Facebook Advertising: Testing, Segmenting, and Optimizing

Facebook Advertising Testing Segmenting and Optimizing 214x300 Facebook Advertising: Testing, Segmenting, and OptimizingThe following is a guest commentary from Marc Poirier, Co-Founder and CMO of Acquisio.

Facebook is becoming an increasingly competitive space, making it difficult for advertisers to just get by on the very basics. You can put your ads out there, but without some extra attention to details, it is incredibly difficult to reach your audience and convince them to find out more.

In addition, Facebook is now a massive gathering place for all things consumer-related, which makes it a critical advertising channel. With these three tips, you’ll be able to refine your Facebook marketing strategy to ensure that you stand out and get your message to the people who matter most: your potential customers.

Multivariate Testing

Many marketers engage in A/B testing — it’s a standard practice and is the very least that one can do as a marketer to ensure that one is getting the best out of their ads. After all, it’s the audience that decides which ad is the most successful. One might actually be surprised by results, but in the end, it’s about your audience.

With multivariate testing, you’re taking this several steps further. You will look at headlines, images, offers, prices, colors, forms, and more. You’ll create multiple versions of your ads with different headlines in conjunction with different images and placements. This allows you to drill down deep and find the very best results based on several criteria, as opposed to just two.

These test can be performed all the way from the ad to the landing page with multiple elements of each. For the best results, it is necessary to be able to generate enough traffic to give you a good benchmark from which to glean results.

Testing with very little traffic can be problematic, so you’ll need to have an idea of the traffic you might get in order to establish the number of variables you might test. The lower the traffic, the fewer the variables you can get honest results from in a test. The more variables you have, the longer it will take to gather usable results.

Full-factorial multivariate testing can aid in this and can be done with just three ads with three different combinations of three elements; in this case they are the Headline, Ad Text and Image. You’ll end up with nine combinations, allowing you to distribute daily budget to each of them evenly. In the end, you get a full set of data that allows you to use the highest-performing element from each of the variations to create the best ad from each variable.

Nuanced Segmenting

There are a lot of targeting options provided by Facebook, allowing you to be very nuanced in your advertising options. Those who are successful with Facebook advertising are those who have not put all of their eggs into one basket.

Being too broad with Facebook advertising is dangerous for more than one reason, not the least of which is the possibility that if you target too large a group and no one is interested in your ad, your CTR will be so low that Facebook will actually cancel your ad.

Being more granular with your ad targeting gives you greater control over your efforts, and you get much higher performance as a result. And being granular requires a lot of different targets. This may mean creating thousands of ads in order to target age groups, regions, and every permutation of these things in order to exercise complete control over your CTR.

If you cannot control your CTR, one ad could be the undoing of an entire campaign. It is crucial to a campaign’s success to monitor ads very closely, as campaigns can unravel due to the poor performance of one ad in minutes, but with more precise targeting, you’re giving yourself more room to breathe.

Start by aligning Interests, Broad Categories broken down further, and Connections and create different combinations of these. From there you can further break down your demographics and geographics. Be reasonable in your efforts, but be aware that true success with Facebook advertising requires thousands, not dozens, of ads.

Obviously, a third-party application that allows you to create this many ads quickly will be a huge part of this endeavor, otherwise you will spend more time creating ads than advertising.

Ad Optimization

The process of testing ads does not stop when you have found one that works best. It is still necessary to find out if you can do better. For example, you may have performed a multivariate test and found a winning combination. This does not mean that the testing ends there. In fact, it’s quite the opposite. Testing and optimizing is a constant process that requires complete dedication to finding out the answer to the question, “What works best for my audience?”

You may find that ads that mention your brand do not work as well as ads that do not. What you find out may be completely counterintuitive and utterly baffling. But, to get the most out of this advertising process you must be ever vigilant and constantly evolving and further refining your ad strategy.

Facebook advertising is not something that one undertakes lightly. In order to really get your message spread over a large group, it is necessary to put in a great deal of effort, but with that effort can come great rewards. Using these processes to refine your ads, your audience, and your delivery will allow you to make the most out of this potentially very profitable marketing channel.

About The Author

Marc Poirier is Co-Founder and CMO of Acquisio, and one of the leading voices of the performance marketing industry. Marc started his internet marketing career in 1996 while studying for his PhD in Cognitive Science, and lives near Montreal where he enjoys fishing in Canadian lakes and rivers on weekends.

 

This post was written by:

- who has written 6031 posts on Mobile Marketing Watch.


Contact the author

0 comments