Google AdWords Tips | This Time We Test The LAST Word Of The Ad

January 17th, 2008 by Mike W. Read more about Google AdWords Tips

We just recently ran an AdWords test showing how important your ads’ first word can be.

So the next logical test would be on the very last word, right?

Right.

Take a look:

boost-ctr-by-not-adding-tod

Our defined goal was click-thru-rate and the ad with the test-word removed did better… much better. Our conversion rate results are not statistically significant yet so we’re disregarding that for now.

This is interesting because generally the more specific you are, the better your ad will do.

But not this time.

And testing that assumption gave us a 50% improvement in traffic.

Not bad for another 1 word difference.

If you are interested in the other critical elements to test, find out in glorious detail right here.

About the Author

Mike is an internet marketer at MindValley responsible for running marketing tests, designing product launches, and leading an eCommerce project.

Check out other posts by Mike W.

If you liked this article, here are some related posts:

If you want to see what's in the private lab...

Ask yourself... which part of your business would you most like to improve?

I can send you 7 tactics related to your specific needs, if you like.

You don't have to buy anything, just take it as a backstage pass into our private course.

Will it give you the solution you've been waiting for?

There's only one way to find out.

6 Responses to “Google AdWords Tips | This Time We Test The LAST Word Of The Ad”

January 18th, 2008 at 7:45 pm Pay Per Click Advertising Roundup | January 15, 2008 through January 18, 2008 | Apollo SEM
[...] Google AdWords Tips | This Time We Test The LAST Word Of The Ad [...]
January 20th, 2008 at 11:06 am Van
hmm…good))
January 22nd, 2008 at 5:31 am Chris Gemignani
The difference in click through rates is about 17% not 50%.

412 / 1.5 million = .027%
4391 / 13.2 million = 032%

This is a statistically significant difference (calculations omitted) but it’s far smaller than you’re reporting.
January 22nd, 2008 at 11:53 am Average Blogger
Great stuff there, will read more about it!
January 22nd, 2008 at 10:22 pm Mike W.
Hey Chris, I rounded the impressions for ease of reading, so the difference is larger.

The key is to focus on the concept, which is testing that last word. Every case will be different, but when you feel like you’re running out of “what to test” you can use this as an idea.
June 15th, 2008 at 7:18 pm Linus
Hi guys,

When you get statistically valid results, I’d be interested to see them.

I have also found that when the number of impressions are so heavily skewed towards one ad, rarely is it a proper test.

In fact, when split testing ads that are equally distributed (50/50), the stats vary considerably depending on what date range I choose to report on! And yes, they have greater than 30 positive results so should be statistically okay.

MindValley Labs Internet Marketing Blog

Conquer the Web