When To NO INDEX (Panda Problems)

[LostOne]

Curious as to what the general consensus may be with pages that have a low time spent (average session duration). I realize it could be different depending on the site.

Trying to weed out some poor page performers. What's a good cut-off point? 30 seconds?

Any guesstimates, ideas?

[aakk9999]

Do they drive traffic to the site?

[LostOne]

Draw traffic? There are some pages that get a decent amount and they're in the 30 seconds to one minute category. They did much better years ago when there was little competition. The others that bring in less than 100 visitors each month (30 seconds and less) are an easy choice to no index.

[RedBar]

The others that bring in less than 100 visitors each month (30 seconds and less) are an easy choice to no index.

Why do you think that?

Do/does that/those page/pages provide their answer?

I have thousands of pages that provide the precise widget information people are seeking, I ain't removing them!

[glakes]

The others that bring in less than 100 visitors each month (30 seconds and less) are an easy choice to no index.

The fact that they are getting some traffic is a good thing. Instead of using noindex on those pages, maybe you should enhance the pages with more in-depth text, relevant images, video, etc. Of course this would depend on the type of site you have and available time, but fixing pages is often a lot easier then starting from scratch.

[seoskunk]

I'm starting to think we are beeing fed a bunch of crap regards Panda. Whilst we carefully atune our sites "brands" get away with absolutely murdering a web page. I am thinking that some of us are overthinking this and whilst "brands" churn out pages we are stuck fine tuning when we should just create.

[Zivush]

When To NO INDEX (Panda Problems)

When? never, unless you want to delete.
You may want to improve low performing pages if you can, but don't play with no-indexing, because a URL is a URL, indexed or not :-)

[[Some of my sites were Panda victims as well and I tried many ways to escape. Finally succeeded.]]

[Zivush]

whilst "brands" churn out pages we are stuck fine tuning when we should just create.

Exactly!

[LostOne]

Do/does that/those page/pages provide their answer?

My information has never really been about providing an answer but looking at it from a different perspective that has been learned from actual experience in the field (construction related) and not the all too common "research."

In 30 seconds no way anyone will get my message. But looking at "all pages" under site content in GA those numbers are vastly different and much higher. So what I'm assuming is those that landed on the page through search didn't get the message they wanted due to the time spent, but others did...by being more patient and actually spending some quality time looking/reading the info.

Yea, I agree with the brand stuff. It's really obvious.

I've reworked this original Panda stricken site for the fourth time (and also hired a Panda "expert" who also happens to own WW now) which is set to go live soon. Maybe they'll like it this time around being responsive and such. Heck it ain't no gorilla site, but only 200 pages now. I went from something like 900 to 500 to 250 and now 200 pages mostly through consolidation.

Thankfully I did get some relief in the May softer Panda release.

Thanks for the comments!

[simonlondon]

If you are talking about average visit duration on Google Analytics, be careful as I think (correct me if I am wrong) the average visit duration is arrived at by dividing total session times (excluding bounces) by total number of sessions (including bounces). If your pages have high bounce rate, this can really mislead you...

Edit: I shouldn't have said "excluding bounces", but that AVD counted 0 seconds for the bounces (where we know even the quickest bounce stayed on your site for longer than 0 seconds, not to mention those who stayed a few minutes and then bounced).

[RedBar]

but looking at it from a different perspective that has been learned from actual experience in the field (construction related) and not the all too common "research."

I'm in the same field along with many others and your "short visit" information seekers may just be checking/confirming what they already believed to be correct.

I have this all the time on my sites when people have told me that architects and specifiers have designated a product as being from X country, and they have a doubt, they immediately come to one my sites to check, they're literally a couple of seconds and gone since they know that information is correct.

You would be amazed at how many sites have the incorrect information for my products, especially so in the USA, plus in Europe we have to give correct technical terminology whereas in the USA it can be generic.

If you feel your pages are good and worthwhile, so will others.

[LostOne]

your "short visit" information

Makes sense and I was about to mention it. I'm the same way on the web. If I need a consensus opinion or fact. I may visit a half dozen sites before I'm satisfied. I'm also big on CTRL +F when I'm on those pages for a split second. For the most part I could care less what the site name is...except Demand media sites that I avoid like the plague.

So maybe I should be focusing on the "ALL PAGES" metrics or forget this freaking myth (bounce, click back button, time on page, user experience etc) that's been spread around for so long.

[tangor]

Page duration can be a bit misleading. As long as the page ranks (and is good info) there's no reason to noindex.

[GreenDog18]

Its too hard to tell if a page is useful for your visitors using the metrics that Google Provides us. Especially since we don't even know they keywords they are using any more.

You have to use your good judgement or solicit help of your peers. Read the article. If you articles solves a problem and is helpful leave it.

If your article doesn't solve a problem or provide any useful insight noindex it.

Metrics can get you pointed in the right direction but can also send you in the wrong direction.