22 Jun
A link, is a link, is a link. Any link will give your site some search engine benefit, well almost. Here are the exceptions and how to get the best from link building!
A Chain Of LinksLink building in one way or another is the life blood of search engine optimisation. You can make all of the changes that you like to the code of a website page, but if you do not have enough links, or enough of the correct links, all your search engine optimisation efforts could be in vain. So, what do you need to be considering?
Links that do not count• Links including rel="nofollow" count for nothing because this explicitly tells the search engines to ignore this link at the request of the website owner. It doesn't mean search engines will not follow the link to see if it is a new page, but no favour is passed.
A scary face, a bad neighbourhood?• Links from bad neighbourhoods are ignored because they are bad. Difficult to say exactly what these are, other than websites that have offended the search engines. But the next point probably covers these also.
• Links from pages that are not cached are not known about, simply because the search engines ignore the pages! If the page does not have a Page Rank, worse still the website does not have a Page Rank, then it is likely that the link might not be seen.
• Link hidden behind redirects and javascript are supposed to be ignored because the search engines do not see them and do not follow redirects. However, I have seen search engines on pages that are only available through redirects, so this might not be completely accurate.
The best linksSo, what is the best link to have? Well the higher the Page Rank of the page your link is on the better, if all other factors are equal. For example, similar number of outbound links, same theme and so on. Without going into full and boring detail of the Page Rank algorithm (which has no doubt changed since it was first published), it is generally accepted that 85% of the Page Rank of the linking page is shared equally between all outbound links. So if there is one link on the page that gets the full 85%. If there are 5 links then each gets 17%. If there are 17 links on the page then they only get 5% each and so on.
It is also generally accepted that because of this splitting down of the value that sites need to have a Page Rank of at least 2 to hold any weight. This is because a site with PR1 is going to have inner pages of PR0, which when they link to you give next to nothing back.
What links to your website?But how are the best links constructed then? What should the anchor text be for the link? Well looking natural is best. Many people say not to use your site's URL or name and instead pack it with keywords, which does have advantages. But, a mixture of everything works best. Search engines have enough data to work out what is natural and what is pre arranged.
I like in link building to have a proportion of the anchor texts being general, for example the URL of the website of the website name. Build a few good links and then start working with new links that use good keywords.
Use The Correct Anchors
Use The Correct AnchorsTo find the best keywords to use, do a little research. Do not go for your most competitive keywords with the highest amount of traffic straight away, you will almost certainly fail. At first, build on the smaller volume keywords with less competition. For example, say I wanted to build for a mortgages site with the phrase
Website Promotion Services, I would start with
Free Website Promotion Services,
Easy Website Promotion Services and
Fast Website Promotion Services.
All terms with traffic and competition, but all achievable. And the best thing is that by the time these are all successful there should be a degree of success with the main keyword as well. It might not be number 1, but it would be a long way there.
Why use keywords as anchors?This is important and something that many people miss. Link building not only gets you the 'votes' for the website because the links exist but also 'votes' for the terms used within the links. If the anchor text is 'Click Here' then that is what you are getting the 'vote' for. However, if it is 'More Website Traffic' then that instead is where you are working towards.
Content really does matter!
Carefully selecting your content!Content of the page you are linking in does matter, for two very important reasons.
First, if the page is about health and your website about mortgages, then they are probably totally unrelated and the Page Rank effect will be reduced. But, it is better than nothing and not to much of an issue when the links are in posts and articles as that can be directly related to your website.
Second, link density. I have had websites built entirely from guest posts, which contain two or three links per 500+ word post to the author's website. These sites are successful, have lots of search engine traffic and the links out do not upset search engines.
However, I have seen sites that just exist for sponsored writing. These have one, two or three outbound links and the posts vary in length from 30 words to 100, sometimes 200 words. On occasion I have even seen these posts just literally containing the link. But quite often these sites are punished by the removal of Page Rank, dropping 2 or 3 ranks.
So what is the difference between the two? Well, just link density. Whilst it is fine to have a couple of links in a few hundred words, constantly writing posts of 30 - 100 words and embedding links alerts the search engines. Whilst they might not punish the site, they probably will punish the page - the page with your links on it.
This means that your links should work best if there are one or two of them on a meaty page - 400 or more words. Shoving 3 links into a 30-word article is asking fro trouble, but burying them in a bigger piece looks natural and appeases the search engines. And that is out aim!
So, what links do we want?In summary, we want links with a variety of anchor texts, some generic and some based on well researched keyword phrases. They should be on PR2 and above websites on content pages with at least a 400 word article that relates to the website linked to. Easy!
I make no apologies that this has been a post nearly three times longer than normal! There is a lot of information to get started with and it is easier to have it all in once place than split over 3 or 4 posts. Do let me hear your comments!
Comments Off
15 Jun
One of the methods frequently described for building plenty of incoming links is creating free themes for WordPress. But, you could be wasting your time! Read on to find out why and how to make them work.
If you have been reading around about SEO and you are any good at designing and building websites, then you might be tempted to build a few WordPress themes and make them publicly available for free. In return, you just slip your link into the footer of the theme and hope that most users keep it intact. That way, you have suddenly created hundreds of inbound links.
But, these links are probably being ignored! In a recent experiment I linked from a well established PR3 website to a new PR0 website and watched it quickly jump up the search results. But, when I put the link on more of the thousands of pages on the PR3 website, the PR0 website fell back down the results to lower than when I started.
This is an obvious sign that Google does not give benefit to anything approaching site wide links and if you have placed your link on the footer, it is appearing on every page of the site. And it could even be a brand new site, so little to no benefit to you.
But my experiment continued to blocking the non home page links to see what would happen then and what I noticed was that the position of the PR0 site slowly went back up, almost reaching the position it had been at before I put the links site wide. Not quite there, but then it probably will take quite a while for Google to work out that some of the deeper inner pages should be ignored.
So, if you are building themes for SEO purposes what do you need to do to make sure that the links are counting? Well, the answer is to not be too greedy and there are two options.
The first is to use the WordPress is_front_page function to detect whether the footer is being displayed on the front page and then, if it is, to display the link. This way your link only ever appears on the one page of the website.
However, this does not get the best results for traffic. If the blog is receiving visitors then some of these might like your work and also want to download a free theme. For these people you need to be displaying the link on every page.
So my preferred method, even though it might not quite work so well on the search engines, is to negate the process. Always display the link on every page through the footer, but if it is not the front page, then add rel="nofollow" to the link. This way you are telling Google to ignore the site wide aspect of the link, whilst getting the benefit of the home page presence (and probably the page with the highest page rank) and also allowing further visitors to discover your work.
A little complicated, but hopefully the best solution all round!
Comments Off
12 Jun
My own experiments with site wide links have convinced me that Google does indeed punish site wide links. Why though, would this be a fair act?I regularly conduct link building experiments between the various sites that I control and one recently has been about the effects of so called so wide links. By using a single link from a PR3 website I was able to move a PR0 website further up the search engine results, but when I put the same link on many pages the PR0 site dropped down to below where it had been before the experiment began. And this was with 20 links detected on a site of thousands of pages, which has been in existence for about 4 years.
To decide why site wide links could be a problem, first let's look at why site wide links might be used. There are a variety of situations:
1) You own both sites and want to link to the other for further information
2) You have paid another site owner to link to your own website
3) You designed the website and want credit for the work
4) You designed the template that the site is using
When you look through the list of likely causes of site wide links, it is a very limited list of reasons and all basically because there is some sort of connection from the site with the links to the site linked to. There are not really many reasons why an existing website might suddenly decide to place a link from every page of that website to another website. Especially not if that link is placed in the footer (as in my experiment) or in the side navigation, such as in a Blogroll.
Now both of these are tried and tested techniques for gaining traffic. I use both on various sites for traffic exchange and see the traffic arriving. For example, I have seen visitors jumping between my sites via the blogroll links and my website design business has had plenty of customers arrive who have seen and clicked on the link in the footer of another customer's website.
But this is not what Google is trying to count when it is looking at links. It is a shame really that we are being penalised for placing links for traffic! However, there are ways around this detrimental effect as my experiment also discovered.
I found out that by blocking the links from the search engines on all but the home page, then the position of the PR0 site did rise back again. It would appear that Google wants us to be honest on every link we place as to whether it is just for traffic, or whether we would like some authority placed on the link.
So, why would Google be ignoring these site wide links - something that seemed to trigger with very few links? I suppose that it is because it is a way of identifying links just for search engines. Rapidly built links across the site are fake links, but odd links here and there are probably for real.
Comments Off
11 Jun
I have sounded off before on these pages about how I consider links directories are a waste of time. Look at the maths and you should agree - you need thousands of links from a PR0 links directory just to have any potential impact.
What surprises me about the constant flow of emails, such as today's (shown above, click on it to view it in detail) is not that people are still link building and wanting directory link exchanges. That is understandable - new website owners research what they need to do and there is enough old information out there and plenty of these convincing emails doing the rounds to convince them that link exchanges might in some way help.
No, what amazes me is that companies are still paying for this as a service. How much money are these website owners are paying so called SEO experts to add a script to their website to create a links directory and a contact form and then just to send an email out to their contact list?
At least usually the emails from this company are associated with a links directory that does actually exist and is usually linked to from the home page, so there is a remote chance of getting a little benefit. But, if suddenly this site is link exchanging, they are going to be seen by the search engines to be adding tons of links to all of the usual suspects that still believe Emily's email and think that link exchanging works.
On a bad day, Ms Ackers and friends have been known to send me dozens of these emails. I'm hoping it is just the one today!
Comments Off
10 Jun
I never really believed that site wide links could be damaging to a site, but I was wrong. I set out to disprove the theory and ended up proving that there is a damaging effect from them.I had frequently read on SEO forums that some people thought that site wide links were damaging for a website, but I was not convinced. I tended to agree with those that said they could not be damaging, reckoning that a site wide link was something out of the site owner's control and, therefore, not an influencing effect that Google would consider.
But, people did keep coming up with evidence that they were damaging. However, these people never provided full evidence wanting to protect their sites. I thought that maybe that the reason there were two different schools of thought is that the truth lay somewhere in the middle, so I set up my own experiment to prove, or disprove, the theories for myself.
I run many of my own sites and one of them is a relatively new site, which I had recently "claimed" on a ranking system. This claim involved the placement of a random claim sentence on a post, which also appears in many other blog posts. This, I decided, was great for my needs as no-one else would be optimising for a junk sentence, but my site was mid way down the 4th page of Google search result when the sentence was searched for.
So I inserted a link to the post page from a well established and ranked site, using the sentence as the anchor text. No surprise when a few days later the blog page suddenly moved to the middle of the 2nd page of Google and a few days later to the top of that 2nd page. Evidence that link building works and that the link had been picked up and done its business.
Next I made the link site wide before Google again visited the linking site. A few days later I checked the search engine results and that post was no down on the bottom of the 3rd page and then the top of the 4th page a day or two later. And yet, this was after Google had only picked up 20 pages containing the link.
So never mind site wide, 20 pages out of 3,000 pages on a PR3 site that is 3 to 4 years old made the blog site go from middle of page 3 to top of page 2 then drop to top of page 4.
Therefore, the bad news is that Google ignores site wide links and the tolerance was very low, about only 20 pages. Although rather than an actual punishment as a lot of people claim, it did seem to be more of just ignoring the effects of all of these new links that had suddenly appeared. But as a website designer, or for those designing blog themes, this could mean that placing a link in the footer of new websites or in new themes could be not getting the benefits you are hoping for.
Comments Off
