Before You Launch A Facebook Ad Campaign
Facebook advertising is like dating. For example, when dating you meet someone then go on a few dates to see if the two of you are a good match. Eventually, you find someone special and worth investing more time into. Facebook advertising is similar in that at first we should test the waters, see what works, Then finally invest more resources in.
Over the years we've discovered that many business, small and large, do not do this! Instead, they put all of their marbles (dollars, time, resources) into one ad targeting a large audience and are puzzled when they don't get the results that they wanted. We recommend approaching Facebook advertising, as well as any other social media or digital advertising platform, like dating by testing multiple ad targets, creative versions, and call-to-actions.
Here are a few examples of tests you should run when advertising on Facebook for the first time.
Let's say your business is a local dance studio that specializes in Latin dancing and you want to increase registrations for dance classes.
Common approach: target everyone within 10 miles of your studio between the ages of 25-45 who like dancing.
Try this: create a few different, relevant target groups:
- People within 10 miles who like dancing. Consider this your broad audience.
- People within 10 miles who like specific latin dance types, such as salsa, bachata, merengue, etc. Consider this your highly-interested audience.
- People within 10 miles who "look similar to" current students at your studio. Consider this another highly-interested audience.
- People within 10 miles who have visited your website before (using custom audience targeting). Consider this your super-targeted audience.
Run your ads for a week or so, depending on your sales cycle, then analyze to see which of the four target groups resulted in the best return-on-investment. Once identified, pause your two lowest performing target groups and test more on the best performing groups. Perhaps now add in additional targeting filters such as those who are single vs those who are not.
Common approach: promote only one ad (or boosted post)
Try this: split test 2-3 ad versions. This way you will learn what type of ad creative results in the return on investment.
Below is a good example of a split test. It's a good test because the only difference between two ads is the image. This specific test will tell you which type of image gives you the best return on invest for each audience.
Below is an example of a bad split-test. It's bad because the image AND the body copy is different. So when you analyze the performance you won't be 100% sure if it was the image or the copy that resulted in the difference in performance.
In conclusion, remember to keep testing, learning, and optimizing toward your conversion goals.