How to Use A/B Tests for Effective Advertising
It’s possible to increase the reach of your brand using effective advertising. If you know which ads work well for your demographics, you have hit gold. Effectively written ads go a long way in bringing more eye-balls to your website or landing page, and therefore more sales.
But what if you are not sure what your audience will click and what it will reject? Well, no one is, at least in the beginning. Yet great campaigns are made.
Enter A/B Testing – a unique way to build effective advertising campaigns.
With A/B Testing you can separate ads that give a good ROI from the rest. Simply put, A/B Testing is a method where you compare two ads against each other. One ad may be different from another by a single factor such as the image, ad copy or heading.
Once you pick this differentiating factor, run the ads simultaneously and see what happens. For example, while promoting a travel booking service, you may pick one image of Tibet for ad A and that of Europe for ad B, and see which one works better. Compare them against your “goal” such as maximum number of Facebook fans or maximum sales. (But remember, often a sale would take days or weeks to close. This doesn’t mean your ad is not working – it may well be impressed on the customer’s mind. Just that he isn’t buying yet.)
At times ad A would deliver more click-throughs than ad B. But does it also lead to most number of conversions (becoming a fan, “liking” the page etc)? If not, it may not be your best ad yet. In other words, a great click-through but a poor conversion rate won’t cut it.
It is better to take other factors into consideration, such as the CPA, or Cost Per Acquisition, which takes both click-through and conversion rates into account.
How to Conduct an A/B Test
Step 1. Pick a Control Ad: Your control ad is your starting point. It will help you to create several variations by changing its one factor at a time. Remember to change just one element because if you change two, you won’t know which one worked. For example, while testing, change the image or the wording of your ad (but never both) and then compare the results against that of control ad.
Step 2. Change Your Control Ad to Generate a Variation: This ad will be compared to the control ad’s performance. If it does not perform better, it will be chucked. The test will continue with another element of the control ad changing.
Step 3. Run Your Ad: Don’t be quick to dismiss an ad if it’s not generated enough clicks, for example, in the first day. Similarly, don’t label an ad as a winner just because it got 50 more likes than its counterpart on day one.
Always run the ads simultaneously. If you run ad A on Tuesday and ad B on Wednesday, which is a national holiday, perhaps people will laze around on Facebook a lot more than usual and ad B will appear as a winner.You can only know if you run them at once and stay long enough to gather sufficient data which can represent your audience.
Step 3: Record the Results: Did the image of Tibet get more clicks than that of Europe? Did the user sign up for your newsletter after clicking the ad? What was the result of changing this one element? Record your results and decide to ignore this ad or change it a bit more to find the best variation. If Tibet wins, change the image with that of India and see how it does this time.
Once you have a better performing ad, throw the “loser” and gear more money toward the effective one. Now keep making subtle changes to the latter and run the ads side by side. Rinse and repeat.
Hopefully, this post gave you the basic idea behind A/B Testing and how it can be used to generate effective advertising, and thereby convert prospects. We’d love to hear your experience with A/B Testing in the comment below!
About Revive ProjectsRevive Projects is an award-winning Melbourne-based leading Marketing Agency with years of expertise in franchise marketing, Health & Fitness marketing, Mainstreet marketing and Distribution-based marketing. www.reviveprojects.com.au.
Photo by faith goble.
Posted by Pooja on 18th January, 2012 | Comments | Trackbacks
Tags: A/B Testing, marketing agency, Marketing agency Melbourne
The trackback URL for this page is http://www.reviveprojects.com.au/trackback?post=25400345 TrackbacksThere are no trackbacks for this post There are no comments for this post Post a CommentHTML is not allowed in comments, http://... will be automatically linked.
|
