Does Alt Tag Spam Work?

I have recently taken over a site that features some serious alt tag spam. I was wondering if Alt Tag spam really works, so I thought this would be an interesting opportunity to find out.

Update 23 September - find out what removing alt spam does to my rankings...

I suspect that Alt tag spam does work. Google often surprises me at how unsophisticated their algorithm is at detecting some basic things, such as Session IDs in search results. So I won't be surprised is it turns out that alt tag spam actually improves rankings.

What is Alt Tag Spam?

All images on a page should have an alt tag, which is used to describe the image to a blind person, or someone with images disabled. Used properly, alt tags are considered to be good for accessibility.
However, it soon became apparent that Alt tags (or more correctly alt attributes) were used by search engines, and spammers realised that they could load extra text into alt tags that couldn't be seen by a visitor. So now you could design a page that looked clean to a person, and pack all your keywords into the alt tags where no-one could find them.

This example

The site in question was a classic case of alt attribute spam. There were approx 40 images named "spacer.gif" on the page. spacer.gif is a 1x1 pixel white image (on a white background), so it's not visible to a person. Each of these 40 spacer images had the same alt text, the phrase that the page is so obviously targeting.
I think most people would agree that these images offer no value to the page whatsoever, and can only be there to trick search engines into giving the page a better ranking. For the record, if you do have to use a spacer image (though CSS is a far better way to add padding), you should simply leave the alt tag empty.


The site in question currently ranks at between 9 and 12 (depending on datacenter) on Google NZ for the phrase in question. I'm not willing to disclose the site or phrase sorry. It's what I would call a phrase with medium competetion.

The changes

Alt tag spam isn't my thing baby. I'm guessing it does work, but I want this site to be around in the long term. So I have removed the alt tag spam and instead rewritten the page using modern css techniques. The titles and meta tags remain the same, but I have increased the keyword density in the opening paragraph of the page, and included the phrase a couple of extra times on the page. I did this by adding a couple of extra paragraphs of content to the page, and it doesn't look spammy to a person.
A more controlled experiment would involve simply removing the alt tags and nothing else, but I got bigger plans for this site.

What next?

It's now a matter of sitting back and waiting. Now that the alt tag spam is gone, and the keyword density on the page is a bit more normal looking, what will happen to the rankings for this phrase?
Will Google appreciate my new standards based layout, and reward me with good rankings? Or will Google blindly hate me for removing the keyword richness that it used to have?

If I was writing a search engine algo, I would probably ignore any alt text belonging to a 1x1 pixel image of any kind, especially one called "spacer.gif". And especially when the same alt text is used 40 times on the page!

I'm going to take a gamble and say that the page will drop to around position 20 when it's next cached. I think Google is dumb enough to reward alt tag spamming, but I would love to be proved wrong. I'll keep you all posted - feel free to post your guess as to where the site lands in the comments below.

Update: What happened a week later

Now that Google has cached the new version of the homepage, it's time to go check the rankings for this phrase. Looking at the site from a couple of datacentres, it seems the page now ranks betweek 8 and 10 for the phrase. So a minor improvement made, and certainly not the drop in rankings I had expected. I have built a few one-way links to the site in the past week, but I wouldn't have expected Google to take any notice of those just yet.


This is hardly a controlled experiment, so I wouldn't draw any serious conclusions from this. However, given that we all know alt tag spam is a pretty rubbish and black hat way of optimizing your site, it's reassuring to see that cleaning up this site offered a small benefit in terms of rankings.
Digg StumbleUpon technorati blinklist furl reddit sphinn