Using rel=”nofollow” Has Nothing to Do with How Serious You Are About Spam (or “Why Your Community Shouldn’t Be a Search Engine Black Hole”)

More than 5 years ago, rel=”nofollow” was introduced by Google as a way to limit the impact of blog comment spam on their index (and the indexes of other search engines that agreed to support the initiative). And (seemingly) every meaningful blog platform or software bundle jumped on board, making it a standard feature for their users.

Essentially, if you add rel=”nofollow” within the HTML <a> tag for a given link, you are telling search engines not to give that website credit for the link and for it not to affect that page’s ranking in the given search engine’s index. It can be a little more complicated than that, but that is a basic explanation.

From there, it was adopted in other contexts. Text link advertisements probably being the biggest one. When Google decided that text link advertisements were impacting the quality of their index in a way they couldn’t control, they required that text link advertisements receive the rel=”nofollow” attribute and, if you didn’t fulfill that requirement, you risked penalty or could be delisted from their index entirely. They even encouraged the general public to report suspected paid links. At any rate, that’s the end of my rel=”nofollow” primer.

Is Any Press Good Press?

On November 26, The New York Times published a fascinating story on a web based business run by a man who made it clear that he cared little about customer happiness. Far from it, he threatens and intimidates customers. And that is a flattering summary.

These negative experiences built up and have been shared online. If you search for the company’s name, DecorMyEyes, you can easily find a seemingly endless number of them. The man behind the store, named Vitaly Borker, shared an interesting theory.

He said that people commenting about his brand, good or bad, and in many cases providing links, was actually helpful because it pushed his site higher in Google’s search engine results. There is some truth there, as links are a part of the ecosystem that leads to results, but there is also a tremendous downside. Anyone who searches for the name of his store will not be likely to shop there.

In the end, does it offset? Who knows. But, one could argue that you can build just as many, if not more links offering good service, coupons and prices.

Google and Get Satisfaction Respond

Google quickly came out and said that they had put together a team and implemented a solution. I didn’t really view it as Google’s fault, necessarily. I don’t think we really want Google deciding who is good and bad, based on what words people use next to the name of the brand online. And Google acknowledged, in their post, that such a solution would be damaging.

Get Satisfaction, the customer support platform (that I have some mixed feelings about due to what I view as predatory business practices), was mentioned numerous times in the article. They allow people to open “communities” for a specific brand, without the brand knowing, in order to make a complaint or comment, and allow others to do the same.

This was done with DecorMyEyes and the portal for the brand is full of complaints. In the Times piece, Borker mentions that the first time he heard about the site was when a representative at the company contacted him to make him aware of said portal and encourage him to participate. In so many words, he told them to pound salt. He did post on the site, however, but it was to tell people that his “goal is NEGATIVE advertisement.”

Get Satisfaction felt that they needed to defend themselves or, at least, clear up what they viewed to be a misconception in the article: that link mentions on their site led to credit with search engines (via Heidi Miller). And that is why I am writing about this story today. In his post, Get Satisfaction Chief Technology Officer and co-founder Thor Muller said this (emphasis is mine):

… The story implies that links on Get Satisfaction positively accrue to the benefit of a company, even if they’re negative. Like any online community that cares to combat spammers, we code our user-submitted links so that Google ignores them for the purposes of calculating page rank (specifically, we attach “rel=nofollow” to anchor tags).

rel=”nofollow” = Caring?

I’ve been managing online communities for more than ten years now. You could say I’m serious about spam (as serious as it calls for, that is) and I don’t use rel=”nofollow” for anything member’s submit, at this time. Do you see the issue here? The implication of Mr. Muller’s statement is that if you do not use rel=”nofollow” on member-submitted content, you do not “care” to combat spammers. I believe this to be completely untrue.

And I don’t write this to target Mr. Muller – I write this because when he says it, I am sure that he is not alone in the belief and that other people share it and that is why I wanted to talk about it today.

I don’t have anything against rel=”nofollow” specifically – I know people who use it and totally respect them. I use it in some contexts now and may use it in more in the future. It is a tool, nothing more and nothing less. It is not the basis by which anyone should be judged when it comes to whether or not they care to combat spam on their community.

Over the years, I have seen many people try to set standards of some sort to determine or pass judgment on whether or not someone is serious about their community. Most of them have been suspect and this is one of them. Another one that I have heard is “if you use phpBB, you’re not serious about your community.” That one is worse, and also untrue. To say it simply, no one is more serious about online community than I am or the proper management of an online community than I am.

Putting rel=”nofollow” Into Perspective

We need to separate, very firmly, the idea of using rel=”nofollow” and the idea of actually caring about your community. They are completely unrelated.

rel=”nofollow” is a tool and it is a part of the arsenal. But, it is not without it’s flaws. And using it as a means to combat spam, comes at a personal cost and a community cost. I don’t believe that either approach is right or wrong – they both have their upsides and downsides – and I believe that they can both be good. This isn’t a be all, end all conversation and that is the point.

There are two main ways people use it, when it comes to online communities. They use it on profiles – meaning any links someone adds to their profile, such as their profile homepage, and their signature – and they use it on posts and other forms of submitted content, outside of member profiles.

The use of rel=”nofollow” effectively eliminates the search engine value of links posted on your site. Notice I didn’t say spam. It doesn’t just affect spam, it affects all links posted, if that is how you use it. And that is a downside. In my opinion, well managed online communities should be impacting Google’s index because online communities are where a nice bulk of online recommendation and influence comes out of.

When well managed online communities completely shut that ability off, I believe that it has a negative effect, not a positive one, on the quality of Google’s index and reduces the impact that your community can have on the greater internet. I’m not saying this should be a deciding factor, but it is something to keep it mind and why I believe that, if you use it, using it for profiles only is the way to go.

In short, think of the message you are sending. “I want Google to ignore all links that my members post because a small percentage of them might be spam (at least until they are removed).” This isn’t good or bad, just something to think about.

Some people use rel=”nofollow” as a means of creating what I refer to as a search engine black hole. They want people to link to them and help them rank well in search engines, but they want to hog all of the Google juice. They don’t want other sites to get credit for being linked to from their community because that would help those other sites rank well. Personally, I believe this messes with the ecosystem, but that is just my opinion.

Combating Spam

What combats spam isn’t an HTML <a> tag element, but actually managing your community. While some spammers will check to see if your site uses rel=”nofollow” and skip it if it does, many (most, in my opinion) don’t care and will post anyway. Combating spam starts, first and foremost, with actual people removing spam when it is posted on your site.

Part of the spam fighting equation is technology and rel=”nofollow” is part of that, but the main thing rel=”nofollow” does is stop Google from indexing all links posted within the area that you define (again, not spam – links period). But, rel=”nofollow” isn’t going to stop (most) people from posting spam on your site. It may, to some extent, combat spam within Google’s index, but when it comes to your community and what you and your members see, it’s impact will be limited.

There are more impactful uses of technology when it comes to your community and to dealing with spam. It’s important to recognize that technological spam prevention will mainly only impact automated spammers. A human can always get around them, if they want to.

I know this sounds funny, but e-mail confirmation and requiring login to post, though easy to get around, probably cut down on more spam than the use rel=”nofollow” would. It doesn’t matter if you require login now and get a lot of spam – if you didn’t require login, you’d get more just because it is that much easier.

Having a good CAPTCHA goes a long way. If a link is being consistently spammed, use something like Censor Block and block it. If you use a popular community software option, try adding CAPTCHA in a place other than the default location (Confirmation on New Posters for phpBB 3 is an example of that). Blacklists can be helpful, but be sure to carefully consider the quality of the data source because you don’t want to block legitimate people.

You can require that a member has a certain number of posts before they can post links. But, people can still spam your site. Even if it’s not hyperlinked, they can still put the name of the site or they can make as many cheap posts as they want until they hit that number.

These are basic concepts, nothing too fancy. If you want more ideas, check out the resources available for the piece of software that you use.

But, again, humans can get around these measures. They can create new accounts, they can have throwaway e-mails, they can complete the CAPTCHA, they can complete it wherever you put it, they can use proxies, they can use link shorteners to get around a link ban (are you going to block bit.ly from your community? Even if you do, there are a million more out there) and they can get around most anything else.

However, each of these measures can cut down the spam that you have to deal with by at least a small percentage and that can make them worthwhile. Just because people can circumvent something doesn’t mean it is without value or that it doesn’t stop some people.

How you demonstrate that you care to combat spam isn’t technology or rel=”nofollow”. Technology is a tool for you, a person. How you demonstrate that you care is by actually being present and removing spam when it is posted and reported. That is how you care. That is what any community that cares to combat spam does.