Pandalized Site to go from 15k pages to 200 Pages: What Will Happen?

[aok88]

I have a site that was nailed by the very first Panda and has never recovered. I have tried everything under the sun to come back even a little, and it has been to no avail. In fact it has only gotten worse and worse. It is now at 100 visits a day and pre Panda it was 4k a day!

So, I am about to give up on this site (I never thought I would say that!).

But I wanted to try a test as a sort of last resort and at the very least I might learn something from this it as well.

The site currently has about 800 unique domains linking to it and about 42k total inbound links, according to MajesticSEO.

Most of the site's inbound links however point to only around 200 pages, which are the best pages and most have unique and useful written content.

Here is what I wanted to try and I would love to hear predictions:

I plan to delete all the pages of the site that don't have at least 1 inbound link to them. So that would leave the site with roughly 200 pages, and every one of these remaining pages would have at least 1 unique domain pointing to it, and a lot of them would have lots of unique domains pointing to them (essentially these are the best/top/most popular pages of the site anyway).

What would happen? Would it do anything?

I ask this because I wonder exactly how PageRank would flow. Is it that currently, with 15k pages, the total PageRank is getting divided up among all 15k pages, so that the 200 pages I'm referring to only each have a small fraction of that PageRank, and that if I delete all the rest, those 200 would have a much larger fraction of the PageRank, thus making them rank much higher than they do now (or at least that is my theory anyway).

So, what are your predictions? How will the remaining 200 pages rank? Better, worse or the same? And why?

[Whitey]

I have tried everything under the sun to come back even a little

Can you outline this to begin with - it will help folks understand how you have applied your priorities to come to the current conclusions and request for advice.

[turbocharged]

If you are removing many "thin" pages, you could see a positive improvement. But if you are removing pages to stop pagerank from being diluted, I don't think you will find it to be successful. Pagerank is trumped by Google's aggressive penalties, unless you are part of the few mega-large branded/whitelisted sites.

[Planet13]

I think conventional wisdom regarding a Panda penalty is to remove poor quality pages - no matter what your link profile is like.

[Masaai]

Turbocharged makes a very valid point. The pages must perform well in addition to having a useful backlink profile (amongst other variables) to perform well.

My site was affected by Panda 1.0 and we have lost approx 85% of our Google traffic. I have thought about doing exactly the same thing. I think that if you have engaging content measured loosely by a healthy time on site and a low bounce rate for those pages then you may succeed. Perhaps you should improve your content on those pages and provide links to relevant internal content where appropriate as well.

[MasterOfPuppets]

i think it wouldn't help at all
probably getting new domain a lot better idea

[Robert Charlton]

The site currently has about 800 unique domains linking to it and about 42k total inbound links, according to MajesticSEO.

Most of the site's inbound links however point to only around 200 pages, which are the best pages and most have unique and useful written content.

This doesn't sound like a very natural pattern to me. It sounds highly coordinated and easily spotted. Chances are that Google has identified all of the linking domains as willing-to-spam.

If the site had been Pandalized, you're now apparently trying to hold onto bits of that linking network that helped low quality pages rank in the first place. It sounds like a disavow-by-attrition process, where you're looking at what you can hold onto from link sources that Google probably doesn't like.

What really needs to be addressed is the quality of the content, along with the quality of the linking domains... and the nature of the relationship you had with them. If, once upon a time, you had roughly 200 freely-given good links to 200 good pages and then you got carried away by, say, "ambition", cutting the site back to the original 200 might make some kind of sense... assuming that all of the original linkers had remained clean.

I don't think this sounds anything like that kind of situation.

probably getting new domain a lot better idea

Repeating the same kind of process on a new domain, IMO, is not likely to be a long term fix. I think you've got to reconsider your business model. In the long run, you will need to have genuinely good content, with a uniquely identifiable value proposition, that's genuinely worth linking to.

[JD_Toims]

I think that if you have engaging content measured loosely by a healthy time on site and a low bounce rate for those pages then you may succeed.

I think you're on-the-right-track, but I would warn [you and all readers] bounce-rate and time-on-site are *not* things you should try to "manipulate" for ranking purposes, because manipulation of numbers really doesn't help in rankings.

Those numbers are things you should use as signals/clues to understand visitor behavior as related to query and information-provided on a given page *internally*, so you can try to better match visitors to the results served by search engines and provide visitors with the answer they are searching for on the first page they visit.

---

As far as bounce-rate goes, I've posted previously I've had [and may still have] pages on sites with a +90% bounce-rate [and that's bad, or so I keep reading] rank in the top 3 [for years] for the queries they were designed to rank for, but I don't mind since I can look at the time-on-site + query made and compare those to bounce-rate and know the bounce-rate is high because, visitors found a "match" to their query on the *first* page they visited.

One thing, contrary to popular belief, a high bounce-rate [shown by most stat programs as one page-view/visit regardless of time-on-site and close rate] can tell you is: You did your job by getting the content right and the search engines did their job by ranking the correct page for a visitor to find what they were searching for with a single click on a result -- I've had more than one page like this [based on query made, on-page content, time spent on the landing page, sales increase] and might still have some* on a site or two.

* I haven't checked on the sites I'm referring to recently, because it's not that important to me as long as the revenue from the sites keeps increasing even though I haven't touched them other than to correct "glitches" that have been pointed out in years.

---

BTW: Welcome to WebmasterWorld!

[aok88]

These are all great replies, thanks. However, I am at a loss as to what to do...

[Masaai]

Again, inbound links alone as a measure of whether to keep a page or not is only one part of the full picture.

To escape panda you need to improve your content and make your site more engaging/useful.

Possible action items may include:
-checking in analytics to see which pages get traffic from Google, and perform well based on the type of the site you have. Keep the quality pages. You may also want to keep pages which don't get traffic but are quality pages (in terms of content.
- 301 all pages you plan on deleting (IF the page in question has any link equity. If in doubt 301 it anyway) to a useful related page that you are keeping.
- Further improving the pages you are keeping (and site as a whole) based on SEO guidelines. A resource you can consider is an article is called 'a visual guide to keyword targeting on-page optimization' by R Fishkin.

[aok88]

What really needs to be addressed is the quality of the content

I think conventional wisdom regarding a Panda penalty is to remove poor quality pages

Perhaps you should improve your content on those pages

Are we looking at the same serps? While I appreciate all the suggestions, it seems like I am looking at dramatically different serps (I am on the East Coast of the U.S.) and am using Google.

When I look at sites that are kicking butt right now, in a few industries (including mine), original and high-quality content are no where to be found. The two 'brands' in my industry alone that are cleaning up right now have NO content on their pages. I mean, an h1 and a big picture and that is it!

I am not so sure that Panda is all about high quality content, at least not in the results I look at.

[Masaai]

I assume you don't have a brand?
Perhaps having a 'brand' is good enough to rank (for a variety of brand-related factors, eg trust which translates algorithmically somehow)

I can speculate for days but it is evident that they are doing something right, whatever it is, and that you need to improve something.

You acknowledge your quality isnt good, so you should start there, (if indeed you it was panda that you were affected by). My understanding is that you will need to do something to change your content from bad to good in order to rank. You got hit by Panda and they didn't so the burden is on you to do something.

This type of things happens all the time. Trust me, the same thing happens in my space.

[heisje]

but it is evident that they are doing something right

Wouldn't bet on that. Rather, it is evident that Google's algorithm is skewed in favor of the "signs" brand sites emit, both on site and off site. You do not posses the firepower to re-create such signs. So, forget it, and don't waste your time.

Therefore, churn & burn - and don't get sentimental with old sites : live and let die. As simple as that.

.

[Masaai]

Are you talking about big-brands specifically?
In all spaces there are websites that thrive, for a variety of big-brand related signals.
In all spaces there are websites that thrive for a variety of reasons non-big-brand related.