeded with very specific questions that don’t have answers. A made up example would be “How do I replace the fender on a 1967 Mustang Fastback”, where top ranking Google results would include pages titled “How to replace a fender on a 1967 Mustang” and on arriving at the page, you’d find the title, some advertising, and an invitation to answer the question. I don’t know if they generate the questions from real searches, automatically, or have a live person doing thousands of variations on a theme, but they waste a lot of when you’re searching for how-to information on very specific subjects.

    Morris

  14. I must be missing something here. A legitimate site that allows people to comment on a certain product or service risks getting banned (or “reviewed”) from Google because Google can’t figure out that a certain page from the site isn’t worth indexing (yet)?

    In other words, the site clearly needs a page that allows for new comments to be added for a product/service (a page one would presumably get to by doing a site search or through internal browsing); it’s the search engine’s job to discover that a particular one of those pages isn’t worth indexing since it doesn’t have a review (yet). The crawler could ignore these when it first crawls the page, or the indexer could choose to tier it higher/lower, or ignore completely. Or the ranking algorithm could rank the pages lower.

    It could be that you’re simply using the spam reporting form as a way to learn this issue because that’s the only feedback channel available; but this is clearly not a spam issue. Scaring webmasters by pointing to webmaster guideline is also silly. This is merely the inability for search engines (Google here in particular) to understand what to do with this class of pages, why blame webmasters for Google’s technical inability (in this case) to deal with such scenarios?

    Amit Kumar
    formerly PM, Yahoo! SearchMonkey

  15. IncrediBILL, given that this was the #1 complaint, I do think it’s worth looking at in more detail.

    himynameis john, I welcome whatever data you want to send us. Thanks!

    Morris Rosenthal, which domain name is doing this?

  16. to name few big one

    http://wiki.answers.com/Q/Example_for_matt_cutts_noresults_pages

    just wonder who is ‘asking’ all those questions

  17. Dave (originial)

    The #1 complaint (20+ comments) was “empty review” sites

    BUT, that doesn’t mean the same complaint is the #1 complaint for Google users, does it?

    I mean, there are x Million unique people using Google each day and 250 complaints posted on your blog, which means the numbers are close to meaningless AND 230 of them DO NOT find “empty review” *pages* a major problem.

    Why not run a survey/competition with normal users (with no agenda/angle) and offer x prizes for x users drawn from a hat? I think Google will find the results actually have some merit then.

  18. Please consider extending this “no results” purge to coupon code sites as well. There are many of them that are essentially black holes of potential coupon codes. They have a page for any term + coupon code even when none exists. I’m not sure why these sites rank well in Google, but the manage to do so with a frustrating frequency.

  19. I don’t think empty review sites are bad and I’m pretty sure that I’m one of the people that suggested that is a problem. My feedback was as a user and not a web developer.

    Just to clear it up. Empty review sites should not show up if you use the word ‘review’ or similar in the search.

    Many sites that have reviews are ecommerce sites or even affiliate sites (not necessarily thin affilate sites). So they should come up i product searches. But if I’m looking for a review I want a real review. I don’t want to see 200 sites that all have the same amazon reviews.

    For instance, there was this one product I was looking for information on and I didn’t find it till like 8-10 pages deep. It was this blog post that used the product exactly like I needed to use it and detailed using it over the course of days with pictures and even a follow up.

    It wasn’t like an iPod or expensive digital camera where there would be a ton of affiliate sites. It was a type of concrete.

    Matt,

    One thing I think you guys are doing, which I hope you continue to do, is keep track of people that create throw away domain names. When they come up with a new one and they’ve been marked repeatedly as spam, do something about it.

  20. James,

    The answer to your situation sounds obvious. You have a page, then pull out the content, that page is just an empty template.

    Through a 404 response. You essentially pulled the page.

  21. I mean throw a 404 response.

    Matt. Please take it easy on grammar and spelling in your changes 🙂

  22. Wow! Thanks for the swift action in this regard. I gave you that example of Baker’s cyst at Merriam Websters with “No entries found” in the earlier post. I will pass on that thru webmaster console now and similar ones that I come across from now on.

  23. Hi Matt,

    I’d also ask that babies not be thrown out…we’re mostly unhappy with sites that focus on offering reviews but don’t realy provide them or only offer scraped content.

    I have clients with eCommerce sites who may add a plugin for their product pages for a review functionality. After a new feature gets added to a site it often takes time for the users to make use of it. If these product pagesm have high authority, they could conceivably rank for review searches even though they aren’t optimized for it. It wouldn’t be justified if these pages were pulled from Google’s index – perhaps some sort of filter for a higher threshold to rank for review searches?

    Thanks for listening,
    Janine

  24. Actually, Matt, I think Bill made a pretty good point, and he made it on at least two levels I don’t think most people would have considered in their haste (as usual).

    1) If a review page for Product X doesn’t have any Product X reviews, yes that seems rather pointless for those who are looking for reviews on Product X. But what if that same page has unbiased information on where to buy the product, maybe a price comparison, a Google map of nearby stores, or some other uncommon or unique functionality that actually serves a purpose for someone who may be searching for “Where can I buy Product X?” or “Product X price comparison” or “Stores that sell Product X”? If you start cracking the whip on sites marketed as review sites because they don’t have reviews, I would suggest (as Bill quite rightly did) that you’d be hurting the user experience in other areas.

    2) The problem with review sites is that they’re quite often biased (intentionally or otherwise). I’ve even seen cases where the “user” posting a business review was clearly an employee or the owner of the business in question. I’m not referring to cases of “yeah, this looks like a paid shill but I’m not 100% sure”, either…these are cases where no ambiguity exists whatsoever.

    Here’s a classic example. I don’t care how much a vocal minority hate “empty review” sites; there is no logical reason to assign preference to a crystal-clear case of self-promotion and an intentional abuse of a review site to boost “ranking” within a voting context over a review site that at least is honest enough to admit that it doesn’t have reviews for a particular product and/or service.

    For those of you who think that this is a one-off example, there are plenty more where that came from…and that’s just very specific query on one site. Examples like the one above are just the ridiculously obvious. Shills are quite often a lot more subtle than this, and the majority of shill reviews would be completely invisible to the untrained eye.

    3) If a crackdown ever came into play on review sites containing empty review pages (which is a scenario that is pretty easy to envision, given the question), shill reviews and/or auto-generated reviews would populate the review pages and there would be no way to tell in most cases whether or not the reviews were legitimate.

    I don’t see how this is a path anyone would want to go down in the Plex, and clearly I’m not the only one.

  25. Whoops…the query link never showed up.

    http://www.google.com/search?sourceid=navclient&aq=h0&oq=“check%20us%20out”%20site:c&ie=UTF-8&rlz=1T4GGLL_en&q=%22check+us+out%22+site%3acitysearch.com

    This is what I tried to link to where it says “there are plenty more where that came from”.

  26. Damn…looks like WordPress has another bug. Guess you guys will have to copy and paste that URL. (Probably why the link failed the first time).

  27. Matt,

    I didn’t say it wasn’t worth looking into, I just said don’t get overzealous with stuff like the robots.txt which is inappropriate for dynamic sites, NOINDEX is more appropriate.

  28. I don’t really get what is wrong with the EXISTENCE of blank review pages. It’s your algorithm that sucks and ranks them higher than useful content. Anyway, when searching for “some gadget” – why are there som many review sites in the first place? That would make sense for “some gadget review”

    Robots.txt is very old standard with extremely limited flexibility and great portion of inconvenience. It is simply not feasible to do the exclusion based on robots.txt.

    You should come up with some better solution – what about HTTP headers that would instruct google how to handle the page?

    Googlebot could simply add ‘i-am-robot’ header and server could decide what to do. I’d love to be able to server google e.g. 412 with some explanation header in the meaning: ‘this-page-exists-but-i-dont-want-it-to-be-indexed-now’ and no content. Machine time saved, bandwidth saved, less indexed pages, less spam.

  29. Dave (originial)

    I don’t really get what is wrong with the EXISTENCE of blank review pages. It’s your algorithm that sucks and ranks them higher than useful content.

    At last, someone with common sense. Agree that blank review pages are NOT the disease, they are ONLY a symptom of Google’s algorithm.

  30. Lukas, i agree, It’s the algorithm that sucks and ranks them higher than useful content.

    Matt, you guys know these results should not show up. Asking for reports is in my opinion merely having a poll to see how many people realy do bother.

    It also could backfire for you, asking for reports and people not seeing any positiv change in the results. Just like lifting penaltys a while ago and poof spamsites pop up again in the results. The algo didn’t pick them up and still doesn’t.

  31. Multi-Worded Adam,

    I take your point, but still, if i search for a review of a product, i would like some results that do actually have a review of the product. Price comparisson and where to buy are nice, but in the mean time i get no answer on which product suits me if i have two or three left to choose from. And that was the main thing i was looking for. 😉

  32. I too am uncomfortable with this, one of the aspects that I have reccomended to site owners is to put reviews and comments in – this is because I am twice as likely to buy from a site which does have reviews on products – especially as they are often negative, excluding an entire product page as it doesn’t have any reviews yet would seriously harm sales, and how do you put on a link which says “review this product” I hate to say it, but thisn’t good.

    Look at http://www.Comet.co.uk for example, not all products have been reviewed, but many have …

    Unless we forge our own reviews, this is a dangerous reccomendation …

  33. Dave (originial)

    NOINDEX is more appropriate

    No it isn’t. By putting the onus on site owners for a short-fall in Google’s algo you will create *even more* fake reviews of products. It is not the site owners fault/problem that Google *chooses* to crawl & index blank review pages and then ranks them above other pages.

    When/IF Google stops sponsoring conferences where spammy tricks are taught, removes TBPR (this alone would stop the most Webspam and email spam), takes a top-down approach on Webspam and gets serious by banning sites (6 months min) instead penalizing or wrist slapping, I will help Google in every way I can.

    Handing out free disposable Web sites to any Tom, Dick or Harry for blackhats to test their latest tricks, certainly doesn’t help the situation.

  34. I must agree with IncrediBILL. There websites that offer unique information on the product, not only reviews.

    Should those sites/pages get removed from the index? Not really, maybe only for searches related to reviews (including the keyword reviews, opinions…).

    Also, I know that some people look for websites where to leave reviews for a product because they’re specially happy/angry with it. Does this means that they shouldn’t be able to be the firsts ones leaving a review?

    Matt, look in to this, but removing the page form the index is not the solution. Look for ways to identify these pages and simple do not return them on related search. THIS will really improve the quality of the index.

  35. Whilst the discussion of reviews is fine, isn’t this just a special case of the problem of negatives in search engines which I posted a comment on many years back.

    If you search for “2147221247” you get results for “-2147221247”, which are probably what you wanted in this case but…. well you get the point.

    If you search for “reviews” you get text including “no reviews”.

    I’m sure Google engineers must know the problem, presumably one could tweak the general algorithm to downgrade keywords prefixed by negatives or negations “-” “no” “zero” (for English), and see if it improves the results of all searches. I’m glad I don’t have to write the code for this – well actually it sounds like quite a fun challenge in many ways but I’m not sure I’d want to be doing it with weekly progress reports.

  36. Cheers, Matt.

    I’ll be sure to report those pesky empty pages from now on!

  37. on behalf of Morris Rosenthal (and since I was less than impressed when I went there yesterday) http://wiki.answers.com/Q/How_do_you_install_rear_leaf_springs

  38. Don’t get too hung up on the “review” example.

    The type of sites this issue is common with:
    1. Review sites (game, movies, software, hardware)
    2. Guides, Cheats (single player cheats, shush you multiplayers 😛 ), Tips, Hints.
    3. Software/Hardware guides/tips/hints
    4. Articles/Information

    Basically sites that have content but also have content pages with no content, aka false positives.
    Sites which have a message “Sorry we do not have any information of this yet” “no review yet” “no cheats yet” “no guide yet” or just a blank article.

    Whether Google’s talented algo guys can find a way to push such result down a few pages for you when you search matching keywords I have no idea, I just hope they do.

    For example, “Good” sites like wikipedia do not really have empty/placeholder pages, the Wikipedia system/managers purge such pages, and pages tat do have content has rules on the quality of content.

    Ever tried searching for say an older game, only to end up with IGN or Gamespot in the top results, go there and all you’ll see is is the title of the game, the publisher if your lucky and the year, and nothing else, no review rating, no description of the game, no age info, maybe not even a genre.

    These are false positives in my view, templates, somebody entered that title into the system, the CMS built the page, but since the CMS can’t magically create the content for it (we stll need humans for that, and preferably good writers too) it remains empty.

    A lot of other sites don’t, some sites do not really know how bad it is to lead surfers down a blind ally, but if Google starts ranking those a bit different then they will fix it, or the CMS devs will improve the way their CMS pages are created (and mght even make use of ROBOTS to help search engines along). There are also other sites that do this on purpose that is the worst, only to draw traffic.

    If Google can find a way to filter these and push them down it will improve search accuracy for the users, the relevancy of the results are better, less false positives, Google search becomes better than competitors (which has to follow suit to keep up), and the well meaning but ignorant sites that has “empty” pages will get less frustrated visitors and thus their image is not tarnished more.

    The end result of this is easier to find and better content when users look for something, improved relevancy.

    What I miss and which I know would be a pain to implement in a user friendly way, would be a search relevancy rating that I could just select.

    I know I tend to either A. open a result by Google in a new tab, or B. click it.
    B might be easier as I usually hit back in the browser, Google could allow a ranking option of 1-10 in that case.

    URL’s with high number of votes (thousands) that either very positive (8-10 real or exploitation?) or negative (1-3 real or exploitation?) for example could be flagged and sent to human review at Google so the algo and spam guys could evaluate it.

    The user rating could be shown next to the search results in Google as well (but not affect the position in the result obviously). Some form of minimum threshold before display would be needed too.
    IMDB uses a system similar to this and it work really well when a lot of people voted on a movie. (like thousands of votes, a few hundred is always a bit sketchy obviously)

  39. Agree with IncrediBILL – NOINDEX is a much better way of handling this – it’s usually extremely difficult to keep the robots.txt updated to block these URLs, and NOINDEX is something you can do at the CMS level reasonably easily.

    I would appreciate Google updating their webmaster guidelines to state “REP” rather than just “robots.txt” (so these other no-indexing options are declared valid)

  40. All reviews sites start out as empty, unless they have made up reviews on them. Instead of complaining about them, how about you help them out and post their first review, get the ball rolling. Even better maybe you can keep ranking them in Google so that actual users can post actual reviews on them because if you flag them all then out with come all of the fake auto-generated reviews. Every action causes an equal or greater reaction.

  41. Good set of arguments to look at especially the one pointed out by Lukas: “It’s your algorithm that sucks and ranks them higher than useful content.”

    I think this is a bitter truth. Google is giving ultimate importance to the titles than to the contents inside, that’s why a “no results” page of a site with attractive title props up on the top than a less optimized title! Am I correct?

  42. Just one more thing to grab your attention Matt:

    People without a Google account will not take the pain to create an account and then report the “no results” pages! Of course, every web surfer won’t be a webmaster! Can’t it be made easy for a common man to report the “no results” pages as a secret ballot without posting the links here?

  43. It’s not just review sites. I get a large number of “links from an external page” where I can’t find my site on their review or search engine but they seem to benefit. These sites seem to have no corelation to the content of my site. for example from toolboxhero.com stats:
    http://ww23.rr.com/index_results.php
    http://finder.cox.net/main
    http://www.become.com/shop?&q=research&reab=100&nofilter
    http://www.staples.com/office/supplies/home?storeId=10001&langId=-1

    These seem to be automated websites that don’t add anyting to the web.

  44. If I have a review site that dynamically creates pages, and one of those pages is a page that says, “no reviews at this time” and Google decides to rank it, please tell me why should I have to now make changes so Google won’t rank it.

    There is nothing on the page, I’ve made no attempt to decieve anybody, I’m telling you there is no review yet Google’s algo has decided to rank it above other sites with reviews for that product. Seems to me the problem is with the Algo and not my site.

    So, yes go ahead look into this while real spam goes untouched but don’t come up with some lame procedure that I as a Webmaster must now take becaue your algo is not smart enough to figure out there is no review on the page.

    Give me a break.

  45. @Ian M
    you can use noindex or the robots.txt for such pages. But the user/webmaster has to tell Google and other search engines, that such pages shall not be indexed, because they have no value to the user.

    Often the robots.txt is an easier way to exclude pages, because you can do it at one place and the webmaster has the control over it. Clients (CMS-users) often makes mistakes, cause they don’t have the technical background and don’t think about noindex or index. They just want to write articles and do their job. That’s ok. But as a webmaster you are responsible for your clients and should prevent them from making mistakes. And the robots.txt is – or better was – a good place to do that.

    Unfortunately it seems, that Google changed the behaviour of dealing with the robots.txt. Excluded pages via robots.txt shown up in the SERPs (without description, only the URL), if they were linked elsewhere in your content.

    For example: If you will exlude the filetype doc, cause you will not have them shown up in the SERPs, you were able to do it via robots.txt:
    User-agent: *
    Disallow: /doc/

    or if the CMS-user put it elsewhere:
    User-agent: *
    Disallow: /*.doc$

    Now this doesn’t work anymore. If you provide the document as a download for your users, the URLs shown up in the SERPs and blow up the index. Now you have to use the x-http-header instead of the robots.txt. robots.txt is obsolete in this case. Maybe to save some traffic, but not to exclude pages or documents 🙁

    greetz
    Robert

  46. Wow Matt!

    The more I read this the more annoyed I become. I can not believe your spending spam team time on this no review pages algo issue. Shouldn’t that be up to another department? You should be fighting the real spam that’s making it’s way into the index instead of quirks with the algo ranking pages that shouldn’t be ranked. Did you move to another department or something.

    The gold old days of Googleguy are sorely missed.

  47. The suggestion that the site be added to robots.txt is problematic on two levels. First, not all sites have reviews as their primary purpose. If they add a page without reviews to the robots.txt page, that page won’t be indexed for the other portions of content on the page. Second, for sites with a large number of products available, keeping the robots.txt file up to date could be resource intensive for the site if they had to list every page that did not have any reviews available.

    Also, I agree with some of the other readers that if Google started penalizing for a lack of reviews, many sites would just start creating junk reviews of products. It’s possible that Google might then be able to use a spam filter to try to determine if they were automatically generated, but at that point it might make more sense to adjust the primary site filter instead.

    As someone who likes to do a lot of research about a product before purchase, I understand why other readers have given you this complaint, particularly after the holiday season. In fact, I’ve had the same reaction to seeing the number of Amazon.com products with no Reviews while still giving the same emphasis to the empty review section on those pages. On the other hand, I’ve often been frustrated by how existing price comparison/review sites (and I include Amazon in this list) provide the information. I would love for sites to be created that provided this information in a better way and I am concerned that a number of bad apples will make such projects impossible, especially if Google decides to penalize sites that provide new tools to better process the existing reviews.

    I’m not sure where Google’s balancing point should be, but sometimes it’s too easy to name the result one gets but did not want as spam, when in reality, it’s how one gets the result that is wrong. Perhaps reviews need to be a special search case where a particular page on a site is placed in the context of the site as a whole. Of course, that would then lead to an issue of trying to determine if a searcher for a product review cares more about lots of shorter reviews (like an Amazon page) or longer, more in depth reviews (like Engadget and CNET).

  48. Adam is right, most of these pages have more than simply reviews, or a lack of reviews.

    Yes I am biased, but my pages show much more than just reviews. How can you say a person is searching for reviews specifically and not just the location of a business or its hours or if they accept credit cards, or if there is any parking.

    The only way to do this is to change the algorithm, blocking or demoting pages is an unfair and biased way of accomplishing a goal and its not right. These are dangerous waters, you are saying that Google will tell people what they are looking for instead of the other way around.

  49. Matt,
    One I did not think of before and do not see above is getting up to date results. I think I am not alone when say this is bad and frustrating, but getting results back that are out dated. When searching for answers to say SEO questions on how this search engine handles “A” and all the results that come up are from 2005 and before. In the ever changing SEO field this information is mostly out dated and useless. Thanks for the post and keep us all on our toes.

  50. My vote goes to doing something about Review Center. That site jas to be one of the wors offending parties. Also, in many cases, one ancient (potentially inaccurate) review keeping the entry high in the SERPS. it’s a reputation management nightmare, as well as being full of empty review pages.

  51. Let’s face it; the algo doesn’t suck anymore then a developer.

    Technically, a programmer can construct a site that if there are NO reviews then there should be no page, right???

    A smart developer would construct the site to only show a page on the site if all the content is available including any reviews.

    It is quite obvious (Matt if I’m over the boundary here just let me know) from reading all these comments and bashing the webspam team, you must be a spammer and you’re afraid Google is going to catch on to what you and the “No Reviews” scheme is up to???

    Now, Matt and the Webspam team are asking for our help to strengthen the search results with REAL information. Allow them to do just that.

    Thanks for looking into this Matt!

  52. I think people missed Bill’s point about NOINDEX.

    The point Bill tried to make (or at least the way I read it) is that NOINDEX is the more appropriate choice for dynamically generated pages as opposed to using robots.txt to list individual dynamically generated pages that should not be indexed. This makes a great deal of common sense when one thinks about it, since listing individual dynamically generated pages within robots.txt would either require repeated manual edits and probably a lot of human error, or a somewhat complicated text file open/edit if necessary/close process.

    It can easily be implemented into any “CMS” that is coded in a half-decent manner as well using the following order.

    Retrieve data from database.
    If data is inappropriate for search engine indexing (e.g. a dedicated review page with no reviews), use meta robots tag with noindex attribute.
    Otherwise, don’t bother with tag at all.
    Display header of page.
    Display data.
    Display footer of page.

    This would make sense for dedicated review pages (I’m not referring to sites here, but the individual pages themselves); however, not all review pages are dedicated review pages. They often contain addresses, phone numbers, possibly a map, possibly a driving directions application, possibly a description, possibly hours of operation, etc. and so on. Why should they be NOINDEXed? They’re of some value to someone. Bill’s right on this (as he usually is…and no I’m not kissing his ass, but I do agree with him on most things and he’s one of the few people online who has merited at least some respect).

    The “no review” problem basically exists in a vacuum. It only applies to a limited set of circumstances. I can understand people’s frustration when they look for a product review and they see a site listed within a SERP with no reviews; that doesn’t make a lot of sense either. But to summarily dismiss and eliminate all such review pages because they don’t have reviews is asinine. A lot of these pages contain useful information beyond just reviews, and a lot of the so-called “reviews” are nothing but shill posts, obvious and otherwise.

    Simon’s comment was the best on the subject; adjust the algorithm to remove any results containing “0 reviews”/”no reviews”/”be the first to post a review”/similar phrases from any query containing “review” and its derivative words and you’re all set. This isn’t an issue that requires everyone and their brother to post query results as of yet…adjust the algo first, see if there are any holes by asking this question again, then readjust as needed.

  53. Onlyme: If the developer only allowed pages with reviews to exist, reviews would never be written because the pages where people would go to write the first review would not exist. Having a blank review page is a prerequisite to having a content rich review page.

    The question is just whether or not the blank pages show up in the SERPS, which seems like something Google should be able to accomplish with the algos.

    Multi-worded Adam: Couldn’t have said it better myself.

  54. “If I have a review site that dynamically creates pages, and one of those pages is a page that says, “no reviews at this time” and Google decides to rank it, please tell me why should I have to now make changes so Google won’t rank it.”

    You shouldn’t be having to make changes *so Google doesn’t rank it*, you should be making them *because the page is BS*. Make your website not generate pages with empty content that nonetheless has the keywords in the title of the page. It’s deceptive. You could just as well argue it from the viewpoint of your own website’s search engine. If I go to Dell.com and search for a 386 DX, I don’t want Dell’s site to tell me there’s a 386 DX page only to clickthrough and find a “Sorry, we don’t make 386’s anymore, but you can always buy this shiny new Dell Dimension for $10000!”

    It really isn’t rocket science…

  55. @Dave – Bad analogy

    “…You could just as well argue it from the viewpoint of your own website’s search engine. If I go to Dell.com and search for a 386 DX, I don’t want Dell’s site to tell me there’s a 386 DX page only to clickthrough and find a “Sorry, we don’t make 386’s anymore, but you can always buy this shiny new Dell Dimension for $10000!…”

    Well, I would want the website to tell me that they don’t make them anymore AND suggest an alternative! thats intelligent – just a no-results doesn’t leave me any wiser.

    OK I really don’t get what we are supposed to do – say as an e-commerce manager I have a website http://www.MySite.com and on it there are hundreds of products, because my customers want(ed) it I added the ability to add reviews to help people in the buying process… As such I added a link to all the pages “add a review” of course to stop people clicking onto the reviews tab I added a “currently no reviews” or something similar, I would want the page to come up in search engines – for XYZ, so I wouldn’t want to exclude the page. I don’t care about being featured for “reviews of XYZ”, if people can’t find the product they can’t buy it – if they see that they can add reviews later after they have bought it, they might come back and add a review after.

    Does this make sense? Google have in the past promoted UGC as being a good thing, they even pull in reviews into maps etc.. so I don’t get what we are supposed to be doing…

    I refuse to reccomend to clients to exclude the entire page! I can’t add a 0 Review(s) tag (or can I ? ) as they don’t exist!!! And I also don’t believe in faking reviews.

    I haven’t yet seen the answer to this!

  56. @htnmmo

    Agreed. Throw a 404 after you remove page content and end up with an empty template.. That’s what seems right to me and most logical.

    I’m just concerned about re-indexing. If we create a page… let it get indexed… then later throw a 404 for six months.. then even later put it back up with fresh content, will it still rank or will the 6 month 404 have some sort of residual knock on effect?

  57. Morris Rosenthal

    Matt,

    I’ll have to do some searches and get back, didn’t write them down. Assuming I find the examples, I’ll post them back through another e-mail so they end up in your moderation queue.

    Morris

  58. I just realized something. The HTTP standard does have a status code relevant for this, and it’s also the code advised for HTML 5’s url ping responses.

    http://en.wikipedia.org/wiki/Http_error_code
    It’s: 204 No Content

    Matt, am I correct in assuming Google already support this one? (due to HTML5’s ping feature) ?

    If so then it should be moderately easy for CMS devs to make the CMS return 204 status codes on empty content pages. (where there is just the template but no actual content).

  59. Gah scratch my last comment, looking at the RFC this wouldn’t work anyway. hmm. any other codes that could be used instead?

  60. Morris Rosenthal

    Matt,

    First one I turned up was Wiki.Answers. Two types of examples below, but first, it occured to me after I fount them to Google the key phrase “This question has not been answered yet” on the Answers domain. Here’s the result:

    “Results 1 – 10 of about 476,000 from answers.com for This question has not been answered yet. (0.24 seconds) ”

    So that’s nearlya half million pages with no content, just ads!

    First example type:

    http://wiki.answers.com/Q/How_do_you_replace_the_TDC_sensor_on_a_renault_scenic

    It seems to me I see lots of these, though it is possible that there was an answer at one time and staff deleted it or something. But I checked Google cache and there was no answer then either.

    OK, I tried an experiment on WikiAnswers and submitted a question, using the made up axample I gave on your blog. There is now a page:

    http://wiki.answers.com/Q/Replace_fender_on_67_mustang&waAn=1

    So apparently they employ user questions to create the blank questions. If they get a real answer, wonderful, if not, I think they should prevent indexing. I suppose they would make the argument that indexing helps the asker find an answer, but given the time delays involved, my guess is their business model deoends heavily on questions and ads.

    Morris

  61. Matt, I really like that fact that Google or at least you on this blog get people involved. I am a fair play by the rules guy and whenever, I see spam pages like, empty review pages pop up, but also other worthless pages it upsets me not just as a user but as a webmaster who feels that there is some sort of injustice going on. However, over the years I would have to say the trend is in the right direction and I see less of this sort of thing, I think the Google algorithm is getting better. On that note ->

    Chess. Think of chess. At the start of computer chess, no decent player would lose to a chess computer. Now, no grandmaster can beat the best chess algorithms. None. This is why I think Google should stay with spam being weeded out by the algorithms, rather than human adjustments as much as possible and only use our feedback to improve the algorithm, as human errors of judgement can be unfair also.

    How is my comment related to your post? ->

    Beyond empty review pages, the next level of spam is empty review pages or other article that are filled with ‘auto generated content’ from programs that create content. Many of these programs are becoming quite good, however, to the human reader they are just as bad as empty pages. I think there should be reporting for 1) empty pages 2) pages that might auto generated in another way.

    Somehow both these offenses should somehow be factored into the Google spam algorithm. However, I am concerned the latter, is more threatening. That being said, think of chess. No matter how clever humans become in creating spam, think of the chess programs that can beat the grandmasters, eventually the algorithm will prevail if its a team effort as was in chess. (BTW I am human and a chess player. :)).

  62. OnlyMe: Amazon has tons of products that they don’t sell (they list third-party products) and do not have any reviews. Should they not have those pages? Should I not be able to buy a product because no one has reviewed it yet? Should I not be able to find the product through a Google search? Having just purchased an item from Amazon that didn’t have any reviews and having found it on Amazon using Google, I firmly believe that the pages should be in the Google index even though the items don’t have reviews.

    I can understand changing certain elements of the page when there are no reviews. For instance, perhaps reviews shouldn’t be a keyword when there are no reviews on the page. Another suggestion might be to remove any heading elements (h1, h2, h3, etc) around Reviews when they don’t exist. This means that the pages can keep a similar pattern between pages of the same type, which aids usability, while lowering the probability of ranking high for reviews of a product for which there are no reviews. At the same time, the page doesn’t get penalized and gets indexed for its other bits of content. The issue is the algorithm that ranks pages has to get far more sophisticated, recognize that these pages have a reviews section, that no reviews exist on the page and that the webmaster took special steps to appease Google.

  63. Say good-bye to all future new retail sites hoping to get consumer-generated content AND indexed in Google.

    This will probably go down in history as one of the most stupid, short-sighted mistakes the SEO industry has made in a long, long time.

    Google — think real hard about how you define “no results review spam” in your algorithm.

  64. [quote]It is quite obvious (Matt if I’m over the boundary here just let me know) from reading all these comments and bashing the webspam team, you must be a spammer and you’re afraid Google is going to catch on to what you and the “No Reviews” scheme is up to???[/quote]

    Yes, everybody who disagrees is a spammer. What a brilliant statement.

    Why should I now have to go back and recode sites the were done years ago just because Googles algo can not figure it out?

    This is called diveresion. Yes, lets get Matt and the spam team looking at something other than the real spam that is in the searches.

  65. Morris Rosenthal

    Matt,

    The operative phrase for finding many sites in Google based on the Q/A model is “Be the first to answer”. In some cases, it seems the sites are highly legitimate and generally taken seriously by participants, like Yahoo Answers. In other cases, the sites look like get rich quick schemes.

    I don’t know where you would draw the line, and I wouldn’t be surprised if some get rich quick schemes turn into genuine resources, despite the ownership. But their results spam up the index when they are empty templates, or when the answers are simply ridiculous, kids fooling around and scoring site expert points or something.

    Sorry about the spelling two posts back and thanks to Gimmster:-)

    Morris

  66. Frankly, I don’t see why much at Google these days seems to be rooted in the high school popularity contest mentality. Just because a site has activity doesn’t mean anything worthwhile is going on. Take blogs, for example, what is so compelling about most of them? Many of them read like bad country music written by Daryl and his other brother Daryl – you picked a fine time to leave me Lucille…yet, they tend to rank better than their equivalent but different platform counterparts, why?

    Then there are the twitter twits?, Is anyone really that interesting that I need to know their every move? Squidoo, the glorified affiliate platform, is another one. Every lens looks basically the same. Why do any of these useless web 2.0 sites even see the ranking light of day? Oh, I get it, so we can all gather and sing Kumbaya.

    Don’t even get me started on these useless directories. Why do we need to go to yet another directory (aren’t you suppose to be doing that?) and click through a gauntlet of links with countless pop ups and ads to find anything? Where is the value there? Oh, I guess having 3000 lawyers on one page makes it easier for me?

    The number and frequency of one’s utterances does not equal quality. We are all alive yet how many of us are so compelling that we would rank #1 just for being in motion? Not many, yet you somehow have managed to equate frequency of activity, sometimes no more worthy of attention than breathing, as quality? It’s pretty damn dull and pathetic.

    Your fuzzy math really requires some attention and review sites are the least of your worries. You are in serious jeopardy of losing my interest altogether. If you want to police something, police stupidly because it’s rampant and it’s viral.

  67. “I’m just concerned about re-indexing. If we create a page… let it get indexed… then later throw a 404 for six months.. then even later put it back up with fresh content, will it still rank or will the 6 month 404 have some sort of residual knock on effect?”

    James,

    If you have a site that Google visits frequently, that shouldn’t be a problem. In a day or so at most, it should come back online once it finds a link to it or it shows up in your Sitemap.xml file again. If it’s a popular site it should come up even faster.

    A day or two for a new page to be indexed isn’t going to kill you.

  68. I like the angle from ‘Dave (originial)’ in regards to data collection with a Google user survey, not just a ‘I read Matt Cutt’s Blog Survey’, but then again, that is what Google Search Algo does on a daily basis isn’t it, measure the masses?

    However, as a starting point at least, I think this blog post serves a good purpose – It is directed to people who work in/with/around the search marketing industry, so in theory, we should have more SE “experience” than most.

    But yes, can I stress the fact that I DO believe usability/listening to the masses is important… but I’m sure a more truthful account of this can be made though tweeking the algo and then looking at how search data and how it has effected usability that way.

    Now, when it comes to usability ONSITE. PFFT, I’m sorry but I disagree with quite a few of you here and this is my reasoning; If I search for “iPod Review”, then no, I do NOT want an empty review page with other suggestions of “Find local store”, “Price Comparison” etc… If I wanted that, that would of been an entirely different search altogether, quite obviously.

    Give the user what they want, not what you can palm them off with.

    And I agree, it is Google that needs to change its algo, not websites change their sites… again, isn’t that quite obvious? Otherwise you’ll get people doing what they’ve been told is ‘best practice’, ie, ‘noindex’ (if no content is found), but then you’ll get others spam away making blank pages regardless. I’m not saying a website should be spam flagged for these pages though, just G tweeking it’s algo. Besides, blank review pages serve a purpose – EVERY REVIEW PAGE HAS TO START SOMEWHERE – and that is with ZERO reviews prior to customers coming along to do just that, but they won’t be searching for “iPod review” to review it!

    I know I’ve said it like, twice now, but I do honestly think that when you search for “buy iPod” for example and x3 Google Product Search results are shown at the top – this should be done for “product X reviews” too, but from reputable companies only… Ciao.co.uk, Amazon.com/uk etc, etc… then the rest of the result can be battled out from all sites.

    Lastly, I DO believe even though 20 or so people picked up on this that;

    a) Algo tweeking for ‘reviews’ can be copied across other similar search queries, so it makes it worthwhile targetting this area first (especially, not that anyone needs reminding, but our economies are still grinding downwards, so reviews are more IMPORTANT to people)

    &

    b) Worthwhile to keep in mind out of that 20, many may of actually gone ahead and read others comments prior to posting. It’s a shame you couldn’t of kept all comments hidden Matt, prior to coming to any conclusions. Sheep will be sheep. 🙂

  69. Sequin – Reviews should be on the same page of the product and if there are reviews it’s displayed right under the product not a separate page…that will eliminate the separate ‘no review’ pages. Separate pages can work if it’s program to display ONLY when there is a review.

    tracy – those products should be appearing whether or not they have reviews. We are generally speaking of blank pages in the results that have no reviews.

    spamhound – well it’s either they are or they’re not. Why complain when webmasters can easily fix the problem…..only show reviews if there is a review.

    Simply put….if no reviews are available then the page shouldn’t be available. Developers can be creative on how to display reviews but just not on a stand-alone page.

  70. Even for the same page, why even mention Reviews if no review is mentioned. This will also help eliminate clogging the serps when there isn’t any reviews.

    Display only when there is a review.

  71. Matt, if Google could do anything about these empty reviews, it would be fantastic.

    But it is not all. Google search for products that are even moderately obscure, yields a lot of junk pages that go beyond “no review, submit your review”.

    For example, you can search for “hy5dv651622”. (an electronic board component).

    What you will see is a huge number of empty generated pages that say “call us for a quote on hy5dv651622”. All of these pages are computer generated junk. They are worth nothing to any user who looks for information on this component.

    I would guess that it is probably difficult to tell a useless page from a useful page, but if Google would do at least something about this proliferation of garbage, I would be delighted.

  72. Just look at how many people are worried about their spammy site…
    This means you Matt are going in the right direction, good job!

  73. Dave (originial)

    Matt, this all came about because 20 people (from roughly 1 Billion) *perceive* empty review pages as “spam”. Why are they spam? Why is the Google spam team wasting time on this, when Google default is to crawl and index every page on the Internet? Does your spam team really have that much spare time on their hands?

    It is so obvious that these pages are NOT spam and only a symptom of Googles algo it hurts.

    Come one Matt, your a smart guy, stop the Witch hunt and address REAL spam (the blackhats who create Webspam and prey on the ignorant) and leave the problem of empty review pages to the right Google department.

  74. Dave (originial)

    spamhound – well it’s either they are or they’re not. Why complain when webmasters can easily fix the problem…..only show reviews if there is a review.Yes, no kidding! The only problem is that Google is ranking empty review pages above relevant on non-empty review pages.

    There is no sure way to make (or rely on) Webmasters bear the burden for a flaw in Google’s algo. Even if there was, why take a band-aid approach and not fix the real problem?

  75. Dave (originial)

    spamhound – well it’s either they are or they’re not. Why complain when webmasters can easily fix the problem…..only show reviews if there is a review.

    RE: “well it’s either they are or they’re not” Err, no kidding. The only problem is that Google is ranking empty review pages above relevant non-empty review pages.

    There is no sure way to make (or rely on) Webmasters to bear the burden for a flaw in Google’s algo. Even if there was, why take a band-aid approach and not fix the real problem?

  76. Dave (originial)

    I think people missed Bill’s point about NOINDEX.

    The point Bill tried to make (or at least the way I read it) is that NOINDEX is the more appropriate choice for dynamically generated pages as opposed to using robots.txt to list individual dynamically generated pages that should not be indexed. This makes a great deal of common sense when one thinks about it, since listing individual dynamically generated pages within robots.txt would either require repeated manual edits and probably a lot of human error, or a somewhat complicated text file open/edit if necessary/close process.

    IF we limit ourselves to NOINDEX and robots.txt and ignore the real problem, then perhaps.

    Point is, there are many choices and the real problem is not Webmasters not using NOINDEX, it’s a flaw in Google’s algo.

  77. What would be awesome is if google recognized a “noresults” meta-tag that we could include in our pages that returned no results. That would be perfect for forums where undesirable posts weren’t deleted before google indexed them. Instead of generating a 404 page that is going to show up on google webmaster tools, we could have that automatically inserted in those “thread deleted” pages. I could find a lot of use for that meta-tag.

  78. what explains in this post means that even the honest review sites who are just starting out slowly and have yet to put more reviews can get booted for no good reason, only because some user reports it cause he thinks it just sucks. This is not the way to handle it. If anything, only the particular detail page of that review for that x product should get lowered in ranking, and not labeled as spam or the whole website getting booted out of Google SERPs. When the actual review for that product is added to the website, Googlebot should just re-crawl and re-calculate its position in the rankings.

    That is the way that things should be done. The other way round is just the Gestapo way of doing things. Reporting your neighbour to keep your neighbourhood clean!!

  79. I have trouble even contemplating the enormity of Matt Cutts’ job. I read about the demands of him to catch every single spammer and think we are expecting a bit too much. That being said, I know many of us would like to know how huge national corporation, who’s scope of violating Google’s quality guidelines is so great even some one with my limited knowledge can discover, can go unpunished. Findlaw was outed for selling link last year. According to one Findlaw rep, they went and showed Google what they were doing on the legal side of their business and that the violation were contain to their corporate side. This can’t be the whole story since hardcoded links from one of their properties, Legal Connection, were deactivated at approximately the same time. If you check any of their website inbound links you’ll see the number of links corresponding to how much money the client is paying. If they have 50 inbound links, 48 are from Findlaw representing them as practicing in areas they don not and practicing in areas of law they do not.You’ll also find invisible links from caselaw.findlaw.com. You’ll also find links from washlaw.edu. Although these would be free links, it is obvious the intention of these links are to game Google when these North Carolina Findlaw customers listed neither work or went to law school Kansas.
    Findlaw’s rep are making lawyers believe Google is their back pocket. It’s understandable conclusion when they deactivate hardcoded links on one site and load them to another while Google is busy increasing its page rank to 8 from the low of 5 when it incurred the penalty for selling link.
    I again say Matt’s job is much more challenging than I can imagine and expecting him to understand the ins and outs of each industry vertical is expecting too much. How long will Google let Findlaw violate their terms of use?

  80. This also isn’t as simple as removing such pages from the index… just because “SELECT * FROM COMMENTS WHERE review_item = ‘widget'” is zero now doesn’t mean it won’t have reviews in the near future.

    I’m not advocating to keep the pages in the results — with no reviews such a page really doesn’t provide value — but they could easily in the (instantaneously-near) future do so.

  81. But Dave you said “The only problem is that Google is ranking empty review pages above relevant on non-empty review pages.”

    There would be no rankings of empty review pages if the page didn’t exist in the first place if there are no reviews, right? SE’s read content and if you have an empty page that says or mentions Reviews or No Reviews that will be picked up. So in that case the page has good standings with Google’s algo (links, quality content etc) so then it’s ranked appropriately according to Google’s algo.

    You also mentioned “There is no sure way to make (or rely on) Webmasters to bear the burden for a flaw in Google’s algo. Even if there was, why take a band-aid approach and not fix the real problem?”

    Rely on a webmaster, they built the site in the first place, it’s much more of a flaw on the webmaster then the algo. The real problem is how the architecture of a site displays content.

    If it doesn’t exist don’t display it on the website, that simple.

  82. OnlyMe: as right as you are about this being a webmaster issue and a usability issue (there’s no point in displaying a blank dedicated review page to an end user), the flaw in your thinking is that it’s far too logical. If we forget about Google for a minute and look at the end user, then most sane human beings with a little bit of common sense would agree that displaying a page with effectively no content on it (and let’s face it, that’s what a dedicated review page with no reviews amounts to) isn’t good practice. There is no non-search-engine justification that could counter this.

    Buuuuuuuuuuuuuuut…there are hundreds of issues just like this issue that boil down to nothing more than webmaster ignorance (and in some cases, outright stupidity) Google has to deal with on a daily basis. Subpar code. Keyword stuffing. Publishing of referrer logs (which leads to sites being indexed before they’re ready on occasion). And about a million other pieces of idiocy that further serve to illustrate the devolution of society. Google as a result has to deal with things that really shouldn’t be of any concern to them…which is slightly unfair but it is what it is.

  83. Dave (originial)

    There would be no rankings of empty review pages if the page didn’t exist in the first place if there are no reviews, right?

    Correct, BUT they only exist because Google *chooses* to index and then rank them above relevant content. The problem is NOT the pages being included in Google’s index, the problem is Google is ranking blank review pages above relevant review pages.

    A blank review page is a Webmaster right to put online as it may just get a review in x time. It starts off a blank page for its OWN site USERS to give a review that Google’s algo ranks above relevant review pages.

    You can try a ban-aid approach (robots.tx or noindex) and HOPE all Webmasters know what to do and then actuallly do as much. I would hazard a guess that many would simply give a bogus review so their don’t need to play about with noindex etc. OR, Google can fix an abvious flaw in its algo and fix it 100% gauranteed.

    So in that case the page has good standings with Google’s algo (links, quality content etc) so then it’s ranked appropriately according to Google’s algo

    ” it’s ranked appropriately” You may say that, but common sense would surely dictate that blank pages ranking above relevant content is NOT appropriate. Google’s algo is always a work in progress, not a static finished product.

    Google has ALWAYS prefered to address Webspam (I dissagree that such pages are Webspam though) so why should this be any different? Think about it, IF the site owners intention is to spam Google, why on Earth would they slip in a robots.txt or noindex tag?

  84. Dave (originial)

    Google as a result has to deal with things that really shouldn’t be of any concern to them…which is slightly unfair but it is what it is.

    So long as Google wants to index and organize the Worlds information, be it online or offline, ALL of this should be, and is, of concern to Google and rightly so.

    This may come as a shock, but Google’s guidelines are NOT law (hence the name guidelines) they are there as an guide for Webmasters who WISH to be included in Google’s Index.

    IF a Webmaster goes outside those guidelines, Google has a CHOICE to remove them. Often they choose to NOT remove them as Google themselves want their content.

    IF the default of Google was NOT to include Web pages in their Index unless the site owner OPTS in, then the use noindex or robots.txt would make sense. But the reality is, Google chooses to index and list other peoples content in their index. Or, put bluntly, Google have a vested interest in indexing other people content *without their explicit consent*. It’s more often than not a win-win situation, but in case like Google ranking blank review pages above relevant review pages, the problem is obvoiusly Google ranking algo.

  85. Hi.

    As for “noresults” pages, I would suggest adding information about the reviews quantity in the search results, for example right under the title.

    In such case if someone’s looking for [product name review] and he sees “Reviews: 0”, he may want to continue searching but there’s also a chance that he may want to add his review.

    If you stop showing pages with no reviews, you could make it harder for webmasters to get new comments on their products. That would be the biggest problem for new websites that have to start from something and that have no reviews at the beginning.

    What do you think about it?

  86. If a web site has no reviews then it does not have the regular refreshed content that is required to get a decent SERP, and it will also not secure the other major factor in getting a good SERP – inbound links. So it will naturally be penalized into obscurity by the current Google algorithm.

    Thus restrictive measures may penalize sites that are at the early stages of driving traffic and trying to secure reviews. Sometimes you find that new site gets an unnaturally high boost at launch and perhaps this is why people come across these sites? As the post notes most of these sites are not set up to deceive people. Thus any restrictive measures leading to a spam label would seem heavy handed?

  87. Matt first I would like to thank for Spam report using Google account sign and hope this change will bring much faster review on Spam complaints?

    Usually when I notice some miserable black hat techniques implemented in sites I spend my time for Spam report. Hope this change in “Spam report” section will bring more people like me and will help Google to have most relevant results achieved through white hat techniques 🙂

    Best,
    Jag

  88. I think better communication via my webmaster tools would be much better for the novices who can only put trust in a SEO company to do a quality job, that we can track if they are doing something wrong. I have now found since replacing my previous SEO chap with a new one that I have 4 porn sites pointing to my horse riding school website which I can only presume is some sort of retaliation. It would be good if in webmaster tools, I could select links that are to be ignored.

    Chow from Sunny Cyprus.

  89. Dave (originial)

    Scott, links from other “bad neighborhoods” to your site will NOT harm your sites standing in anyway with the main SE’s. It’s the link FROM your site that can harm you.

  90. I would like to see some page to report malware infections on web pages. I often search for stuff that is in the current news. I searched today for “Arthur Nadel” and the second result on Google (mondaq.com) tried to install a fake virus scanner. Google didn’t warn me (with a warning page) like it is supposed to.

  91. Hi, I’m trying to make my CMS more intelligent and do not create or accept empty pages, also, I’m working to use the rel=”nofollow” to list pages, I mean, those just have list of pages with no real content, just a list of chapters or articles that you can click over any breadcrumbs… I hope it make the users target the real content pages at my web sites. Also, I’m working with the Robots.txt and with the URL Rewrite, to create more easy to find the real target pages… it’s not a very ease work, but, I think it will help people to get more interesting content.

  92. Dave (originial)

    nofollow is NOT for humans, only search bots.

  93. Matt,

    Here is one example and look the page it has nothing else than some ads from various network. The title says “MHS-CM1 Webbie HD Camera Review” but has nothing inside, not even an image.

    http://pclaptopreview.com/search/MHS-CM1+Webbie+HD%E2%84%A2+Camera+review

    Thanks

  94. Dave (originial)

    Why does Google insist on listing search results within its own search results???

  95. Am I missing something? All these people complaining that it’s not the fault of empty review sites, but how Google’s algos suck? I agree, and obviously so do many others that wrote their comments to Matt. Am I missing something? Wasn’t that the idea behind soliciting comments? To find out what sucks? Now that we figured out what sucks, let’s get rid of empty review sites, because they obviously are not review sites.

  96. Hi Matt
    qarchive.org is one site that comes up for a lot of searches.. But no useful content at all. I have already reported many times using the comment link that appears in the search results.

    Regs
    Jawahar

  97. Hi Matt,
    I think spam will always be around. The most important thing is to educate Google users how to use the search engine and how to recognize spam. Google help files could be much clearer on many topics. I miss examples in most help files. Google owns YouTube, but the Google channel is a bit of a mess with all topics not categorized too well. You could start a channel with “how to’s”.

    I also noticed there’s funny update / Google dance going on right now. I suddenly see results appear of one of my sites with URLs I redirected 2 years(!) ago. For a lot of (Dutch) queries the top 10 results are becoming “same, same but not different”. Looking for a map? Why not display 10 results that point to Google Maps clones (?)

  98. Matt,

    What do you want to realize this is certainly good, but it can also lead to consequences, as if at the initial stage will be to check people, then it will analyze and make the robot. But we all know that the work of the robot does not make perfect, and often by these “floating tests” are developing new and old sites, even if they are not as popular, but they have specific information and they can be blocked simply by accident.

    While I have repeatedly seeking feedback before seeking to buy and do not argue that some of the goods it is difficult simply because of the variety of the reasons (lack of description of the manufacturer, a brief description, users can leave comments, waiting for the description, etc.). I believe that in this there is nobody to blame, because no one is obliged to leave feedback when he bought merchandise through the site, but also prohibit the sites to make parts for the feedback, or the descriptions, too silly, as all who creates his shop hopes that after the purchase write a comment or a wish.

    Take for example a forum, a forum with many interesting questions and answers. One user left a question at the forum, while another who knows the answer, and accidentally got on the board is not so easy to answer because it is necessary to register and can expect the approval of the moderator. And all because a moderator protects the registration process to reduce spam.

    What do you think to do with such resources, as well as weak developing forums, blogs also fall under this criterion.

    Thanks

  99. Dave (originial)

    I agree, and obviously so do many others that wrote their comments to Matt. Am I missing something?

    No, the others are missing common sense. I feel sure Google will stop showing blank review pages via its algo.

  100. Any word on the ROR? I noticed some complaints about them and their extortion business. Why does Google support them and no other search engine??

    Thanks,
    Marko

  101. Sidenote: spam report is fairly difficult nowadays. How is the “normal” (not-SEO-inclined) net user to know what do “masked page” mean. Or which page does not match description in Google. Or … Not to mention it is in no way reachable from the result page.

    Those [x] buttons recently introduced seem like more reasonable attempt to let usual crowd mark sites they don’t like. Maybe Google could make some wiser use of it? If I click [x] whenever I see any result from swik.net, Google could really omit this site from my search results on any term, not only those where I already clicked this.

    This would solve the review sites problem too. It would be a site’s owner interest not to have poor landing pages

  102. Matt, how about http://www.dialaphone.co.uk/phone/Samsung_Armani_M7500/ it comes up second when searching for Samsung Armani M7500 review on http://www.google.co.uk. All it contains is a “reviews” tab that tells us “There are currently no reviews for the Samsung Armani M7500”.

  103. Hi Matt,

    You’ve had around 14 days to decide what to do so far, what are your thoughts?

  104. Talk about filling Google with empty garbage. They are really just useless pages if someone is trying to review something and there are actually “no results” to even look at. Good post!

  105. Today is Abe Lincoln’s 200th birthday too. Charles Darwin is important too, but Abe was one of our presidents, and we need to recognize his 200th birthday too.

  106. Latecomer here I think…

    Besides the non reviews or blatantly fake ones I cannot take the Local fake listings – this is terrible.

    I did apply for the “feet on the street” situation that was floated a year or so ago and felt that would be a solution to flag all the fake businesses addresses.

    David

  107. Matt, it seems the servive is unavailable:

    https://www.google.com/webmasters/tools/spamreport?hl=en
    It says:
    “This service is unavailable. Please check back later.”

  108. How about sites that just take reviews from other review sites?

    That annoys me. I’ve seen my own reviews on Yelp end up on other sites. Of course, there’s nothing I can do .

  109. I’d like to enforce the web spam team’s efforts on removing spammy search results, but I have an issue with sitemap building/submitting.

    My site has about 8 million indexed url’s in Google SERP and about the half of the sites are submitted via sitemaps.
    My question is how to manage such a large structure. Review sites many times become outdated the minut they are submitted to Google what produces all the empty pages. Rebuilding sitemaps of this size takes time, effort and kills my databases :-/ .

    How can I proceed to make everyone happy? Will Google enforce the effort like the webmasters do?

  110. Hi Matt,
    It’s probably too late to comment on this article, but I have just one little issue.
    If Google cares about user experience (which I don’t doubt) and wants to provide the best possible results, then, how come it completely ignores content of the page it serves to users in SERPS? How is it possible that the algorythm suddenly goes around what users actually want, read content?

  111. Hi Matt,

    I just saw your post and the answer seems to be write there.

    Couldn’t Google simply scan for “0 reviews” and use this to mark “review” as not being included in the page. Similarly “0 reviews” would cover your specific example. Even better, it would be broadly applicable to “0 s” or “0 s” or even “no s” to indicate that the particular is not a search term for the page. It would make it easy for honest web developers to have guidelines to follow as well.

  112. OK, that last comment got munged.
    It should contain a whole pile of <noun> and <adjective>
    View the source for info…

  113. Josh Beckhusen

    Just wanted to offer a suggestion for those that hate the useless comparison sites that always popup in the top results:
    givemebackmygoogle.com
    It helps filter out most of those sites by including a long -() string with the site names so you get generally cleaner search results. It’s not for everybody, since some people find those sites useful. It would be nice if google provided an option to include a filter like this as an option or make it easy to exclude a long list of them.

  114. Hi Matt,
    Going back to the question of reviews. I noticed today that some sites are getting review links in their search results http://www.google.com/search?hl=en&q=barnacles+temple+bar+house
    Just wondering how this works with the little stars? I couldn’t find any info anywhere on how to make the reviews crawler friendly?
    Thanks,
    Suzanne

  115. When I type Afghanistan war in google, it gives me an empty page (except for 1 line) and I have to scroll down a ways to get the regular results? And when I type in Lindsay Lohan for example, the results show at the top of the page like usual? Is this politically motivated? Or just a glitch?

    Any comments would be helpful.

    Regards,

  116. have been using google gmail for a while and it seemed to be working o.k. but for the last 2 weeks it has slowed down so slow that it is almost going backwards.i had to go to yahoo and that is working just fine..i would like to have gmail again but not the way it is working..i’m on dial up so i know it is slower then cable but it has been working rather good until for about 2 weeks..please is there any thing i can or you can do to help..i use chat so really would like to have google all the way..thanks..eugene

  117. Hi Matt,

    Suppose i have a web page with feeds that’s come through third party & sometimes these feeds are not shown on the page means web page is totally blank again & again, than how Google take this page?

  118. i want to get rid of google picnik,,, i don’t like it so don’t want it

css.php