Panda's Take on Message Boards and Forums

[Pjman]

I was just wondering as to whether anyone formed an opinion on how Panda quality assessment views forums (message boards)?

Obviously this forum is the exception to your standard 3-4 posts per thread in a forum. But I can name many forums where each thread maybe totals 300 words, but the information is great.

Has anyone been hit due to the quality of their forum?

What are your thoughts on this?

[JohnRoy]

On the other side, one of the sites in the Sistrix list of the biggest losers was daniweb.com - a forum through and through. I've heard that traffic is back for them since Panda 2.1.

According to compete.com (http://siteanalytics.compete.com/daniweb.com/) daniweb traffic didn't get back up (as of 4/20). They lost in July 10 much more than in February 11 though.

How did Webmasterworld fare ?

According to the above - no change. but dropped 85% since last year. Ouch!

[tedster]

Then Compete is wrong - which is not the first time. Their data is only dependable when the site has done the dedicated Compete tagging in their source code.

[JohnRoy]

Immediately after being pandalized

Add to dictionary ;)

[JohnRoy]

@tedster,
wrong - agree.
But to which degree?
I assume their traffic range is still within limits. [although their "referring sites lists" is not worth looking at]

- Oh,
and do you refer to daniweb or to WebmasterWorld?

[tedster]

Neither of us have tagged our sites for Compete. Their data is incorrect in both cases. Do not assume anything, not even about "ranges". Errors can be in the area of a full degree of magnitude for sites that are not Compete tagged, unless they are full enterprise behemoths like Microsoft.

[shri]

>> On the other hand, if your forum is absolute dripping with awesome and the community is alive and kicking then it deserves to rank well.

What are some of the factors these forums have.

1) Regular return visits ( stickyness ).
2) People loading multiple pages ( bounce rates )
3) People spending time on the site ( dwell time )

If you run more than forums on your site and parts are hit ( ours are ) look at those pages and I suspect, even though you're providing the best information possible, your pages might have been degraded a little bit.

May be I'm not sophisticated enough like most people here to look into things like ads above folds and speelings and what nots... to me in my little world, the above factors seem to point out a weakness.

>> one of the sites in the Sistrix list of the biggest losers was daniweb.com

Was majority of their traffic coming from search engines? People found the answers and bounced right back after 30 seconds? I suspect it might have been the case.

[JohnRoy]

Was majority of their traffic coming from search engines? People found the answers and bounced right back after 30 seconds? I suspect it might have been the case.

Not sure this is correct.

When I search how to make chocolate milk and I get the answer through ehow, I don't bounce back to Google. I get busy with the cocoa ;)

[rogerd]

One thing ehow has done is create strong "related" links. So, when the first article about how to make chocolate milk is useless, you'll be drawn to a related article "The Easiest Way to Make Chocolate Milk," and from that one to "How to Make Extra Tasty Chocolate Milk." By the time you are totally frustrated and bail out of the site, you haven't bounced, you've scored multiple pageviews and been "engaged" for more than a minute. All signals of great quality to the mechanical observer.

(Actually, chocolate milk is something ehow can likely handle. Home improvements and technical topics require actual expertise.)

[Simsi]

Just thinking out loud here: Are UGC sites scraped more frequently than non-UGC sites?

Many forums (vBulletin)have RSS feeds of posts which makes scraping far easier so you may have a point. There are plenty of forum aggregators out there that just hook into the standard feeds and duplicate content.

It's also probably not uncommon for a forum owner to think they are being scraped when in actual fact they are unwittingly providing a feed that facilitates it.

I only have experience with vB (and even that is dim and distant) but from memory it's easily turned off if it's a concern.

[madmatt69]

Yeah, Panda killed off lots of traffic to my forum, which is a shame because many of the popular threads could easily be considered authoritative on the subjects they cover.

And there's a lot of UGC, but why should I no-index older threads to bring down the overall site page count - as has been a strategy many have advocated to get de-pandalized? It's not good for the users, so if Google practices what they preach about rewarding sites that engage their users..I mean isn't that what a forum is all about? Long before facebook and twitter those were and still are the avenues for sharing information about specific subjects.

[webdesignkc]

As said before the emphasis seems to be on quality.

[webconnoisseur]

I keep a close eye on UGC sites (of all types) & their traffic, especially pre and post Panda and can confidently say that UGC-alone has little indication of pandalization. A few sites on Google's hitlist (like eHow & TripAdvisor) did get hit by Panda 2 on their UK sites. Either it was manual, or the UK versions of their sites didn't have the massive backlink shield the US versions did.

Daniweb is best measured on Quantcast where they are quantified (have tags on their pages to provide accurate numbers). Their traffic has not bounced back and is still maintaining traffic levels 50% below pre-Panda.

Considering Panda appears to be a quality judgement algorithm, potentially using site design or other visual elements, I'd be curious to hear if there are certain bulletin board or discussion forum platforms that were affected more then others. It would make sense that certain platforms are designed better then others and that a greater percentage of sites on one platform may be affected.

[Uber_SEO]

I didn't think Tripadvisor got hit? Sistrix data mentions Tripadvisor.in, but not their UK or .com sites. Would be interested if they had been effected, as I'm considering them to be an outlier case at the moment

[webconnoisseur]

@Uber_SEO Are you paying for Sistrix data? The reason I ask is because I've found the free data they share to be highlights and not all encompassing for those affected by Panda. Some sites lost 30% of traffic, not 70%. And a million phrases is a lot, but still a very small sample set for sites that receive long-tail traffic (which is many UGC sites).

If I recall correctly, About.com never showed up on the systrix lists, but their losses shared during the Q1 NY Times: "About also experienced a moderately negative impact on page views from the algorithm changes Google implemented in the quarter."

[Uber_SEO]

I'm not paying, although I contacted Sistrix and asked them for the complete list, so it's more than the stuff they blogged, but I imagine less than the stuff you pay for

[ascensions]

FYI, I thought I'd pass this along- especially for PHPBB owners. I discovered someone out there is trying to do something (not sure what though by it) by appending a directory in the name of the viewtopic.php page to the page:

ie:
/viewtopic.php/viewtopic.php

See for yourselves in Google and do a site: command on your site.

[google.fr...]

This would definitely suggest "thin" content if not properly canonicalized.