Meta descriptions better empty or with duplicate content?
-
I am working with a yahoo store. Somehow all of the meta description fields were filled in with random content from throughout the store.
For example, a black cabinet knob product page might have in its description field the specifications for a drawer slide. I don't know how this happened. We have had a programmer auto populate certain fields to get them ready for product feeds, etc. It's possible they screwed something up during that, this was a long time ago.
My question. Regardless of how it happened. Is it better for me to have them wipe these fields entirely clean? Or, is it better for me to have them populate the fields with a duplicate of our text from the body.
The site has about 6,500 pages so I have and will make custom descriptions for the more important pages after this process, but the workload to do them all is too much. So, nothing or duplicate content for the pages that likely won't receive personal attention?
-
Thanks, you were a big help. I'll do the A/B you are talking about.
I am thinking at this point I'll probably go with the body text. The site I'm talking about has well written text as the body of most pages. And, as I said, I'll be writing custom descriptions for the most important pages.
-
To be more specific, if you have good body text, Google/Bing can pull that into the SERPs if there is no meta description. That shortens your efforts. What I'm saying is, A/B test a page with Fetch or some other headless browser tool to see what the SERP is like without Meta description. I'm sure you've seen cruddy SERP results with Alt-text or code or unpronouncable characters: that's a coding issue. In many cases the result will be the H1 text, or the first sentence of the body.
As for what Luke said, yes, if bots aren't pulling good text into that space, a dynamic programmatically generated meta can work. It depends on goals. The downsides are that it can lose you a click if the searcher doesn't like what they see, as in, if the CTA or hook is ineffective. With body text they might give you the benefit of the doubt.
-
Thanks for the response.
I understand what you are saying. It sounds to me like you think (as Luke does below) that if duplicating the body text (which is good quality) will work then that's the best way to go?
What about Luke's suggestion of using dynamic text? Do you think dynamic text could be better than quality body text? I've never worked with any dynamic text. Are what are the downsides?
I'll investigate the questions you posed as well.
-
Thanks, we are thinking along the same lines here. The text from our body will 95% of the time be of good quality for a description, so it might work just fine.
I didn't think about creating dynamic text. Good idea. This might be the best middle ground for all the pages I don't plan to give personal attention.
Looks like I have a couple options to consider.
-
I think this depends a lot on what the text of the body looks like. If in general, the first couple of lines of the body is a good introduction that would inspire someone to click on the search result, then that would be a fine way to go. Otherwise you may want to trust Google. They do a pretty good job of selecting some relevant text for you.
If all of these are product pages, another option may be to dynamically create a generic yet enticing first sentence that the name of the product could be inserted in to and follow it with the first line from the body. So something like "Our <insert product="" name="">is the greatest thing since sliced bread. <insert custom="" text="" from="" the="" body="" to="" fill="" rest="">". So you would yield results like "Our door slide is the greatest thing since sliced bread...." and "Our black cabinet knob is the greatest thing since sliced bread....".</insert></insert>
Note my choice of initial phrase was more for comic relief. I would especially avoid that if the store also sells sliced bread
-
Whew, that is a tough one. IMHO, you are better off with a useful Meta description--one that is accurate to what the SITE is about--than none, IF there's a risk that bots will pull something other than useful text (like the social button or image alt text). Just think how the SERPs would look if only Title is visible, or a mess.
But, better with none, and let the bots pull in their own, than an inaccurate one (what you have now).
Have you talked to a dev about a dynamic and programmatic way to make unique meta descriptions for these 6500 pages? What kind of result do you get if you delete the meta description? Can you use a testing tool to fetch the site without meta description, just to see what searchers will see? If it's not bad and is more useful than a sitewide duplicate, just blank the majority out,
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
-
Exclude sorting options using nofollow to reduce duplicate content
I'm getting reports of duplicate content for pages that have different sorting options applied, e.g: /trips/dest/africa-and-middle-east/
On-Page Optimization | | benbrowning
/trips/dest/africa-and-middle-east/?sort=title&direction=asc&page=1
/trips/dest/africa-and-middle-east/?sort=title&direction=des&page=1 I have the added complication of having pagination combined with these sorting options. I also don't have the option of a view all page. I'm considering adding rel="nofollow" to the sorting controls so they are just taken out of the equation, then using rel="next" and rel="prev" to handle the pagination as per Google recommendations(using the default sorting options). Has anyone tried this approach, or have an opinion on whether it would work?0 -
Blog on server or embedded? Duplicate content?
Wondering what would be best in terms of SEO. Should I install some blog software actually on the website or can I just embed say a blogger.com blog? if I did that would they consider it duplicate content?
On-Page Optimization | | Superflys0 -
Duplicate Content
I'm currently working on a site that sells appliances. Currently, there are thousands of "issues" with this site, many of them dealing with duplicate content. Now, the product pages can be viewed in "List" or "Grid" format. As Lists, they have very little in the way of content. My understanding is that the duplicate content arises from different URLs going to the same site. For instance, the site might have a different URL when told to display 9 items than when told to display 15. This could then be solved by inserting rel = canonical. Is there a way to take a site and get a list of all possible duplicates? This would be much easier than slogging through every iteration of the options and copying down the URLs. Also, is there anything I might be missing in terms of why there is duplicate content? Thank you.
On-Page Optimization | | David_Moceri0 -
Duplicate meta data for 301 redirected items
Google webmaster tools tells me that I have duplicate title tags for two items. They have the same product ID. When there is a change in SEO URL, the system does an automatic 301 redirect of the first item towards the second item, like in the example below:
On-Page Optimization | | Madlena
22"x28" White Foam Board School Project Kit|Geographics.com
/p48998/foam-board-school-project-kit-22-x-28/product_info.html /p48998/white-foam-board-school-project-kit/product_info.html What are we doing wrong that Google asks for different meta data for a 301 redirect? Thanks!0 -
How to avoid duplicate page content
I have over 5.000 duplicate page content because my urls contains ?district=1&sort=&how=ASC¤cy=EUR. How can I fix this?
On-Page Optimization | | bruki0 -
Duplicate Page Content Question
This article was published on fastcompany.com on March 19th. http://www.fastcompany.com/magazine/164/designing-facebook It did not receive much traffic, so it was re-posted on Co.Design today (March 27th) where it has received significantly more traffic. http://www.fastcodesign.com/1669366/facebook-agrees-the-secret-to-its-future-success-is-design My question is if google will dock us for reprinting/reusing content on another site (even if it is a sister site within the same company). If they do frown on that, is there a proper way to attribute the content to the source material/site (fastcompany.com)?
On-Page Optimization | | DanAsadorian0 -
How much constitutes duplicate content in your opinion?
Mornin' In your experience, how much constitutes duplicate content? A sentence, a paragraph, half a page, etc? What about quotes - are they considered duplications, too, if there aren't quotation marks? Over the years, the client has been a bit bad in taking a paragraph from here, a sentence from there, and coupling it all together as daily news on their site. I'm now in the middle of a purge. Oh boy! All hail originality.
On-Page Optimization | | Martin_S0 -
Would it be bad to change the canonical URL to the most recent page that has duplicate content, or should we just 301 redirect to the new page?
Is it bad to change the canonical URL in the tag, meaning does it lose it's stats? If we add a new page that may have duplicate content, but we want that page to be indexed over the older pages, should we just change the canonical page or redirect from the original canonical page? Thanks so much! -Amy
On-Page Optimization | | MeghanPrudencio0