What to do when your Facebook ads are not performing as expected
To begin with, it is always good to keep in mind that a possible reason for poor performance in your ads may be too high expectations. That is why expecting realistic results is part of understanding the standards and metrics of your industry, segment and type of conversion funnel.
There are also a variety of reasons why your Facebook ads can fail, which can be divided into two large groups: Technical errors and errors in creativity.
Technical errors can include situations such as incorrect installation of the Facebook pixel, poor campaign structure, and incorrect budget selection. Creativity flaws are usually in copy, headlines, images, and landing pages.
To fix creative errors, you need to identify whether they are found on your landing pages, in your ads, or in both places. Finally test and optimize the elements of your messages so that you have the basics to start working.
How to test and measure messaging elements in your Facebook campaigns
To determine if you are making mistakes in your Facebook campaigns look at your link CTR (Link Click-Through Rate) and not the total CTR. Total CTR identifies all clicks made by users anywhere in your ad. On the other hand, the link CTR indicates specifically how many people clicked on your landing page.
A good reference for link CTR (Link Click-Through Rate) is between 1% and 2%, which means that if 1000 people saw your ad, 10 or 20 people should click on your landing. If your metric is lower than 1%, it is likely that message targeting is disabled.
To change the focus of your ads you must test the text or copy, the images and the headlines, in that order and one variable at a time.
In many cases, the length of the copy is one of the great factors by which the performance of an ad can vary. However, this variable can change for each particular case. Some brands do very well with super long copy and others with clear and concise text ads. To find out what works best for you, you should try three copy lengths:
- Short: a few sentences
- Medium: Short paragraphs
- Length: As long as necessary to tell your story, whether it is 1,000 words, 2,000 words, or 3 paragraphs.
Try all three text length variations with the same graphic elements. If you have a small business you can run your ads for 2 or 3 days until you reach a thousand users or get about 7 conversations. If you have a larger company, you should probably allocate between 5 to 7 days to do your tests.
Once you've determined the length of copy that gives you the best link CTR, you can start testing your visual assets. For this I recommend that you work with a minimum of 3 to 5 images. I recommend combining static content with or without text, both illustrations and photos, videos and GIFs.
In the same way, you should test the piece that works best with the copy that has performed better in the previous stage, to determine the combination that gives you the best performance.
You can then move on to the headline test stage. At this point you should keep in mind that the most creative headline is not always the one with the best performance. On the contrary, sometimes a headline that contains a direct message may be the best option. For example, a message as simple as "Download now" or "Sign up" may work better in your ads. That is why in this case I recommend you try between 3 to 5 types of messages to communicate.
Run each headline or call to action with the graphic element and the text that you have determined in the previous points to obtain the best combination, for a good link CTR.
How to test and improve elements of your landing pages
To determine if you are making a mistake on your landing pages or your ads, you need to look at your conversion rate.
If your landing is getting emails for a training, downloading content or getting a promotion, your conversion pass should be from 30% to 40%. If instead you are looking to generate user queries, your conversion rate may be between 5% and 15% depending on your industry. For example, the conversion rate for the landing of an online store is usually 5%, however this indicator varies according to the type of industry and the price of the product it offers. If your conversion rate is below these values, I recommend you review the points previously seen.
In the same way, I recommend you test your pricing strategy. If your landing offers an incentive, try what works better, if you offer a 20% discount or show the final discount in money, for example “Save $20”. For some cases it may be clearer to have a $20 discount than a 20%.
How to use Facebook's dynamic ad optimization
Once you've tested all of your messages and landing page elements, you can improve the performance of your ads with dynamic content optimization. A function that you can find when setting up your campaigns in the Ad Set section.
By using this feature you can upload up to 10 visuals, 5 copy variations, and 5 headlines and then Facebook will automatically mix the various options, until it finds the best match for your audience.
For example, if a user tends to interact more with video content, the ad platform will show them that version of the ad that has a video. In the same way, if a user clicks on ads with overlapping images, Facebook will show them this type of content.
To have a good mix of elements, I recommend that you use a list like the one I show you below, with the best content from your tests:
- 5 still images, 3 videos, 2 GIFs
- 2 variations of copy
- 2 headline variations
- 2 button variations with a call to action.
Dynamic ads may take a while to generate results, however with a little time they can perform better than other types of ads. That said, if you don't have a large budget, I recommend that you be careful with this type of piece creation.
Conclution
Having winning ads requires understanding your metrics and expecting realistic results. Also, to know where you have room for improvement, it is advisable to test one variable at a time, in order to determine which combination of elements works best for your brand and industry.
I recommend that your tests be as exhaustive as possible so that you can understand which ones to keep in future communications and which ones not. Only then can you generate campaigns that perform well over time.
