* Mass directory submission
* Bulk article submission
* Forum signature links
* Comment spam
* Link exchange requests

Of course sometimes these techniques *might* give some results but the way that they are generally approached makes the chance of any ranking improvement almost zero.

Overall the author of the post is correct in the context of what he says about these link building techniques regarding “the way that they are generally approached”. Let’s see in detail what happens in case of each one of the mentioned techniques:

- Directory submissions: if we take a look at automated directory submission services that offer you to submit a site to a gazillion of directories/search engines/whatever they call them for $20 or so – this is useless waste of money. The way they collect the directories/search engines/other places to submit to is by scraping the search results for something like “submit URL”. They do not analyze the quality of the sites they find, nor do they check if the submissions ever appear on the targeted sites. Moreover, directories in
general have long lost the value they once had – hence it requires a long and thorough research to find the ones that are still worth using. The more valuable ones are niche directories – but there are still some general directories that Google sees as authority that still give good links if you submit to them. Local directories are also good – but everything needs to be checked at the level of each given directory and generalising is of no use here.

- Bulk article submission – useless unless you modify the article at least slightly for each article directory and only submit to a few carefully chosen quality article directories. Moreover, I’ve seen many times how backlinks acquired by a site through article submissions disappear due to the particularities of the article directory site structure. Article directories do not have the same value they used to have either any more – but some still preserve their authority status.

- Forum signature links – if you are using these on a forum where you have hundreds or thousands of posts they may be useful for direct traffic – if the link really appeals to the forum users – but not as backlinks. Neither Google nor Yahoo count more than a couple of links off the same domain. However, I have found some forums to be good for indexing new sites if you place a link into the profile and make a dozen posts or so. Again, depends
on the forum and on the individual tactics used.

- Comment spam – if done intelligently (and manually) might be good for direct traffic again – but not if it is automated. Blackhats use it in its automated form for getting mass amounts of links as Yahoo doesn’t care about the nofollow.

- Link exchange requests – again, depends. Reciprocal link exchange definitely doesn’t make sense – but we have been getting some really nice links with more complicated schemes (3-way, 4-way, etc.) It all boils down to what sites you target with your link exchange requests, how you word them, how your schemes are set up, etc. I for example always stay away from any pages marked as “Link Exchange” and the like – the more natural it looks the better. And of course 99% of the linkexchange requests you get into your mailbox sent by other people can simply be deleted right away.

Finally, let’s not forget that every link can have two kinds of value: it can pass link juice and improve a site’s rankings, or it can bring direct traffic to a site. The best links are the links that manage to accomplish both goals – no matter what methos was used to acquire them.

And indeed, some time later what do we see but an updated, all web 2.0-looking Digg! Thank God still no nofollow crap there…

Look at your backlinks and answer these questions:

1. Yahoo is normally faster at picking up your new links. Do your links get indexed by Yahoo? Which ones of them show as omitted results?

2. Do the sites linking to your site get indexed by Google? Are they in the main SERPs or marked as supplemental results? Does Google show these sites in your backlinks after it picks them up or does it omit them?

3. Does your deep link ratio (link to DLR calculation explanation) change over time? If so, in what direction? How does it relate to the general DLR tendencies in your industry?

4. And of course, while this might sound like common sense, are the sites linking to you topically related to your site?

These are just a few things to analyze, of course there are more factors that influence your rankings (you might want to read the series I did on site evaluation, especially the part on link evaluation). But taking a quick look at these few things can help you keep your link building campaign on the right track and minimize wasting efforts.

(When looking at links, we need to look at the source code – and Firefox is my best friend in this case – I can simply select any portion of a page, right click it and view the source code for this selected area only – so much easier than in IE!)

Compare the following link examples:


< a xhref="http://www.somedomain.org/" mce_href="http://www.somedomain.org/" >some text< /a > - a normal link

< a xhref=”http://www.somedomain.org/” mce_href=”http://www.somedomain.org/” target=”_BLANK” > some text < /a > – a normal link that opens in a new window

< a xhref=”http://www.somedomain.co.uk” mce_href=”http://www.somedomain.co.uk” rel=”external” >some text< /a > – a normal link that is marked as external – this is just an indication that it leads to a different site, no more no less

< a xhref=”http://www.somedomain.com/” mce_href=”http://www.somedomain.com/” rel=”nofollow” >some text< /a > – a link that is not supposed to be taken into account by the spiders (however, Yahoo ignores “nofollow”)

< a xhref=”http://www.somedomain.net” mce_href=”http://www.somedomain.net” rel=”external nofollow” >some text< /a > – a link marked as external that is not supposed to be taken into account by the spiders

< a xhref=”http://www.archive.org” mce_href=”http://www.archive.org” class=”external text” >The Internet Archive< /a > – a normal link

< a xhref=”http://redirect.alexa.com/redirect?www.wdvl.com%2F” mce_href=”http://redirect.alexa.com/redirect?www.wdvl.com%2F” isdata=”true” >Web Developer’s Virtual Library< /a > – a link with a simple redirect

< a class=”yschttl” xhref=”http://rds.yahoo.com/_ylt=A0SO5ig9xSJENIMAGOdXNyoA;_ylu=X3oDMTE2YTlpczVrBGNvbG8DZQRsA1dTMQRwb3MDMQRzZWMDc3IEdnRpZANTTkdZMl8x/SIG=119qeehko/EXP=1143215805/**http%3a//www.alexa.com/” mce_href=”http://rds.yahoo.com/_ylt=A0SO5ig9xSJENIMAGOdXNyoA;_ylu=X3oDMTE2YTlpczVrBGNvbG8DZQRsA1dTMQRwb3MDMQRzZWMDc3IEdnRpZANTTkdZMl8x/SIG=119qeehko/EXP=1143215805/**http%3a//www.alexa.com/” >Alexa Web Search< /a > – a link with an encoded redirect

< a xhref=”http://www.cre8asiteforums.com/redirect/jump.php?url=ten.niamodemos.www%2F%2F%3Aptth” mce_href=”http://www.cre8asiteforums.com/redirect/jump.php?url=ten.niamodemos.www%2F%2F%3Aptth” target=”_blank” >some text< /a > – a link with a redirect, the target URL is reversed

< a xhref=”/weblink?MGWLPN=CATA&MGWAPP=g&id=778073″ mce_href=”/weblink?MGWLPN=CATA&MGWAPP=g&id=778073″ class=”link” onmouseover=”window.status=’http://www.acu.ac.uk/’; return true;” onmouseout=’window.status=”";’ target=”_blank” >Association of Commonwealth Universities< /a > – a masked redirect aimed at misleading those looking at the link on the page – when you mouse over the link the status bar displays the target URL but the actual URL in the href is encoded and redirected

So, while for traffic all these links would be pretty much equally good if sitting on a really good and targeted site, for the search engines they would all hav ea different value. While you want normal regular links, you might want to avoid those with all kinds of redirects and “nofollow”. It is argued whether or not the search engines can and will pay any attention to target=”_blank”  and style class names such as “external”, but this sum up should give you an idea.

Other people’s posts on this topic:

Michael Martinez at SEOMoz on link quality

Andy Hagans on what’s a link worth

I have put together a quick checklist for evaluating sites. These are the things I’d be looking at when evaluating a site:

• the site’s overall look and feel;
• the site top level domain (com/org/edu/gov etc.);
• site age;
• Google PageRank (ok don’t laugh, I will explain later on what this is good for);
• number of indexed pages in the main search engines and how it correlates to the actual number of pages on the site;
• quality of listings of those indexed pages;
• number and quality of backlinks indexed by Yahoo;
• the site’s history as seen through Web Archive;
• the site’s traffic estimation.

Let’s now look deeper into each criterion of this checklist.

The site’s overall look and feel – very often, this will be the only thing you will look at. Think of spam sites (a.k.a. built for AdSense/insert any other ad network here) – you will probably make your judgement as soon as you see the site – but not all sites are so easy to evaluate.

The site’s top level domain – .edu’s are the best out there of course, and .gov are also very good. Country-specific domains might be of importance to you as well if you’re targeting something local/non-English speaking audience.

Site age – you know Google loves old sites, so the older the better. I use either a whois option in Firefox SearchStatus plugin or whois.sc which gives you a lot of information about a domain. You might also want to pay attention to the period of time the domain is registered for – the longer the better.

Google PageRank – ok even if you don’t believe it matters much any more, you still want to look at it. If you see a grey bar instead of any pagerank it might mean the site is not indexed/banned by Google – so that’s a sign you need to dig deeper to find out why it’s grey.

Number of pages indexed by the main search engines – this will give you an idea of a site’s level of exposure in the search engines, and also if there are any problems with the site. If a site is thousands of pages and you only see a few indexed – and the age of the site suggests it hasn’t been launched just now – beware of possible deindexing going on.

The quality of indexed pages listing – are there any supplemental results?

Number and quality of backlinks Yahoo shows for the site – ok you may want to check Google’s backlinks for the site as well but Yahoo shows more (however, there are ways to get Google to show you more backlinks it knows about than it intends to ) Keep in mind that link: search in Yahoo might not show you all you need to see and you might also want to do a linkdomain: search.

The site’s history in Web Archive – this is especially helpful if there are problems with the site and you want to see whetehr there has been something on the site earlier that could get it banned.

The site’s traffic estimation – I use Alexa for this. Although Alexa’s data is not particularly accurate and is prone to manipulations, it can still provide you with an estimate of the overall tendency (e.g. if the site traffic is increasing over time or vice versa).

The important rule, however, is not to base your judgement solely on one of these criteria. Use all the data you obtained when making a decision about a site’s quality.

This is the checklist for any type of sites. Now, if we’re evaluating a directory, there are some additional criteria that can be considered:

• I found that using Aaron Wall’s Directory Archives can help you maike sure you’re dealing with a quality directory – if it’s listed there it’s a sure sign you’re dealing with a quality directory. Aaron’s “directory of directories” is all hand checked edited stuff – however, he says he’s not maintaining it any more so some stuff there would be outdated.
• Consider page rank of not only the home page but also the actual category page your link will end up on.
• See how many links are already listed on that pagea nd how far from the top your link would be.
• Find out the terms of submission (free/paid/both options available) and see which one of the available options makes most sense and serves you best.

If you’re evaluating a blog, there are some more blog-specific things to consider:

• Questions to consider when evaluating the blog: Does the blog look like it’s being maintained by a human or does it look like auto generated content? How many outbound links does it have? How recent are the last posts? How often new posts are added? What’s the blog’s blogroll like? Where did you find the blog? Are the comments/trackbacks allowed?
• A blog’s age cannot always be discovered through whois as many blogs are hosted on blog hosting servers (e.g. Blogger, WordPress.com, etc.). But a good indicator of a blog’s age is its archives – how far back do they go?
• PageRank is not necessarily much of an indicator for blogs – a blog can have a large following but a PageRank of 0. The blog’s audience can be estimated through the number of comments – keep in mind, however, that most of the blog readers just read it and hardly ever comment – though this can vary from blog to blog.
• Technorati data: in Technorati, we can see the number of blogs linking to our blog in question. High profile blogs (A-list blogs) are also included in the Technorati top 100. Blogs that people have been adding massively to their Technorati favourites are listed at Most Favorited.
• There is a tool that might be of some use as well when measuring a blog’s “influence”: BlogInfluence – it shows links found on Technorati, Yahoo, Bloglines subscribers and Google PageRank. OK, Bloglines is just one RSS reader and this doesn’t give you an estimate of the total number of subscribers – but if you see a lot of subscribers in Bloglines you know you’ve found something of a good value.

When making an estimate of a blog, mix these criteria in with our usi\ual checklist and you’re sure to dig some valuable data. But even if a blog’s current traffic and following is insignificant, the author might still be active enough to be able to become a new A-lister tomorrow – so maybe it’s worth it to make friends with that blogger

Other people’s posts related to this topic:

Todd Malicoat’s post on figuring a site value when buying/selling sites
ShoeMoney’s podcast on directories

Suggested reading:

A Librarian’s Guide to Finding Web Sites You Can Trust – from Google’s Newsletter for Librarians
Purdue University tutorial on Site Evaluation
Google’s Spam recognition Guide for Raters (a Word document, provided by Henk van Ess)

Be sure to catch the next part of this series where we will talk of link value.

So what do you need to create linkbait?

1. Throw together a lot of buzzwords like viral marketing, SEO, iPod, Podcast, Google, Yahoo, whatever

2. Make sure you mention popular people and/or A-list bloggers like Matt Cutts or Jeremy Zawodny or Robert Scoble or Steve Rubel or Seth Godin or Mickey Mouse or all of them together in the same post. Throw in other names depending on the industry you’re targeting

3. If there’s no news, make it up. The more your writing looks like you’ve been hallucinating when writing it the better

4. Take some rumours and present them as facts, invent some additional details to make them more impressive – nevermind the lawsuit those affected by your interpreteation would possible threaten you with later on – by that time the traffic and the clicks on your ads would have generated you enough income to cover the cost of a good lawyer. Heck, the lawsuit will even generate more traffic

Disclaimer: the author bears no responsibility for the consequences of applying this advise partially or completely to your blog/web site.

Whatever it is you write, it has either a long term value (e.g. site content) or short term value (e.g. a press release), though sometimes something that was originally meant to serve as material to attract short term attention can end up serving you long term (e.g. syndicated articles).

1. Finding stuff to write about and researching the topic’s popularity (most important for effective exposure of short term value content like press releases):
1. what do people talk about currently? – to find out, you can use:

i. Bloglines – create an account, subscribe to blogs posting on your topic of interest – this is good for both short term and long term value content writing, in the meantime you can subscribe to your own blogs to create extra exposure for them

ii. Technorati – tags by popularity

iii. Feedster

iv. PubSub – subscriptions to searches on certain terms – the info is updated as new posts matching the query are found, sample subscriptions are available

v. Furl: latest headlines, most popular

vi. Free Press Release site – for ideas what to write about for immediate publication (long term as well)

vii. Google news – the latest news at a glance, can be researched in depth for a certain specific topic

viii. Google Blog Search – what’s been posted recently on a certain topic – when sorted by date, shows what’s hot short term, when sorted by relevance can be used for long term material research

ix. Yahoo News – same idea as Google News

x. Yahoo News Search – searches news and blogs, search results can be sorted by date

xi. Wikipedia’s main page – they display a featured article as well as current news. Both can give you an idea of what to write about both short term and long term.

xii. Blogpulse – Top news stories, Top Blogs, Top Blog Posts, key phrases

2. Watch what your competitors do (e.g. at i-newswire.com or any other press release site, search for your competitors’ site URL and look through their press releases) – but instead of simply copying them, critically analyze their activities and try to make your work better by using and improving their good ideas and avoiding their mistakes.
   
   
2. Link bait concept – think what people will be interested in and likely to link to. Not necessarily something positive – negative reaction is good too for linking (see this article by Rob Sullivan on types of linkbait)
   Keep in mind, however, that in some cases negative stuff for link bait might not be a good idea – if it undermines the site’s reputation in the potential customers’ eyes, for example.
3. How not to write: www.clickhereyouidiot.com (the ideology of sales letters and what’s wrong with it when driven to the edge – got this one from a Threadwatch post by Aaron Wall)
   Consider an article title like this: Unimaginable Power Of [Our Product/Service]- Most Sacred Secretes Of [Whatever It Is We Offer] Revealed - what’s wrong about it? Well, everybody smells an exaggeration and hype before they even start reading your stuff.
4. Checking visitor logs for topics to write about – this concerns sites/blogs rather than articles/press releases but still can give you an idea
5. Checking rankings – are there any unusual terms the site ranks for? If they are still somehow related to the site’s topic they might be worth covering/expanding (tools: data.rankcount.com, googspy.com – thanks Aaron for these two links as well)

Recently, however, I have read this post on sitewide links by Patrick Gavin and it suddenly dawned me: THERE IS NO SUCH THING AS SITEWIDE LINKS PENALTY!

If you haven’t understood yet what I mean here’s a simple explanation: say you have a site that has sitewide links on 3 sites and each of those sites has, say, 20 pages. This means, according to the old algo, that you had 60 links pointing at your site. When the new algo has been enforced all the sitewide links have been discounted, ach of them now being considered as one link. How many links does your site have now? – Right, 3! If that caused a drop in your SERPs it only seems logical…

The interesting point about nofollow links is a different approach to links and their value. Loren Baker’s post says:

Goes to show that being covered by, participating in the comments of or obtaining a link from a blog or site using “no follow” link condoms are still of some search engine value (not to mention the real value – traffic and reputation).

- while I’d say, the real value might be much more important. I like to say, a link is a link, period. Don’t know whether “nofollow” stops the links from telling the search engines anything about a linked to site’s reputation – but the traffic bit is certainly there, and don’t forget that it’s not only the bots that are looking at the links but people do too, as well as click on them.

This bit is especially insightful:

So, even links from no follow coded pages can appear in Yahoo’s backlink index, as those links are actually being followed by search engines. I’ll also go out on a limb to say that the aspect of the no follow link is a bit outdated when it comes to search indexing & ranking, as even if those links don’t pack “ranking juice”, they will help to define you and your site’s:

* online reputation
* profiles of users who visit your site and sites they have visited in the past
* clustering of similar pages and content
* and identifying links to pages which may have not had SEO in mind when obtaining those links or contributing to such discussions.

Such variables, included in a new age algo (and some in published Google & Yahoo patent apps), sound quite revelant to me, even if those links have no “PageRank” or “SiteRank” juice.

Who cares about the old Page Rank when nowadays the number of variables involved is so much greater, making this so much more interesting?