In my previous article How”>http://www.bukisa.com/articles/306062_how-to-create-incredibly-effective-internet-ads”>How to Create Incredibly Effective Internet Ads! We discussed the basics to:
– Grab the reader’s attention
– Spark some interest
– Create some desirability
– Get the reader to take action
Today, we will be focusing on ad effectiveness via a simple measurable approach. It really does not matter how much you like a particular ad unit, what really matters is how your intended audience is responding. Often times, very slight and hardly noticeable changes can significantly impact your click through rates (CTR), positively or negatively. The simple method to finding this out is called split testing.
Most advertising networks allow you to run 2 or more ad units simultaneously. Split testing means creating your basic ad unit and a very similar alternate. Then, after some amount of activity, one ad unit will undoubtedly perform better than the other. At this point, all you have to do is delete the poorer performer and create a small variation of the winner for yet another round of testing.
For banner ads, simply adding or removing the border can make a huge difference. Any word changed in text ads should also provide different results. Split testing is not new, but it is extremely effective and most people don’t do it. When running a Cost per Impression (CPM) campaign, just a 0.0001% in CTR improvement can dramatically increase your traffic for the same cost. The correct split testing procedure is as follows:
1. Create your ad unit normally, and then create a slight variation.
The new, variation ad should include a subtle change. Although you can compare two entirely different ads, you really won’t know exactly why one is outperforming the other. A slight variation will give you the marketing research you’re looking for since you know what the exact difference among the ad units being compared is.
2. Test the two ad units long enough to know the difference is significative.
It is tempting sometimes to disregard a slightly worse performing ad too quickly. A sufficient amount of impressions need to be considered. As a good rule of thumb, try to compare once 50 or so clicks have been obtained
from both ad units. Some people are always in a hurry and will want split testing to occur instantly.
3. Compare only TWO ads at a time.
I have seen marketers split test three or more ads simultaneously. The correct number of ads to test is TWO. The basic idea behind split testing is to find an ad variation that will beat the current running ad. Running more than two ads may tell which ad performs better, but you will not know why!
4. Always split test, even when running a proven ad.
In advertising there is no such thing as perfection. Every ad, no matter how good and for how long can always be improved. Split test and try to beat your most effective ads.
5. Once you’ve discovered a better ad copy through split testing, go back to step 1.
Split testing is an ongoing trial and error type of procedure. The benefit of creating ad copy that beats the original is improved CTR. This means more traffic for the same budget. Sometimes it could mean more traffic for less money since many ad networks reward higher CTR campaigns.
If your ad looks good and sounds well enough to you, split testing allows you to see if your public agrees. I’ve had ads perform better in the past by simply misspelling a word on purpose. Go figure!