Here are the essential factors the search engines look for:

Ultimately, “anything for a buck” directories do not enforce editorial judgment, and therefore the listings do not convey value to the search engines.

To take a closer look at this, let’s examine some of the key statements from Yahoo!’s Directory Submission Terms:

These statements make it pretty clear that Yahoo! will in fact reject your submission if your site is not a quality site, and it will keep your money.

You can divide directories into three buckets:

The process is relatively simple for directories that pass sustainable links, as defined earlier:

The process is a bit harder for directories that do not conform to the policies search engines prefer. There are still some things the publisher can do:

Content syndication can be a powerful tactic, and we will cover this in more detail in More Approaches to Content-Based Link Acquisition. The concept is to develop content with the intent of publishing it on third-party sites. The site that is publishing the content provides a link back to your site in return for being permitted to publish the content. Ideally, this content should be distinct from the content on your own site to avoid any risk of duplicate content issues.

Widgets are another way of syndicating content to third-party sites. They have the advantage of coming with a way to package the content so that it is not seen as duplicate content. This is because they are usually implemented in JavaScript such that the web page publishing the widget calls back to a remote server to fetch the content. The result is that a search engine crawler does not see the content.

This also results in any links embedded within a widget being invisible to the search engine as well. However, it is possible to implement a widget in such a way that it has an HTML wrapper around it with a simple HTML text link in it, a link that is visible to the crawler.

Popular widgets can get adopted by a large number of websites, and can result in a large number of links as a result. Widget campaigns can also result in links to deep pages on the site.

Of course, the incentives can be over the line too. For example, proactively going out and buying links might be a way to rapidly acquire links, but the search engines are increasingly good at detecting these types of links algorithmically (and discounting or even penalizing them).

Google also provides a method for third parties to report paid links directly to Google on an anonymous basis. There is every incentive for those third parties to do so.

Based on these risks, buying links is a dangerous practice, and it is harder to safely execute than it appears to be on the surface.

Another dangerous tactic is doing a large percentage of your link building through reciprocal links. Once again, this is easy to do in principle. It is not hard to find sites that will accept the “link to me and I will link to you” value proposition.

However, the search engines potentially view this as barter. They are not likely to question a few selected link exchanges with sites closely related to yours. It becomes a problem when the link swapping becomes a significant portion of your backlink profile. That is a situation that looks manipulative, and they will most likely discount those links.

One of the key points is how the publisher can make its pitch interesting enough for the potential linking site. As noted earlier, this starts with understanding what content or tools the publisher has on its site that would be of interest to the potential linking site.

Sites do not link to other sites for the purpose of helping those sites make money. They link because they perceive that the other sites’ users might value the content or tools on their own sites.

This relates to the fundamental structure of the Web, which is designed around the notion of interlinking related documents. Positioning a site or a section of a site as being related to the potential linking site is a requirement of each link-building request.

With high-value sites and very-high-value sites (as defined in Table 7-2), it may be worth spending a bit of time and energy on defining a compelling value proposition. In many cases, it is even worth doing custom content development to increase the perceived value and relevance of the content to the potential linking site.

In all cases, developing a link request value proposition begins with understanding the nature of the potential linking site’s content, and then deciding how to match up the content of the requester’s site with it.

Once you have done this you are in a position to contact the site publisher and communicate what you have to offer, and request a link.

The most important thing to remember is that the person you are emailing to request a link probably did not wake up this morning wondering what links she was going to add to her site. And certainly, she was not expecting or waiting for your email. Basically, you are interrupting her to ask her to do something for you, and she has no prior reason to trust you. Based on this, there are few simple guidelines you should follow when making a link pitch:

One of the more popular techniques is to buy links. This has two significant advantages:

In the early days of search, publishers developed link farms and link networks as tactics for gaining cheap links. A link farm is a website or a group of sites whose primary reason for existence is to cross-link between themselves and other websites. Generally speaking, the links are created through aggressive reciprocal linking.

Since these sites are typically very heavily interlinked, they can be pretty easy to detect. Part of the reason is that since they have little redeeming value, they typically do not have high-value links coming in to them from other sites, and most of the links result from various cross-linking schemes.

Link networks are a similar concept. The network exists for the purposes of creating links between sites, and it can be a bit more sophisticated than a link farm. For example, you could create a club where publishers agree to contribute a link in return for getting a link from somewhere else.

If managed with great care, the clustering of links between sites can be limited, and this can be a bit harder for search engines to detect. However, the tactic remains highly exposed to a disgruntled webmaster simply reporting the scheme to Google.

A related concept is the notion of three-way link swaps (a.k.a. triangular link swapping), where Site A links to Site C in return for Site B linking to Site A. In this scenario, Site C may be the site the publisher is trying to promote, and Site B may be a site it uses to provide low-value links to people it trades links with.

This is almost always a scam, because Site B is probably a low-value site with little to recommend it. So, the publisher of Site A is providing a good-quality link in return for a low-quality one.

Spam tactics can include the concept of creating a bot that crawls around the Web looking for open forums and blogs and leaving behind automatically generated comments. Clearly this is spam, as no human is involved in the comment process (other than the programmer) and no effort was made to read the blog post or forum where the comment was left.

The great majority of these comments are deleted or NoFollowed by the blog or forum software content management system (CMS), but the spammer does not care because she is operating on a large scale. If she leaves behind 1 million comments and 98% of them are filtered by one means or another, she still ends up with 20,000 links.

This is, of course, a very risky tactic. The search engines may be able to detect this behavior algorithmically, or competitors can recognize it and turn you in via a spam report. We do not recommend this tactic.

As we previously discussed, you can use widgets as part of a link-building campaign. However, this tactic can be abused as well. One popular way of doing that is to develop a JavaScript-based widget and then embed a link back to the widget publisher’s site in the NoScript tags. Since this is the code the browser presents to users who do not support JavaScript, and the search engines do not run JavaScript, the crawlers see the link and not the widget.

An example of someone who did this is discussed in the post “Another Paid Links Service Disguised As Hit Counter” (http://blogoscoped.com/archive/2008-02-06-n19.html). Four days after this post went up, the sites referenced lost all of their high rankings in Google, and therefore lost most of their traffic.

However, be aware that making the link visible is not enough to make this practice legitimate in Google’s eyes. Google has confirmed that it considers tactics such as the use of unrelated widgets for link building a no-no, even if the link is visible.

The underlying reason for this is that if the link is unrelated to the widget, it is pretty unlikely that the link given actually represents an endorsement of the web page receiving the link. The goal of the person installing the widget is to get the benefit of the widget’s contents.

As another example, Google considers the use of so-called sponsored WordPress templates with embedded links, even if they are visible to be spammy as well—unless, of course, the publisher distributing the WordPress template is in the WordPress template business. The key issue that needs to be addressed in these types of link-building campaigns is relevance.

One of the most common uses of NoFollow is to NoFollow all links in comments on a blog or in a forum. This is a common and legitimate use of the tactic, as it is used to prevent spammers from flooding forums and blogs with useless comments that include links to their sites.

However, as you might expect, people did implement some common scams using NoFollow. One simple example of this is that someone proposed a link swap and provided the return link but NoFollowed it. The goal of the site using the NoFollow was simple: to get a link that passes link juice, but not to provide any link juice in return.

However, Google announced that it had changed the way it handles NoFollow in June 2009. At this point in time, NoFollow still prevents the passing of link juice to the site receiving the link, but it is lost to the linking site as well. As a result, this scam no longer benefits those who attempt to use it.