It can be extremely difficult to test the effectiveness of your SEO changes on many elements of your site. Controlled environments are rare, and even then you don’t know what the search engines are going to change come tomorrow that is going to affect your SEO strategies today. However, despite all the challenges there is one area where you can do some “live testing” without worry – on your anchor text for internal links.
By now we all are aware of the value to not only linking to high quality external sites, but also the value of linking to other relevant content on our own site (hence the name internal linking). The big question becomes at what point does that internal linking stop helping us, and does SEO optimization of anchor text have a positive or negative effect on the rankings?
Since anchor text is invisible to the end user for the most part it is relatively easy to test it on a live site to see how it affects rank. Here are some of the SEO techniques I’ve found that works and doesn’t work well when dealing with internal links:
- When it comes to the homepage, the anchor text doesn’t matter that much. Your main landing page of any site is not going to benefit from spending too much time on anchor text for internal links. It appears for the most part the search engines will judge its worthiness from links coming from outside your site.
- Adding links to other pages internally provides benefit, but only up to a point. After the first dozen or so links to a page it looks as if the search engines stop assigning any new value to internal links and anchor text that point to the page you are looking at. This does not mean you shouldn’t link to the page, but if it’s a popular page you find yourself linking to a lot in other content then don’t spend a lot of time worrying about the anchor text every time you form a new link to it.
- Don’t overdo it – especially with keyword stuffed anchor text. We’ve recently seen where Google will penalize you for trying to keyword stuff anchor text of internal links, especially when you have a lot of them. Remember, you should be linking for value – not for the heck of it. If every other word on a page is linked to something else you might want to re-examine your strategy!
SEO techniques are not easy to test – especially since you are dealing with live data on a constant basis. However, by following good SEO practices and making tweaks where appropriate you can get a lot of value out of your testing, plus learn what works and doesn’t work for rank – lessons you can keep using for client after client, site after site.




February 9th, 2009 at 9:10 pm
What methods do you recommend for testing? Any tools you use for this that you recommend? Or do you just use the standard analytics?
February 20th, 2009 at 10:51 pm
Do you know or have you heard anything regarding any differences between an internal link to a different page in the same site (ie. http://www.site.com/1.html linking to http://www.site.com/2.html) and an internal link linking to a different spot on the same page (ie. linking to a named anchor )?
I was curious if it was better to do one or another. We have a small description on one page, and are trying to figure out if it’d be better to link to the longer description on a different page, or just toss it at the bottom of the page and have the link scroll to the page to that longer description.
thanks,
Mike
February 21st, 2009 at 1:36 pm
@Josha we usually use google tools, like analytics and google j=keyword tool.
There is also some other tools out there that are really good.
Thanks for commenting.
February 21st, 2009 at 1:38 pm
@Mike, I’ve never heard about linking to the same page and its influence to SEO, what we would do your case is link to another page.
Thanks for commenting