3 Problems with Facebook Advertising


I have been advertising on Facebook for West Boca News and for other projects. Facebook advertising shows substantial promise, and is it far better than it was in 2010.

With that said, here are three problems that are frustrating as an advertiser.

1. Poor Version Optimization
Up at the top of this article is a message Facebook shows on my home screen, recommending that I try multiple versions of ads. I do. Facebook’s advertising system fails to run effective comparisons of different versions. Below is a report on two different types of ads I’m running now, each with two versions:
The first ad is for sponsored stories, which show up in the news feed. One version has been shown over 5,000 times, while the other has barely been shown 100 times.  The second is for ads that appear on the side of the user’s window. The first version of that one has been shown over 6,000 times. In the same period the other version has been shown less than 100 times.

Google’s AdWords interface is far more sophisticated. An advertiser can let Google optimize, or the advertiser can choose to run the ads in equal proportion. With equal proportions, the advertiser can figure out for himself which ads are more effective. If Facebook allows this, it’s not easy to find out how to do it.

To work around this, I pause the “dominant” ad version (the one Facebook shows heavily) which then effectively forces it to start showing the disfavored version. This is not ideal. To effectively test multiple versions of the same ad, the advertiser should be able to run them in roughly equal proportions at the same time.

2. Targeting Inconsistencies
During the process of setting up an ad, Facebook has an interface that allows you to choose who you are targeting. This works well most of the time and in the long run will be one of the great strengths for the company. But its implementation is spotty.

I’ve done some ads where the same targeting results in dramatically different audience sizes. On the right you can see that one of my ads reportedly targets only 20 people.

I have an ad with very similar targeting that Facebook says targets over 12,000 people. It’s not a perfect example because the first ad is a sponsored story and the second is a regular ad. But the sponsored story should reach far more than 20 people under the circumstances. I’ve had other experiences with identical ad types and targeting options that led to bizarrely different numbers.

3. Weak Customized Reporting

As you can see above, the options for viewing reports of advertising numbers are very limited. The user has a choice of a 1-day, 7-day, or 28-day period, running back from a chosen date. Google’s AdWords is a far more robust interface allowing advertisers much greater flexibility. One could argue that it’s not fair to compare to Google, which has been around much longer. But the AdWords interface was better that this back in 2006, and Facebook has the advantage of being able to see how Google blazed the trail ahead of them.

Despite these concerns, the effectiveness of advertising on Facebook is compelling. They will make this experience better for advertisers like me, and that will lead to dramatically increased revenue for the company over time. I’m so confident of it that I recently bought stock in the company.

About Warren Redlich

A co-founder of West Boca News, Warren is a resident of West Boca Raton. He is a criminal defense and personal injury attorney in both Florida and New York. He can be reached by e-mail at wredlich@gmail.com or by phone at 888-RED-LAWYER (733-5299).
This entry was posted in Business. Bookmark the permalink.

2 Responses to 3 Problems with Facebook Advertising

  1. Kristina says:

    Hi Warren, great job running Facebook ads – they can be really effective. And thanks for stopping by the AdEspresso.com blog and pointing us to your post.

    I think maybe I can help clarify a few things to help you further improve your campaigns.

    1. Version Optimization

    When you create multiple versions of an ad inside the same campaign Facebook automatically a/b tests them against each other. As one version begins winning over the other version, Facebook automatically stops showing the poorer performing version. This is to help you get the best results. So, I would say that you should not pause the better performing ad as this is probably hurting your results.

    2. Targeting Inconsistencies

    The reason why some ads have a much greater reach than other ads even with the same targeting is because they are two different type of ads. Sponsored Stories are ads where it says “your friend likes this page/checked in at this place.” That is why they will always tend to have a smaller reach then ads that can reach anyone regardless of whether they have a friend who has interacted with your page or not.

    3. Customized Report

    You can select any time frame you want for ad reporting. Simply go to the area in your ads dashboard where it says Lifetime Stats and it will open a dropdown that lets you choose from: Today, Yesterday, Last Week or Custom. When you select Custom Facebook will allow you to select from any dates you want.

    I really hope this information helps and if you have any other questions feel free to reach out to me at kristina@adespresso.com. Thanks!

  2. @kristina:
    1. This assumes that FB is doing such split testing and doing it well. When I add a new version it generally does not get displayed. Split testing doesn’t work if you don’t actually test.

    2. Yes, I agree, but that does not explain how our page, with around 20 Likes (for now), only targets 20 people. So each of our Likes has an average of one friend with the same interests? When there are over 12,000 people with such Likes in the county? Seems questionable. And that also does not explain my experiences when the exact same ad type and interests showed radically different targeting reach.

    3. The custom option is nowhere near as flexible as the one on AdWords.

Leave a Reply