On 8 February 2018, we organised our fourth Facebook Advertising Workshop at Uptown coworking space. We are sharing here the answers to questions that our audience asked us, so we can help our online readers as well!
You can find the Chinese version of this article here: 座談會4個主要重點 – 中文版
Background: this workshop’s goal was to provide actionable tips on audience targeting & multivariate testing. We also explained how the Clickful platform helps to achieve better performance in these key components of Facebook Advertising.
Question #1: How does Clickful platform communicate with Facebook?
By using Clickful for your advertising on Facebook & Instagram, you will not have to use the Facebook Ads Manager at all anymore! You can manage all the parameters of a Facebook campaign via our platform: ad banner, ad copies, audiences, setup, etc.
From a technical standpoint, we are plugged into the Facebook API. Our special developer-level access allows us communicating with Facebook at a high rate (above a thousand calls per second).
Question #2: What are the text limitations in banners and videos?
The rules are straightforward: no more than 20% of text in a regular banner, no more than 20% for the thumbnail text in a video, and finally no more than 20% of text in a static image within a video!
What if I do not want to / can not change my creative? Facebook will limit the reach of your ads, from low to high. You can find more information here: https://www.facebook.com/ads/tools/text_overlay
We make sure that the limitation is not affecting our users by testing directly the banners created within our banner generator!
Question #3: Which type of A/B Testing can be done with Clickful?
We are providing our users more than a simple A/B testing by taking a more efficient approach called “multivariate testing“. We are testing the following components for every Facebook campaign:
- Ad banners
- Ad copies: ad text, headline, and link description
The platform automatically sets up all the tests so that our algorithm can use them to optimise the campaigns. For example, when a user sets up 3 audiences, 3 ad banners, 3 ad texts, 1 headline, and 1 link description, the platform will create 9 variations that will be shown to 3 audiences. Then, using live data and anticipating on performance via machine learning techniques, the algorithm will mitigate the budget to the most performing variations and audiences.