Crawl reports urls with duplicate content but its not the case

[MakMour]

Hi guys!
Some hours ago I received my crawl report.

I noticed several records with urls with duplicate content so I went to open those urls one by one.
Not one of those urls were really with duplicate content but I have a concern because website is about product showcase and many articles are just images with href behind them. Many of those articles are using the same images so maybe thats why the seomoz crawler duplicate content flag is raised. I wonder if Google has problem with that too.

See for yourself how it looks like:

http://by.vg/NJ97y
http://by.vg/BQypE

Those two url's are flagged as duplicates...please mind the language(Greek) and try to focus on the urls and content.

ps: my example is simplified just for the purpose of my question.

<colgroup><col width="3436"></colgroup>
| URLs with Duplicate Page Content (up to 5) |

[SEOKeith]

Those pages are very similar so it's probably throwing the duplicate content switch in SEOmoz, you might want to ignore it in this case.

I would add some more text to those pages personally to aid with ranking, you can position the text over the images with CSS.

[MakMour]

Thanks Keith. I ve been using seomoz for some days so I wasnt sure about this.

Client wants website with as less text as possible so I guess my only hopes are title and alt attributes.

[SEOKeith]

Maybe explain to the client it's not going to rank as well without text and has less chance of getting found by searches (generally speaking...).

I get duplicate content flagging as well sometimes, I check the pages manually when it happens.

[MakMour]

Yeah, thats what I m going to do in my next meeting. Either way I also feel such websites need to have more pics than anything else, maybe a blog page or separate pages with articles could link to those products one by one with related description having a side content website for the actual product pages.

[Cyrus-Shepard]

Disclaimer: I just answered a question just like this on another thread, so I literally copied and pasted my response from there, and edited where necessary.

---

The SEOmoz web app uses a similarity threshold of 95% of the html code. This takes everything on the page, both hidden and visible into account.

In this case, it's counting all of the navigation and sidebar as well, which is significant. What's left of the unique content - the part that matters, makes up less than 5% of the code.Here's a tool you can use to check the similarity: http://www.duplicatecontent.net/

I ran the pages through a couple of tools which showed 98% similarity. (but only 75% text similarity, which is good, but not great)

SEOKeith is absolutely right that there's very little on those pages to help them rank. Without text, you're fighting an uphill battle.

Hope this helps! Best of luck with your SEO.