Why is OSE showing no data for this URL?

[danatanseo]

Hi all,

Does anyone have any ideas as to why OSE might not have any data for this URL:

http://www.ccisolutions.com/StoreFront/product/shure-slx24-sm58-wireless-microphone-system-j3

It is not a new page at all. It's been on the site for years.

Is OSE being quirky? Or is there an underlying problem with this page?

Thanks in advance for any light you can shed on this,

Dana

[AndieF]

Hi Dana,

I think your page is corrupted, I have copied a link to the sourcecode I am seeing http://pastebin.com/BRfFT4RR

It looks like Google Cache is also having problems with this page. Perhaps OSE had trouble too and so skipped the page?

• Andrew

[danatanseo]

Thanks so much for responding Andrew. I have suspected problems with our code for a long time, but I am not a coder, sp it's been a challenge to attempt to identify the specific problem.

I believe this is not just a problem with this page, but could be a problem across many pages on our site.

Can you are any of my fellow Mozzers point to what you are seeing in the source code that leads you to believe it is corrupted?

Many thanks for any help. I truly appreciate it!

Dana

[danatanseo]

I also see what you mean that there is a problem with Google's cache. The cache date is really old (April 11) and there is no preview of the page.

Anyone who can point me in the right direction?

[danatanseo]

Could the coding issue be related to the drop down box that's located just above the pricing on the right hand side? That is one thing that makes this product page different from others on our site.

Thoughts?

[danatanseo]

I did some comparisons with other pages and it doesn't seem that the drop-down frequency selector is the culprit. This page also has one: www.ccisolutions.com/StoreFront/product/shure-slx24-sm58-wireless-microphone-system-h5

but the cache in Google seems to be fine for this page and OSE displays data for it just fine.

[danatanseo]

I just ran the source code for this page through the validator at: http://validator.w3.org/

There are a multitude of problems that need to be addressed. Thanks very much Andrew. I do have enough HTML knowledge to provide guidance to our IT manager on how to fix the problems. I don't have access to much of the source code, so it will certainly be a "project" to fix the issues.

I am sure these problems are everywhere all over the site, as many people with very little experience in coding and design have had their hands in the pot (so to speak) over the years.

At least this will allow me to prove to our CEO that our underlying code is indeed presenting a problem for indexing and crawling.

[ThompsonPaul]

Dana, there's no content on that page.

The massive head section with all it's JavaScript is there, making it look like there's lots of code, but the actual body content has somehow been deleted.

This is all I see in the actual body of the page:

|

<form name="headerForm" action="IAFDispatcher" onsubmit="return submitQuery()" method="post">

That's it. There's no actual content, no footer, no closing or tag, which makes me think someone's actually deleted the content part of the code by accident.

Good luck figuring out who borked it

Paul

</form>

|

[danatanseo]

Thanks so much Paul,

Yes, when I ran a "Fetch as Googlebot" it returned a "Success" message, but when I looked at what Google is seeing there is no content on the page.

"borked" - great term...I am definitely going to have to file that one away for future use!

If the problem is isolated to this page, that's one thing. I am more concerned that this problem is effecting a larger number of pages.

Once I figure it out, I'll come back here and post what we found/fixed.

I really appreciate the comments from you and Andrew very much!

[ThompsonPaul]

Glad I could help, Dana.

And yes, "borked" is a technical term. It's defined as existing in a badly broken state as a result of an inexperienced/inattentive user making unauthorised/incorrect changes to a website's code or content

Can also be used as a verb: "he borked the database so badly the whole site went 503".

Not that it's ever been applied to me or anything.

And yea - sometimes our tools can mislead us, even though the info they provided was "technically" correct.

Suggestion for a fast way to test the rest of the site for this kind of error: Use the paid version of Screaming Frog to program a search for a snippet of code that should be in the content area of every product page. Limit the crawl to the product pages category. (Or whatever sections of the site you're worried about.)

You could search for something as simple as class="productExtendedDescription" which would at least ensure the content container was there. Still wouldn't prove there was any content it it, but if you wanted to get fancy with regex, you could even do that too. You could also search for the tag, which would indicate that the rest of the pages' code likely exists.

Just an idea to speed up the testing process.

Paul

[danatanseo]

Hi Paul,

We discovered that the problem was being caused by a trailing "comma" at the end of the keyword string that we once used  to populate the Meta keywords tag. Unfortunately, the keyword information in those fields is still being parsed. The parser did not know what to do when it encountered a comma followed by nothing.

We did run a query and found that this problem was affecting 128 of our product pages and had been for a long time. We haven't been populating the keywords for almost a year now, so the problem is at least that old.

The commas are now gone.

Thanks again to you and Andrew!