Google, Content Farms, and the Changing Face of UGC

In February, Google updated its algorithm in a significant way. The update impacted nearly 12% of all queries and, in Google’s own words, was “designed to reduce rankings for low-quality sites [and] provide better rankings for high-quality sites”. But it also might represent a fundamental shift in Google’s philosophy about search and user-experience.

Search, Relevance, and UGC

Nearly all of Google’s revenue comes from Adwords, but Google is so much more than just a search engine. The company’s mission statement, after all, is “to organize the world’s information and make it universally accessible and useful”. Search might provide a platform (and revenue model) for indexing everything that’s online, but what Google really wants to become is the go-to cloud-based app for the entirety of human knowledge — i.e. the how-to and what-is guide for everything.

But Google’s algorithm update also happened to napalm certain content farms and competitors. Indeed, when you look at the top 25 losers, you notice that Google seems to have devalued user-generated-content (UGC).

This represents a significant shift in Google’s approach to content because, traditionally, Google has been partial to UGC. The logic behind their UGC bias, moreover, seemed sound because UGC doesn’t have the same vested interests as so-called “commercial” sites.

Whereas commercial sites produce content (and link to other sites) for the purpose of creating revenue, UGC is more about actual people/users sharing their opinions and experience through content because they find it useful. We are, after all, social creatures, and since Google wants to give the most valuable results to real human beings, UGC seemed like a promising resource to help them do that.

But then you look at the top 25 sites to lose out from Google’s latest algorithm update, and you notice just how many of them are UGC-driven. From shopping sites powered by user-reviews to article submission sites, many of these losers had built their business model on UGC to the extent that they leveraged it to rank well in the SERPs and attract new users/customers.

Source: Search Metrics

Of course, these sites had found a loophole in the Google algorithm and were exploiting it. In other words, their revenue models were (on one way or another) piggybacking off of Google’s own technology and user-base. But depending on just how altruisitc you are about content and search, some of these sites deserved to penalized, and some didn’t.

On the less extreme end of the spectrum, there were the sites that were leveraging Google’s preference for UGC and how-to content to generate traffic that they could then monetize. One example is which used user-reviews to rank better on product searches, and then make a commission on the users they then referred to various online merchants.

On the more extreme end of the spectrum, sites were actually incentivizing users to generate content content. These sites were more problematic because UGC loses its value when its incentivized. Basically, if there’s something in it for the user to produce the content, the motive behind the content becomes less altruistic and it becomes that much less likely that the content is actually share-worthy.

Search and the Changing Face of UGC

So what does this all mean for the future of UGC and search? Well, it depends on which search engine we’re talking about.

Most UGC these days is being produced on social networks, not on content sites. It’s coming in the form of tweets and wall-posts and likes. This is even more valuable as “altruistic” UGC because (1) it’s being created at a much higher volume, and (2) it’s more closely tied to “real” user identities and less likely to be spam.

The only problem is that most of it is happening on Facebook, and Google can’t get at it there That is, its crawlers can’t get in behind the Facebook registration wall, so they can’t see what’s happening there and factor it into their index.

Bing, on the other hand, has a deal with Facebook, and is actively factoring data such as Facebook Likes into its index. Specifically, it’s able to personalize your search results based on (1) how much something has been liked on Facebook, and (2) how much it’s been liked by people in your network.

This, of course, gives Bing a significant edge over Google in offering a “social” kind of search. But the real point is that as search evolves, the kinds of UGC that are going to matter is going to be the stuff that social graphs are made up of. It’s going to be about the content (1) real users have produced, and (2) how what each user has produced can be used to personalize their own search experience.

CT Moore

CT Moore is a Senior Strategist and Account Manager for a SEO and web design company based in Montreal, Canada.


  1. Ken Durham

    The issues I have most often run across for people with unique content, that did get hit by Panda, are two things:

    1: The presence of too many affiliate ads. I know most people will say to nofollow them and they will be fine, but I don’t believe this matters. I believe what matters is that the affiliate ads are relevant to your niche, and not packed in so tightly that it is hard to distinguish ads from content. The ads should be for products that you truly recommend. How many people that you know would recommend more than one or two lawnmowers to buy? Yet one site that was hit was recommending a dozen or so, on the front page. Advertisement like this is a sure way to get you Panda’ed.

    2: Purchasing SEO plans that include mass backlinking. I have seen this time and again. People will argue that if this is so, then all you need do is purchasing such backlinking packages for your competition. I’m not recommending this but I say this would work, to an extent. But no one ever tries it in fear of helping the competition. In fact, I had a high ranking site that we tested these packages with. And always, in any of these “Get 5,000 Backlink” package tests, we noticed a decline in SERPs. Not just sometimes, but always. Times have changed. Those 5,000 dofollow forum profile links and mass bombardment of blog comment tactics are over. And frankly, I am glad for it. My suggestions are to keep your advertisement and your backlinking strategies relevant. Go for quality and not quantity. And when you do link out to a website in your content, leave it a dofollow link if it is relevant to your niche. Who you link to and give credit to does matter in the New SEO world.

  2. Emory Rowland

    I keep hearing people say that only low quality content sites were impacted. Not true. Sites with high quality content, low number of ads, high number/quality of links lost large amounts of traffic or existing traffic was replaced by traffic they were not optimizing for. It was ostensibly because the sites linking to them were devalued by Panda.

  3. Ken Durham

    Yes, there have been some innocent casualties. I’ve seen a couple of log files that show a personal visit from Google (a real person) just before being hit. Sites that you would think would be fine, as you mentioned. Why the real life visit and then the hit? I know Google said it was all algorithmic, but what they say and what they do seem to often be two different things. The one item in common was that these sites where in large markets (such as health), markets that some pay Google fairly well via Adwords. I’m not suggesting anything, but Google is a for profit business… I wonder what Matt C would have to say if presented the log files and then the immediate hit?

  4. Emory Rolwland

    “Why the real life visit and then the hit?”

    Maybe the personal visit means they aren’t so confident about what Panda was taking down. There have been what, four releases since the first?

Leave a comment