Thin page threshold with Panda

[austtr]

Had coffee this morning with a colleague and we got to talking about how encyclopedic, informative type sites often end up with a natural mix of "fat" and "thin" pages. eg.. a destination guide might have 300 pages, with the capital cities and tourist hot-spot pages being full of good "fat" stuff but regions off the beaten trail are "thin" by comparison because there just isn't much to say. That's just the way it is.

There is nothing wrong with these pages and they might still be the best information source available... but they are "thin" when compared to the rest of the site. (Fewer visitors, less time on page etc)

Is it possible that sites with a pronounced variation in "fat" vs "thin" pages might be on Panda's radar? From a human perspective, we can look at pages and say they are not "thin" because of what we see..... whereas an algo might see a ratio of "fat" vs "thin" that triggers a bad site slap.

Lots of folks are/have been complaining about sites of lesser quality now starting to dominate their serps..... and I have to agree. The sites I'm seeing are mostly bland, under researched, lacking in substance, with a templated format and certainly not the best resource available.

Could it be that sites are escaping Panda, not by the quality they deliver, but by having a "fat" vs "thin" ratio that keeps them below some magic threshold?.... and conversely, real quality sites are being hit because there is too big a difference between their key pages and the rest?

[deadsea]

I think it depends on how often the pages are promoted internally. If the "off the beaten path" thin pages are hard to navigate to, then it shouldn't be a problem. If you click around the site and get to both thin and fat pages regularly, then it could hurt.

I work with a site with about 30,000 pages. Of those, there are probably only about 1000 fat pages. They do a great job of promoting the fat pages to visitors, but allowing visitors to find the thin pages if they really want to.

[Sgt_Kickaxe]

I think that you need to step back and look at the bigger picture. There is no one magical solution but there are a lot of things you can do to improve results and improving quality is definitely one of them. How much thin is too much? It's probably relevant to the other pages on that primary topic from other sources, how does yours stack up?

In SEO when you overdo any one thing something else begins bulging out the other side as a result, everything in moderation!

[tedster]

Especially in light of Matt Cutts recent comments about local pizza pages only needing 2-3 sentences to rank, I don't think that many of these issues are a fruitful way to understand Panda and shallow content.

I don't think word count defines shallow content. If there's not a lot of information available on a topic, that's just the way it is. As long as those words are useful and not just a re-hash of any other page about the topic, it should be fine. And if a topic is not a widely researched topic, then not many visitors would be expected.

I agree that there seem to be a lot of pages ranking these days that are lower in quality, and I don't have any real answer for what's going on.

[claaarky]

What I'm finding is that 'quality' is whatever your users decide it is. Your entire site could be pages with 2 sentences on (for example, a great question followed by the perfect answer) and if that information was so incredibly insightful and useful that users found every page vital, you'll have no problems with panda.

I don't think it's about how nice your site looks, how much text there is, whether you use images or video, it's purely about whether your site delivers something people like, and more so than your online competition.

[MarvinH]

2 sentences on (for example, a great question followed by the perfect answer)

Claaarky, I cannot agree more. I have one such page on my forum. The page consists of a very short question consisting of fewer than 30 words, and an even shorter 8 word answer with a single link to a downloadable .zip file. Certainly one of the shortest threads on my forum. It's a typical page many people in SEO communities would consider "thin content" deserving noindexing, excluding in robots.txt, or even deleting. Yet, this page, for the past well over a year, has been receiving a lot of clicks from Google search results, a lot more than product reviews consisting of thousands of words, numerous images, and hundreds of replies (no ads, no aff links, just pure reviews, it's a non-commercial site, a hobby site).

[SnowMan68]

Especially in light of Matt Cutts recent comments about local pizza pages only needing 2-3 sentences to rank,

What Matt said there people should be paying more attention to. I think it is a huge statement. Run with it!

[aakk9999]

Especially in light of Matt Cutts recent comments about local pizza pages only needing 2-3 sentences to rank,

What Matt said there people should be paying more attention to. I think it is a huge statement. Run with it!

We have to be careful how we interpret what Matt Cutts said.

Matt Cutts did not say that pizza page needs only 2-3 sentence to rank. What he said is (emphasis mine):

In addition to address and contact information, 2 or 3 sentences about what is unique to that location and they should be fine.

This does not mean they will rank (since rank can be misunderstood as = rank well). I believe this merely means the page will not be seen as thin/shallow page. Ranking well is whole lot of different and would depend on other factors too.

[tedster]

Agreed - what we're talking about is avoiding evaluation as shallow content and the associated Panda problems that brings.

[diberry]

Well, WikiAnswers keeps ranking well, despite a really small word count, so I'm doubting word count alone will get you in trouble with Panda.

[austtr]

I don't think word count defines shallow content.

I don't think so either, but obviously there has to be some cause and effect(s) which arouses Panda, and everything stated and reported to date indicates that thin content is the primary trigger.

Given that websites cover a huge spectrum of topics, have very different formats and styles, different content delivery platforms etc etc... then obviously a thin page in one niche might be very different to a thin page in another niche. A determination of "thin" cannot be made about a page in isolation... surely any such assessment has to involve a comparison with the surrounding pages.

The OP takes this hypothesis and asks if its possible to trigger Panda if too many pages fall below some arbitrary threshold when compared with the remainder of the site pages

[SnowMan68]

This does not mean they will rank (since rank can be misunderstood as = rank well). I believe this merely means the page will not be seen as thin/shallow page. Ranking well is whole lot of different and would depend on other factors too.

I never said anything about rank well...this thread is about "thin page threshold with Panda" and how to escape or avoid Panda. The insight that Matt gave is a very important one. What he mentioned about 2-3 sentences of unique content I think can be applied to more than just pizza shops :)

[rango]

What Matt said also ignores other factors on the page. Let's say that a page with 2 sentences also has 2 ads on it. In fact, the 2 sentences could drop below the fold. In that case, I don't think it's unreasonable to expect it to be considered thin again. And I don't think his statement rules that out. If however it's an ad-free page with 3 unique sentences, then it does make sense for that to considered enough.

I think the fat/thin ratio is a good theory and much in line with my own thinking. I'd imagine it's not the *only* factor though - I believe there's also a duplicate/non-duplicate ratio at play, though I think that existed prior to Panda.

[aakk9999]

Maybe the problem is not in number of pages with thin content (few sentences) versus other beefed up pages. Instead, maybe the problem is in what subject these thin pages are about and how many other sites write about the exact same subject these thin pages are covering.

Going back to pizza restaurants and comparing this to the imaginary site that lists cities in a country. Lets say that for each city it has only 2-3 sentences and you have many of these pages. I think a site that list cities would have greater chance to be hit by Panda as all this content is already covered many times over and over throughout the web.

I would also say that this is perhaps very different between ecom site and informational site, and further, that ecom site is more likely to be hit if the same products it is selling are sold widely by other sites.

Again, take a comparison between pizza restaurants site with one page per restaurant that has a couple of sentences and lets say, compare it with a site that is selling televisions, where each of a particular model of TV is widely sold on many other websites, and where page describing TV has also only 1-2 sentences. I think that in this case the TV site has bigger chance to be hit by Panda - especially as other sites will repeat pretty much THE SAME information, whereas there would be fewer sites on the web talking about this same pizza restaurant.

Think of this in the light of the following sentence from the same interview:

Matt Cutts:

It is still not a good idea to repeat a ton of content over and over again. (...) The value add disappears.

So perhaps it is not how much you say - it is maybe whether it was said before and how many times and on how many other pages of your site and/or other sites.

With regards to the ratio to the other "beefed up" pages of the site - I am sure it is not so simple, it is also likely to be connected with the authority the site itself has on the subject it is covering.

[claaarky]

This discussion perfectly illustrates the difficulty in defining thin, shallow, low quality content. It's not one thing and it doesn't standstill. The threshold is whatever tips your site from being generally well liked to being a bit annoying.

Over the past 3 weeks I've been looking at my site in a different way, looking at how people react to my site and how google reacts in corresponding fashion. Google likes the pages people like. What people like today may not be the same tomorrow.

I'm thinking of my site like a real world thing such as a hotel, restaurant, nightclub, tv show, smartphone, etc.. These can be popular one minute because they offer something new or better, then be gone because others arrived on the scene doing the same thing better. Their offering (content) was high quality until others came along and with a better offering.

How we all decide which hotel or restaurant to visit or which tv show to watch changes all the time. We can even get bored if the things we like don't innovate and move forward with the times. Even google knows they need to keep innovating to keep competitors at bay.

I think google is now has a democratic ranking system based on what most people like and that is constantly changing. Panda is all about weeding out the website equivalents of old hotels trading on past success, tv shows that people are now bored with and so on.