noindexing pages to try and escape penalty

[proboscis]

I'm going to try noindexing duplicate and thin pages to try and escape a penalty and Panda. I have both now.

Anyway should I just noindex or nofollow too?

[Robert Charlton]

You definitely don't want to use the nofollow attribute along with noindex, as nofollow would stop natural PageRank circulation on your site.

I'm not a big fan of noindexing. I think of it as a bandaid approach, but it's not a long term fix for thin content (more appropriately for Panda, "shallow" content). You've got to do the hard work of creating new content.

For certain kinds of duplicate pages, if you can't fix the duplication problem structurally or on the server, the rel="canonical" tag is a better approach than noindexing.

[proboscis]

Thank you Robert! I'm glad I asked.

I've got about 150-200 pages that need new content and it takes me a day or more usually to do one, so I'm just going to noindex them, then either fix (with new content), merge, or delete.

[balibones]

Proboscis, I and several other folks I know have tried this and it didn't work for us. We didn't come out of Panda until we deleted the pages and either 301'd them to better content or let them 404. It may work out different for you, but I just wanted to share my experience.

[proboscis]

Thanks balibones.

So if I try reinventing the pages first and it doesn't work, then I should try moving them to a new page and 301?

[balibones]

Proboscis, if you "reinvent" the pages, as in make them higher quality, I don't think there will be a need to 301 them or to delete them. What I was saying is that redirecting or removing a bad page seems to work better than simply noindexing it with a robots meta tag. Good luck!

[Robert Charlton]

Keep in mind too that it may take a while for the reinvented page to come back. Ditto for the site.

It's possible that some Panda factors at this point are hard wired into a continuously updating algo, but it's likely that most, if not all of Panda is still being recalibrated and adjusted periodically... so I would not make any judgements based on short term results.

Also, in my experience looking at Pandalized sites, I feel that most of those who built them are in denial about what the problems are and about how much work it's going to take to fix them.

I don't think that it's a fruitful approach, though, to think you can turn an stock automobile into a race car by removing the entire engine. You can strip down some parts for weight, but at a certain point you really need to beef up the essentials.

[proboscis]

Also, in my experience looking at Pandalized sites, I feel that most of those who built them are in denial about what the problems are and about how much work it's going to take to fix them.

Yes that was me. I'm just starting to get it after being hit by the first round of Panda almost a year ago.

I really appreciate these extra tips, I'm getting more clear on what to do and how to do it.

Strip it down and beef up the essentials.

[Planet13]

There were a few things that stuck out at me:

I've got about 150-200 pages that need new content and it takes me a day or more usually to do one, so I'm just going to noindex them, then either fix (with new content), merge, or delete.

Which led me to re-read this statement by Robert Charlton:

Also, in my experience looking at Pandalized sites, I feel that most of those who built them are in denial about what the problems are and about how much work it's going to take to fix them.

I have to ask whether the type of pages that can be created in a day are really going to rank well? This is not a personal insult, but just a question about what the requirements are for content to rank well in the post-Panda world.

I know that a day's worth of work might be quite a bit of effort compared to the average ehow article, but it still seems like for a quality page, it is minimal effort.

I remember when I worked as a journalist, you were expected to take about three or four days to produce a quality story. Usually, the actual process was longer, because you would have to spend time contacting authorities on the subject, and of course you had to wait for them to get back to you.

Anyway, no offense.

[jkdt0077]

Planet13, surely the amount of time required to create a useful, content filled page is dependant upon several things; the person writing it, the subject matter, the number of different pages about that topic you are going to write etc..

What may take you three or four days to write about may take somebody else, perhaps more knowledgeable about the subject matter at hand, only one day. That isn't an insult either by the way!

Your question is a good one though, what are the requirements for content to rank well post-Panda?

[proboscis]

I'm back. You were all pretty much right. I've seen some movement but no increase in traffic.

A search for my domain name appeared on page 6 and over the last few months has slowly moved up to page 2 or 3.

Some search terms have moved from around page 400 to 200 but I don't know if they're just moving around back there or actually increasing. A few have fallen further but most appear to be increasing according to WMT. Not much of a success story.

And it is taking me way longer than a few days to improve each page. I think because I'm trying so much harder now, making sure not to cut any corners and trying so hard to be interesting.

So what would you do at this point?

1. Keep the noindexed pages and continue trying to slowly improve them before letting them back in the index.

2. Just 404 them and loose all the incoming links some of which are good and almost 14 years old plus have a silly half-done looking site with only a small amount of pages.

[Planet13]

2. Just 404 them and loose all the incoming links some of which are good and almost 14 years old plus have a silly half-done looking site with only a small amount of pages.

is it possible to serve up a 404 status BUT STILL SHOW THE ORIGINAL CONTENT without cloaking?

that way, any people who land there will still be able to see content, while google will deindex them (for now).

[Robert Charlton]

is it possible to serve up a 404 status BUT STILL SHOW THE ORIGINAL CONTENT without cloaking?

I understand you can do that with a 410, but I've not done it. See this discussion... [webmasterworld.com...]

My initial reaction is that, in this situation, rather than go to that extreme and kill the pages and the inbound links, you should keep the meta noindex,follow, but only on the pages eventually worth keeping. More thoughts on that below.

Strip it down and beef up the essentials

That made sense, but I realize that "strip it down" may be ambiguous. Some pages may simply need to be dumped completely, others noindexed but worked on in a prioritized fashion. I'm not sure if that's what you've done. It may be that you've tried to hang onto too many pages. I don't know. There are likely to be tradeoffs.

I think you need to look clearly at the user experience and at Google, and trim your site down to a point where it's useful. Check backlinks, try to preserve the core pages and pages that have backlinks which are worth preserving, and work on those first.

While normally I don't like to fiddle with pages once they've been launched, I'm wondering whether, in this case, it's possible to improve the pages incrementally, in stages, because it seems the otherwise you're going to have a site sitting hidden in the basement for much too long. Is there a core site online that's in the index? How much of the site online is noindexed?

You should also assess the quality of the initial pages that did get backlinks. If, in your opinion, those pages really aren't very good, it may be that those backlinks aren't worth much either. Were they freely given editorial links? Are the sites that gave them still good sites? It may be that you need to further evaluate what's worth keeping, and maybe get rid of some of those pages too, perhaps redirecting the links to related and reworked pages, but only where appropriate.

[proboscis]

Okay, I'll take a carefully planned middle road.

At this point a portion of the reworked pages along with new pages are in the index. About 2/3 remain noindexed.

The quality of the initial pages is good, but I think the type of content is bad (in Google's opinion). It a subject guide for research, so basically just lists of links. They're links to authoritative resources on specific subjects that you can't find by just doing a search but still that's not good enough anymore. Right?

I have another site that has thrived with Panda and continues to do well while I do nothing. So I'm switching to that type of content which is (hopefully) thoughtful well written information. Maybe it's important that the user finds what they're looking for on the page rather than a collection of links to what they're looking for.

My good links should still be good but it's an old site that was popular back in the day, so lots of scraping and lots of low quality links too.

Am I on the right track?