Is it a good idea to use models in remarketing content?
May 7, 2022
Using models in remarketing campaigns seems like the perfect combination for success.
In general, using models and faces in content can improve performance because they catch people's eye and connect product and consumer on an emotional level.
Remarketing to audiences is a great strategy because they’re already familiar with your brand.
But is it a good idea to use models & faces with a remarketing campaign?
Seems like a no-brainer, but does the data agree on it ?
Will using models & faces in your remarketing content be effective in bringing back customers that have visited your site but not made a purchase?
Let’s explore the use of models and faces in remarketing content to see if it's a good idea!
Faces in the content typically increase performance by 40%
In general, it’s usually agreed upon that using models and faces in your content is a good idea and helps boost performance.
Research has shown that human faces are the first thing people notice when they look at images.
This is based on human biology, which states that faces are the first and fastest thing that humans process when looking at images.
This means that in your content, it’s a good idea to let the face do the attention grabbing, and allow the rest of the ad content to do the actual selling.
This means that content with models and their faces is more noticeable, which can lead to an average increase in performance of 40% - when looking at all audiences of all types and categories of business.
40% increase in performance sounds great, but can the same be said when running remarketing campaigns?
Let’s look a little deeper in it.
Faces in remarketing content decrease performance by 31%
In remarketing content, the opposite seems to be true.
We found that faces actually decrease performance by 31%!
This is very surprising because we usually think of faces as a way to get people's attention.
To understand why this might be happening, we need to think about the stages of the marketing process.
Let’s take a typical marketing funnel consisting of Awareness, Interest, Consideration, and Action.
Someone who is being targeted through remarketing will probably be in the consideration state as they are thinking about different options.
You already have the customer's attention in this case and they already know about your product.
This is where you can truly make the product shine by showing what the USP (unique selling proposition) is, whether it’s price, discounts, shipping, quality, etc.
When a customer is making a decision whether or not to buy, they need information to help them make a transaction.
Keep in mind that this is only one way to explain what is happening.
You might want to do some thinking yourself as to why faces in remarketing content decrease performance.
It’s also important to note that remarketing isn’t all the same.
There are many different types of remarketing messages, and showing faces/models affects them all differently.
This might be different for some message types in remarketing
Even though in general faces in remarketing content decrease performance, there are still some cases where they can improve conversions.
These messages are trying to get customers' attention and create curiosity about the offer, which is what faces are good at doing.
However, faces don't seem to be as effective in Social proof and Shipping messages.
This is a scenario where showing a face and any USPs is probably not a good combination.
You are most likely just overloading customers with too much information.
In fact, they cause a drop in performance of 54% for social proof and 86% for shipping!
One possible reason is that these types of remarketing content include specific information that customers need to make a decision, so faces may not be that helpful.
Also, using faces in these types of content might take up space that could otherwise be used to give important details about the offer or product.
These types of messages, especially sales and new, are more focused on grabbing the attention of consumers.
It creates curiosity about the offer and the general goal is to create a somewhat of an “impulse buy”, which as we mentioned, is the function of utilizing faces in content.
Overall, using faces affects different messaging types in remarketing.
Test your own remarketing content
Even though in general it helps to use models and faces in your content.
But for remarketing content, it might not always be a good idea.
While faces are usually effective at increasing performance in content, they seem to have the opposite effect in remarketing.
This could be because of the stage of the buying process that customers are in and the type of message being presented.
While faces can be effective at getting attention and creating curiosity in certain types of messages, they may not be as helpful in others where the goal is to provide specific information to help customers make a decision.
When deciding whether to use models and faces in remarketing content, it's important to consider the message being conveyed and the stage of the buying process that customers are in.
And above all else, remember the importance of conducting A/B tests on your specific situation to determine the effect of using faces in your remarketing efforts!
More to read
- The key differences of remarketing ads to be aware of
- How much text to include in remarketing content for best performance?
- Why you should use a sales message in remarketing content
- Should you show the price in remarketing content?
- Including a logo in your remarketing content - Yes or No?
The dataset from this article is based on 215M impressions and 6K pieces of individual content on Meta (Facebook) platforms. It takes into account data from all countries and industries, with the campaign objective of Conversions. The period is from 1. September 2021 - 16. September 2022.
Numbers are looking at correlation only, not causation. Remember to check your own data: numbers for different brands, industries, and contexts will vary.
If you were intrigued about a specific insight, you can go in-depth into the various dimensions and how data can be segmented in Confect, here.