If you feel your site was wrongfully hit by Google’s Panda update, there might be hope for you yet. We recently looked at a couple sites who have seen some minor recovery since being hit hard by the update, and since then, we’ve spoken with Dani Horowitz, who runs the IT discussion forum DaniWeb (one of those sites) about what she’s been doing to get back into Google’s good graces.

Should forum content rank well in Google search results when relevant? Comment here.

DaniWeb’s US traffic went from about 90,000 visitors per day down to about 40,000 per day after the update, she tells WebProNews. This sent her into “complete panic mode”.

“I just went into crazy programmer SEO mode, just removing duplicate content and things like that,” she says. She thinks duplicate content may have been a big factor, but duplicate content and its relationship to backlinks, specifically.

“We syndicate our RSS feeds, and there are a lot of websites out there that syndicate our content, duplicate our feeds legitimately…they just take our RSS feeds and they syndicate that,” she explains,noting that many of these sites were linking back to DaniWeb.

“My hypothesis right now is that Google Panda figured out all these sites are really content farms – are really just syndicators, and we just lost half our backlinks,” she says. “So I think it might not necessarily be that Google is penalizing us for being a content farm, but that Google is penalizing all the content farms that are syndicating our content, effectively diminishing the value of half of our backlinks.”

What DaniWeb Has Done to Aid Recovery

First off, she says she entirely redid the site’s URL structure. The actual URL of every single page has changed, Horowitz says.

She removed tag clouds, which were at the bottom of every single page, saying that Google frowns upon these because they can look like keyword stuffing. “What I went and did was made my tag clouds actually populate via javascript in such a way that it actually improves page load time for the end user because they’re no cache, except Google can’t actually spider the actual tag cloud pages, because I added them to the robots.txt file.”

It’s been established that Google takes page speed into consideration as a ranking factor, so certainly this could only help (though it does make you question Google’s whole philosophy of “creating pages for users and not for search engines”). In fact, Horowitz recently showed the correlation of pages Google was indexing with the improvements in page load time:

Horowitz says she added a robots.txt to all search results pages, because Google also frowns upon actually having search-like pages in its index. Google wants to be the search engine itself, and point to the content – not to other search results.

She made heavy use of nofollow and noindex tags. “Basically what I did was I took hundreds of thousands of pages out of Google’s index from our domain, but hopefully the advantage being beneficial to the end users…”

Specifically, she noindexed forum posts with with no replies, hoping that Google will recrawl, and start indexing them after they do get replies. She notes that this is simply an experiment.

Finally, she made the Facebook and retweet buttons more prominent. Clearly, Google is moving more and more toward social as an indication of relevancy, so this can’t hurt either.

Horowitz notes that it is entirely possible that the uptick in post-Panda traffic might also be related to other updates Google has implemented since the Panda update. They make changes on a daily basis, and it could simply be that DaniWeb was positively impacted by a different tweak.

Forums and Their Value to Search Results

With the Panda update being all about the quality of search results and the content they deliver, we asked Dani about her thoughts on the value that forums have in this department.

“Forums are in my opinion the best way to get content online, and to get the answers to questions that people want online, where you have not just a single publisher or an editor and team of staff writers, but actually [are] able to poll the entire Internet and [are] able to get expertise from anyone who has it,” she says. “I definitely think that forums are growing. They’re not going to end anytime soon,” she adds, noting that they may change in format.

“It is a double-edged sword, because you have all this great content that’s contributed by the people who know the content best – know the answers best – as opposed to being limited by a team of staff writers, but the flip side is you have people who are not talking in 100% U.S. English, and you have people that don’t have correct grammar, and you have spelling mistakes,” she continues. “So now, we’re leaving it up to Google’s algorithm to try to figure out which…if someone is querying Google…which page has the correct answer. Is it the page that is written by some staff writer that doesn’t necessarily have a complete interest in the topic, but does have a three-paragraph/five-paragraph article that’s written in full-sentence English or is it written by someone who’s a complete expert in the topic, and knows everything…but maybe isn’t a native English speaker and is writing in broken english with lots of spelling and grammar mistakes. It’s hard to have an algorithm try to figure out which is the better result to show.”

Google did include the question, “Does this article have spelling, stylistic, or factual errors?” in its recently released list of “questions that one could use to assess the ‘quality’ of a page or an article”.

Better Google Results?

When asked if she thinks Google’s results are better now, she told us that the rankings for DaniWeb content have gotten a bit weird. She says that they were ranking for “round robin algorithm” (a computer science term that would make sense in terms of DaniWeb’s content) before the Panda update, but not after the update. Meanwhile, DaniWeb started ranking for the odd keyword “rectangle” after the update (though this was no longer the case after she posted about it in the Google Webmaster forum).

“Before Panda, we were ranking number one for some really great articles that were very relevant,” she says. “Post Panda all of our number one rankings for all of these great articles went down, but we started ranking for some really weird stuff.”

She also noted that her experience searching with Bing “sucked”.

Recovery?

To be clear, it’s not as if DaniWeb has experienced a full recovery since the uptick in traffic began. “We’re still nowhere near where we were before,” she says. “We’re still down nearly 50% but literally we just stopped the bleeding, and there’s [been] a very small improvement week after week the past three or four weeks, but if nothing else, it’s not going down anymore…”

She’s still looking at other things that can be done, and concentrating on building backlinks – trying to create great linkbait.

Do you think DaniWeb should have lost Google rankings? Tell us what you think.