How to Heal Your Site after Google’s Panda Update

Did you get mauled by Google’s Panda? Are you sure? And if you did, do you know how to nurse your Web site back to health on the world’s foremost search engine?

To recap, Google is in an almost constant state of revision of its search algorithm, the mathematical formulas it uses to decide which pages are the most authoritative for a given keyword and therefore should get displayed first in a search on that term. But the system saw two big spikes in changes last January and again in early March. The first, as Google explained after the fact, was an attempt to cut back drastically on the search prominence of “content farms”: sites that produce reams of low-quality content, drive traffic to those sites through organic search, and then earn revenue by selling pay-per-click ads on those pages.

Good riddance, you may think.

But that effort then broadened into a much larger revision of the Google algorithm in February, nicknamed “Panda” (apparently after one of the update’s engineers), that had the express intent of downgrading the page ranking of many pages within non-farming sites.

On its official blog, Google announced that the aim was to cut the rankings—and thus the position within search results—of “sites which are low-value-add for users, copy content from other websites or sites that are just not very useful.”

It’s been estimated that the first revision, the one targeting content farmers, had an effect on about 2% of all Google queries. Not huge as a proportion, but since Google logs upwards of 2 billion searches a day, pretty substantial in real numbers: 40 million queries a day affected by the Farmer update.

The second, more extensive Panda update is ballparked to have affected some additional 15% of search queries. Together with the first change, that’s now about 340 million Google searches a day that produce search results pages with links different than those they turned up before the update process began. And tweaks to Panda are still rolling out.

That’s queries. But how many actual sites have reported a drop in their rankings for their main keywords? The Web is so vast that it’s hard to know, but a poll conducted by Barry Schwartz on the Search Engine Roundtable site about five days after Panda found about 40% of Webmasters reporting that they were seeing less Google traffic than pre-Panda.

Pretty certainly the large majority of sites downgraded in these Google moves are informational or publishing sites on the lines of AssociatedContent.com (down 93% in U.S. Google traffic right after the update, according to calculations from search research firm Sistrix) and Ezinearticles.com (90% dropoff).

But ecommerce, B-to-B and general product or service sites can’t afford to be complacent, for the simple reason that many of those also contain some of the telltale signs Google’s looking for of poor quality content. An article in the Wall Street Journal pointed to ecommerce sites such as OnTimeSupplies.com, an office-supplies Web retailer that experienced a 50% decline in Web traffic.

So what can a Web merchant who has felt the Panda’s claws do to recover? There’s plenty of advice floating around—some of it contradictory.

First, to evaluate whether they are in fact being hurt by the algorithm change, marketers should go to their analytics toolkit and check the number of key search terms that are referring visitors to their site from Google. “That’s an important metric for search engine optimization,” says Chip Rice, National accounts manager with digital marketing firm OneUpWeb. “It goes beyond the vanity of search position [on a Google results page] and gets to the core of how big a presence you have in the natural search space. If you’re seeing a smaller variety of search terms driving traffic, then there’s a good chance that you’re experiencing a negative impact from Panda.”

But one evident key is for marketers to check their Web site’s content and evaluate it honestly for originality, value, and usefulness to visitors.

“A lot of ecommerce companies face a real challenge in producing good quality unique content,” says Rice. “Their product descriptions often come straight from manufacturers and are duplicated across many, many Web sites—perhaps hundreds or thousands with the exact same copy. So as this update goes through and looks to reduce the amount of scraped, duplicated or poor-quality content and to reward unique value-added content, there’s definitely a potential for ecommerce companies to suffer an impact.”

The answer might be to rewrite that content substantially to make it unique. The situation may be complicated by the fact that in some cases, manufacturers require resellers to post their product content exactly as written. If enough resellers feel the pinch of Panda in their rankings because of content that Google sees as duplicative, those agreements may have to change. In the mean time, etailers may need to consider adding elements to those product pages that counterbalance the shared content with what Google will detect as valuable original data—such as video, consumer reviews or ratings.

“Reviews and ratings can be a wonderful way to add unique, relevant and fresh content that’s being updated continually,” says Chris Keating, director of SEO with Performics. “That way you don’t run into the problem of having only scraped content that never changes. Ideally you want to show Google a page that’s alive. That sends the message that this page is relevant.”

Web marketers should also be aware that bad pages can drag down the good ones around them in Google’s eyes. If you’ve got a page that has very little on it except a short product description and photo, you’re asking Google’s bots to see that as shallow content of little worth, and it can affect the overall ranking of your site.