Google using descriptions from other websites instead of site's own meta description
-
In the last month or so, Google has started displaying a description under links to my home page in its search results that doesn't actually come from my site.
I have a meta description tag in place and for a very limited set of keywords, that description is displayed, but for the majority of results, it's displaying a description that appears on Alexa.com and a handful of other sites that seem to have copied Alexa's listing, e.g. similarsites.com.
The problem is, the description from these other sites isn't particularly descriptive and mentions a service that we no longer provide.
So my questions are:
- Why is Google doing this? Surely that's broken behaviour.
- How do I fix it?
-
I (finally) see the confusion - a good reason for me to be careful in word choice. I didn't say "duplicate content" I said "duplicated" content. What I meant was "repetition" not duplicated but I guess because we see "duplicate content" every day as SEOs I chose the wrong phrase. What I meant was the duplication / repetition that can happen in the title, as in my example:
"Brisbane SEOs and digital marketing services in Brisbane | SEO | Marketing"
I have many times seen replaced title/description if keywords are repeated in the titles. I have always cleared it up with noodp and noydir. In this case I stated that I didn't think that was the real issue but it is one that causes problems.
So the examples I copied in didn't have to do with "duplicate content" as it relates to rel=canonical but it has to do with "duplicated" title keywords. Obviously I wasn't clear enough in the original post and I'm glad to know that. I still think my advice will work and for the reasons I stated, just with better phrasing. I definitely didn't mean to be confusing so thanks for pointing it out.
Hope that clears up the misunderstanding and thanks for helping me give better advice - appreciate it.
~Matt
-
brisbane web development may get more searches but I also don't rank nearly as well for it as I do for terms with freelance in it.
Most of the enquiries I get follow on from searches that contain freelance and brisbane in the query, whereas brisbane web development is the only one of about 30 keyword phrases that I've been tracking that is showing the correct description.
As far as changing over time: it's only in the last month that these incorrect descriptions have shown up; everything's been fine up until now.
-
If you look at the conversation between Matt and I, you will see that your meta you do not want is showing in dmoz and a few directory sites. Since the query, freelance web design brisbane is a low volume query, and brisbane web development is getting 2400 searches per month, I would not worry too much about it.
Every search I did that I was able to find you had corrected meta. The one you don't like was last used on your site in mid 2011 it appears. I think over time it will change, but putting too much into it is not worth the time.
All the best.
-
Matt,
This info from google doesn't have anything to do with duplicate content.
The first one is about title tags and even that says they (Google) may try to improve the title. Nothing about the meta.
The second is from a Google forum in 2008 and says to first check dmoz to see if the meta is appearing there. If you just say, hey use noodp, noydir, you are making an assumption that is problematic.
The third where you have John Mueller, it is again about Titles and not using keyword stuffing.
Here is the issue Matt: When you state something like that (and I have made the same mistake) and leave a lot out, someone who doesn't do this day to day, assumes something that is simply not true. Frankly, I know of no instance where duplicate content has caused a SERP snippet to change.
Yes, you can use noodp, noydir, but you need to explain why and not say its because of duplicate content. The snippet he gives says "Provides a range of web design and print design services." If you put that search on Google.com.au, there is no duplicate content issue.
Yes, that does appear on dmoz, not on Yahoo Directory. But, it also appears on several directory type sites. Will using noodp keep it from happening? Only if that is the source.
So, I have thumbed you up for the courtesy of a reply to me (evens out the thumb down). Thanks for the reply and, feel free to let me know if I stray or if you believe something here is incorrect. I am open to being wrong and having it pointed out.
All the best,
Robert
-
Robert, if you do a search for freelance web design brisbane (result is on the first page for me in google.com.au), you'll see the sort of thing I'm referring to. This is what's coming up for most of the keywords I'm tracking for my site.
If you do a search for brisbane web development (result on page 9 although a few days ago it was page 15), you'll see the snippet saying what my meta description tag for the home page says, i.e. what I want it to say.
-
http://support.google.com/webmasters/bin/answer.py?hl=en&answer=35624
"Avoid keyword stuffing. It's sometimes helpful to have a few descriptive terms in the title, but there’s no reason to have the same words or phrases appear multiple times. A title like
"Foobar, foo bar, foobars, foo bars"
doesn't help the user, and this kind of keyword stuffing can make your results look spammy to Google and to users."If we’ve detected that a particular result has one of the above issues with its title, we may try to generate an improved title from anchors, on-page text, or other sources.
Google's suggestion is basically what I said above:
If you're concerned about content in your title or snippet, you may want to double-check that this content doesn't appear on your site. If it doesn't, try searching Google.com for the title or snippet enclosed in quotation marks. This will display pages on the web that refer to your site using this text. If you contact these webmasters to request that they change their information about your site, any changes to their sites will be recognized by our crawler after we next crawl their pages.
In addition, John Mueller gave this advice in a post on one of Google's blogs:
"In general, when we run across titles that appear to be sub-optimal, we may choose to rewrite them in the search results. This could happen when the titles are particularly short, shared across large parts of your site or appear to be mostly a collection of keywords. One thing you can do to help prevent this is to make sure that your titles and descriptions are relevant, unique and compelling, without being "stuffed" with too much boilerplate text across your site."
---------------------------------
(Pretty much sounds like what I said but you thumbed me down for.)
-
Matt,
Where are you getting: **Usually it's duplicated content that gets your meta replaced **
I cannot find any reference to it anywhere in GWMT, etc.
Thanks,
Robert
-
I think I understand, but want to be sure. The first img attached is my listing in SERPs here in US with my homepage drumbeatmarketing.net/ Below the SERP link for the query are sitelinks.
The next is your Home Page meta per SEOmoz tool
The next is your About Page meta per SEOmoz tool
The last is your SERP page from Google.com.au showing ABOUT page as the first page to show. Note query was Tyssen design australia
Note that the SERP snippet and the meta you have are the same for that ABOUT page. This would mean that Google is showing precisely what you are asking for it to show.
If this was recently changed, it may not yet have been reindexed hence the need to resubmit sitemap or do a fetch as google on that page as I previously gave you.
I thought this might be a site link issue originally and should have done a bit more investigating and asked you more. If what you are seeing and what I am seeing is the same, then the issue is that you are assuming your homepage is what is first in SERP and it is About page. Short of that, I would need to know what meta you have for homepage, what query gives wrong result, etc.
Hope this helps you out.
Robert
-
Hi Robert, no they're definitely not site links, can verify that they're search results snippets.
-
John,
I would be curious to see if this changes anything for you. I have sites that are listed with the Open Directory Project (aka dmoz) and with Yahoo (we paid for the listings in Yahoo for our client). I do not see Google grabbing those descriptions any longer for use in the SERP snippet.
One thing I would suggest if you believe the noodp, noydir (both good links) change will make a difference is to resubmit your sitemap and/or run a couple of Fetch as Googles on some of your site url's like your home page where you are seeing this. I doubt anything will change.
Also, it sounds as if what you are talking about are site links as opposed to the search snippet that draws from the meta description. With site links, you can turn those off (demote them) for ninety days if incorrect. Go into WMT and on your site you will see: Configuration. Click and you will see sitelinks. Note that if you are doing this for your home page you add nothing to the url that you see first.
Here is info straight from GWMT:
Demote a sitelink URL:
- On the Webmaster Tools Home page, click the site you want.
- Under Site configuration, click Sitelinks.
- In the For this search result box, complete the URL for which you don't want a specific sitelink URL to appear. (How to find the right URL.)
- In the Demote this sitelink URL box, complete the URL of the sitelink you want to demote.
I think you will find this much more effective.
Best
-
Thanks, I've just done both of those.
-
You should include a noodp, noydir tag to try to prevent this. I saw your title & description and they look fine. Usually it's duplicated content that gets your meta replaced (say Brisbane SEOs and digital marketing services in Brisbane | SEO | Marketing" That would get you replaced in a heartbeat.
For yours, I don't see that - but they must think the Alexa is more relevant.
-
Change Alexa and ping those changes to Google.
-
Add noydir and noodp to your meta tags.
Hope that helps!
-
Got a burning SEO question?
Subscribe to Moz Pro to gain full access to Q&A, answer questions, and ask your own.
Browse Questions
Explore more categories
-
Moz Tools
Chat with the community about the Moz tools.
-
SEO Tactics
Discuss the SEO process with fellow marketers
-
Community
Discuss industry events, jobs, and news!
-
Digital Marketing
Chat about tactics outside of SEO
-
Research & Trends
Dive into research and trends in the search industry.
-
Support
Connect on product support and feature requests.
Related Questions
-
Duplicated titles and meta descriptions
Hi, Dealing with both my duplicated titles and meta descriptions i'm wondering if there's a "quick" win I could potentially implement asap. A bit of background:
Technical SEO | | GhillC
Say I've 4 pages structured that way: domain.com/us/productA.html for the US domain.com/gb/productA.html the UK domain.com/fr/productA.html for France domain.com/de/productA.html For Germany At the moment, both my page titles and meta-descriptions are duplicated all over the place for product A.
Title is reading "Product A - company name"
MD is a bit better, being translated in all 3 languages (En, Fr, DE). Therefore being the same for the US and for the UK. Ideally, I would get unique page titles and MD all over the place. However, due to time and resource constraints, I can't make it happen overnight. So my questions are pretty simple:
1. Can I create a rule for page titles to be "Product A - country - company name" or similar? Would that be enough to make the page titles unique? Is there any value doing so?
2. Can I "localize" duplicate MD by simply naming the country? I assume it is not enough in this case as all the rest would be copy/pasted. Ideally speaking, both my page titles and MD would be completely unique but I can't afford doing so in the short term. Thanks!0 -
Can you force Google to use meta description?
Is it possible to force Google to use only the Meta description put in place for a page and not gather additional text from the page?
Technical SEO | | A_Q0 -
IT's Hurt My Rank?HELP!!!
hi,guys,john here, i just began use the MOZ service several days ago, recently i noticed one thing that one keyword on the first google search result page, but when i done some external links,the rank down from 1 to 8, i think may be the bad quality external links caused the rank down. so my question,should i delete the bad quality links or build more better quality links? which is better for me. easy to delete the bad links and hard to build high quality links. so what's your better opinion,guys? thanks John
Technical SEO | | smokstore0 -
Links under Meta Description when performing a search
Doing research for clients, I have came across seeing sites displaying hyperlinks underneath their own meta description. keywords that I have googled that result with hyperlinks displaying under meta descriptions: Google'd: iacquire (brand) bmw wheels (Beyern Wheels, position 1) aftermarket bmw wheels (MMR Wheels, position 2) These companys have hyperlinks underneath their descriptions. Anyone have any ideas why this happens or how it happens?
Technical SEO | | frnprz0 -
No confirmation page on Google's Disavow links tool?
I've been going through and doing some spring cleaning on some spammy links to my site. I used Google's Disavow links tool, but after I submit my text file, nothing happens. Should I be getting some sort of confirmation page? After I upload my file, I don't get any notifications telling me Google has received my file or anything like that. It just takes me back to this page: http://cl.ly/image/0S320q46321R/Image 2013-04-26 at 11.15.25 AM.png Am I doing something wrong or is this what everyone else is seeing too?
Technical SEO | | shawn810 -
Can JavaScrip affect Google's index/ranking?
We have changed our website template about a month ago and since then we experienced a huge drop in rankings, especially with our home page. We kept the same url structure on entire website, pretty much the same content and the same on-page seo. We kind of knew we will have a rank drop but not that huge. We used to rank with the homepage on the top of the second page, and now we lost about 20-25 positions. What we changed is that we made a new homepage structure, more user-friendly and with much more organized information, we also have a slider presenting our main services. 80% of our content on the homepage is included inside the slideshow and 3 tabs, but all these elements are JavaScript. The content is unique and is seo optimized but when I am disabling the JavaScript, it becomes completely unavailable. Could this be the reason for the huge rank drop? I used the Webmaster Tolls' Fetch as Googlebot tool and it looks like Google reads perfectly what's inside the JavaScrip slideshow so I did not worried until now when I found this on SEOMoz: "Try to avoid ... using javascript ... since the search engines will ... not indexed them ... " One more weird thing is that although we have no duplicate content and the entire website has been cached, for a few pages (including the homepage), the picture snipet is from the old website. All main urls are the same, we removed some old ones that we don't need anymore, so we kept all the inbound links. The 301 redirects are properly set. But still, we have a huge rank drop. Also, (not sure if this important or not), the robots.txt file is disallowing some folders like: images, modules, templates... (Joomla components). We still have some html errors and warnings but way less than we had with the old website. Any advice would be much appreciated, thank you!
Technical SEO | | echo10 -
Duplicate Meta Description in GWMT
We've just discovered that there are multiple duplicate URLs indexed for a site that we're working on. It seems that when new versions of the site was developed in the last couple of years, there were new page names and URL structures that were used. All of these seem to be showing up as Duplicate Meta Descriptions in Google's WMT, which is not surprising as they are basically the same page with the same content that are just sitting on different page names/URLs. This is an example of the situation, where URL 5 is the current version. Note: all the others are still live and resolve, although they are not linked to from the current site. URL 1: www.example.com/blue-tshirts.html (Version 1 - January 2010) URL 2: www.example.com/blue-t-shirts.html (Version 2 - July 2010) URL 3: www.example.com/blue_t_shirts.html (Version 3 - November 2010) URL 4: www.example.com/buy/blue_tshirts.html (Version 4 - January 2011) URL 5: www.example.com/buy/bluetshirts.html (Version 5 - April 2011) Presumably, this is a clear case of duplicate content. QUESTION: In order to solve it, shall we 301 all of the previous URLs to the current one - ie. Redirect URLs 1-4 to URL 5? Or, should some of them be NoIndexed? To complicate matters, there is Pagination on most of them. For example: URL 1: www.example.com/blue-tshirts.html (Version 1 - January 2010) URL 1a: www.example.com/page-1/blue-tshirts.html URL 1b: www.example.com/page-2/blue-tshirts.html URL 1c: www.example.com/page-3/blue-tshirts.html URL 4: www.example.com/buy/blue_tshirts.html URL 4a: www.example.com/buy/page-1/blue_tshirts.html URL 4b: www.example.com/buy/page-2/blue_tshirts.html URL 4c: www.example.com/buy/page-3/blue_tshirts.html URL 5: www.example.com/buy/bluetshirts.html URL 5a: www.example.com/buy/page-1/bluetshirts.html URL 5b: www.example.com/buy/page-2/bluetshirts.html URL 5c: www.example.com/buy/page-3/bluetshirts.html Since URL 5 is the current site, we are going to 'NoIndex, Follow' URLs 5a, 5b and 5c, which is what we understand to be the correct thing to do for paginated pages. QUESTION: What shall we do with URLs 1a, 1b and 1c? Should we apply the same "No Index, Follow" OR should they be 301'd to their respective counterparts in 5a, 5b and 5c? QUESTION: In the same way, since URL 4 is the version just before the current live Version 5, does it make a different on whether the paginated pages (ie 4a, 4b and 4c) should be No Indexed or 301'd? Thanks in advance for all responses and suggestions, it's greatly appreciated.
Technical SEO | | orangechew0 -
Does 'framing' a website create duplicate content?
Something I have not come across before, but hope others here are able offer advice based on experience: A client has independently created a series of mini-sites, aimed at targeting specific locations. The tactic has worked very well and they have achieved a large amount of well targeted traffic as a result. Each mini-site is different but then in the nav, if you want to view prices or go to the booking page, that then links to what at first appears to be their main site. However, you then notice that the URL is actually situated on the mini-site. What they have done is 'framed' the main site so that it appears exactly the same even when navigating through this exact replica site. Checking the code, there is almost nothing there - in fact there is actually no content at all. Below the head, there is a piece of code: <frameset rows="*" framespacing=0 frameborder=0> <frame src="[http://www.example.com](view-source:http://www.yellowskips.com/)" frameborder=0 marginwidth=0 marginheight=0> <noframes>Your browser does not support frames. Click [here](http://www.example.com) to view.noframes> frameset> Given that main site content does not appear to show in the source code, do we have an issue with duplicate content? This issue is that these 'referrals' are showing in Analytics, despite the fact that the code does not appear in the source, which is slightly confusing for me. They have done this without consultation and I'm very concerned that this could potentially be creating duplicate content of their ENTIRE main site on dozens of mini-sites. I should also add that there are no links to the mini-sites from the main site, so if you guys advise that this is creating duplicate content, I would not be worried about creating a link-wheel if I advise them to link directly to the main site rather than the framed pages. Thanks!
Technical SEO | | RiceMedia0