Image titles and SEO

[JazyLcan]

Hi,
Can any of you tell me if adding titles to my images with keywords in there will help my SEO efforts?

Eg. Web hosting site
<a href="/" title="Web Hosting">Logo Image</a>

Thanks,
JL

[BigAlReturns]

In my opinion this would give you some SEO benefit, albeit a very small one. The problem with link titles is the same as we see with things like meta tags, i.e. they are not visible to the standard end user, or at least not without some effort on their part. As such, the webmaster can insert whatever they like into them, without affecting the look/usability/readability of their site, leaving them wide open to abuse.
However, they can be useful to use, as they will give some semantic meaning to otherwise meaningless links. By meaningless I mean that SEs can determine nothing about the linked-to pages content from an image. Adding a title gives some information, albeit info that can easily be manipulated.
BUT
There is a very good reason why you should always use appropriate titles (and alt tags for that matter) for your image links. Users who use screen readers cannot obviously see the images, and will be left without a clue what they link to. This is especially important if the image in concern is a link. Title and alt tags are a simple easy way to massively increase the accessibility of your site, and as such you should always use them, even if the SEO benefit is going to be small.

[JazyLcan]

Thanks BigAlReturns,
That helps a lot. I will try not to abuse not neglect them keeping my keywords as relevant and straight to the point as possible.

I would love to actually see some kinda of test done on this to see if a page with a certain rank for a specific keyword actually changes rank if these are added. Maybe one day I will try and set something up.

Thanks again.

Cheers,
JL

[BigAlReturns]

It would be possible to test, but not by looking at changing ranks. The problem here is that there are so many (hundreds at least) of factors affecting rank, and it is very difficult to isolate one factor when looking at changing ranks, if not impossible.
I would propose to test the theory this way:

1) Think of a new keyword that yields zero results on Google, a random string of letters should do it. E.g. shfbwuigbv Name it something unrelated like testerpage1.html

2) Create a new page with some junk content, but it must not contain your random new keyword, this is vital.

3) Include the following code on one of your reasonably well ranked pages:
<a href="testerpage1.html" title="shfbwuigbv"><img src="img.jpg" /></a>

4) Wait for testerpage1.html to be indexed - should be reasonably quick if the page you're using to link to it is well ranked.

5) Search Google for shfbwuigbv

Now the only single place that "shfbwuigbv" is mentioned is in the title of the anchor tag - nowhere else in Google's index does that phrase appear, nowhere on testerpage1.html does it occur. Therefore, if searching Google for shfbwuigbv brings up testerpage1.html as a result, we can categorically state that Google is using the title tag from the anchor as a means of determining the content of testerpage1. From this we deduce (although we can't be certain it will hold true in all situations), that the title of an anchor tag is taken to bear some relevance, and therefore some SEO benefit, on the page being linked to.
Of course, we must also be prepared to accept the contrary: if after a reasonable amount of time testerpage1 does not appear in the results for shfbwuigbv then the title tag cannot be conferring relevance onto the linked to page.
We can also look in other SEs to see if testerpage1 appears in their results for the new word, and use the same logic to determine if <a title attributes make any difference in their algorithms.

p.s. shfbwuigbv shouldn't be used for testing now - make sure it's a brand new unique word with no results in Google.