Using models in story ads to improve performance

December 22, 2022

The following article consists of Confect's interpretation of actual data. We analyze real-life marketing, advertising, and sales data using Confect's proprietary AI engine. Below, we're breaking down our findings and sharing our results so you can supercharge your future sales efforts. Click here to learn more about Confect.

When it comes to advertising, using creative, eye-catching visuals is key. 

Something as simple as using colored background versus using an image with a human model can significantly impact your ads' performance. 

According to research, as humans, our brains are hard-wired to recognize faces, even in inanimate objects.

This makes one thing clear: faces are more recognizable than other objects.

Therefore, using human models in advertising should play a role in an advertising campaign's success or failure. 


But more importantly, how much impact can a live model have on your ad performance and overall ROI? 

Below, we're taking a closer look at how effective models can be in boosting story ad performance and whether or not you should use one for your next ad campaign!

Models in content improve the performance of story ads by 46%

As mentioned, faces naturally attract our attention, inciting an immediate emotional response. 

Therefore, it shouldn't be surprising that we found ads with models performed 46% better than similar ads without models.

Psychologically, when we see an object worn or being used by a model, we can instantly relate more by imagining ourselves in that same situation. 

Seeing a model use a product also makes content appear more genuine, so it will feel less like an advertisement. 

Therefore, using a model can increase the likelihood that a viewer will want to read the content, engage with it, and hopefully, want to click through and learn more. 

So yes, we found that overall, models increase ad performance.

Read more about the impact of models on different messages.

But depending on what you’re selling, the impact of using a model might not always be the same. 

In story ad sales, models in content don't have a large impact on performance

When we compared sales ads with and without models, we didn’t find much difference in performance levels.  

In sales ads, we found that using a model only improved performance by a small margin of about 2%. 

While 2% is still an improvement, advertisers need to question whether or not that 2% increase is worth the initial investment of hiring a model, paying a photographer, and setting up a photo shoot, which can quickly add up.

So, yes, using a model in a sales ad can improve performance. 

But the boost is slight, and the cost of hiring a model can sometimes outweigh the return. 

So it’s important to note that different ads for different industries, companies, products or services might obtain different results.

And that highlights the crucial importance of A/B testing your ads during your next campaign. 

Regardless, the fact is that most of us have grown tired of sales ads, period. 

We see so many ads daily that we immediately recognize ”sales-y” ads. 

And we move on without a second look. 

Using a beautiful live model isn’t going to change that! 

When launching new products, models improve story ads performance by 102%

So beautiful models can’t save sales ads from being ignored. 

But what about ads for newly launched products or services? Can a model increase performance for new product ads?

To find out, we compared new product launch ads and ads mentioning news in any form, both with and without models, and found a drastic difference in performance.

According to our data, story ad performance increased by an impressive 102% when models were used.

That’s doubling your ad performance, and just by using a model!

So if you're launching a new product or rolling out a new service, you should use a live model to interact with and display your latest offering. 

The science is there that, as human beings, seeing a live person already makes the ad feel more personal and relatable. 

So when we see a person using a new product, we can easily envision ourselves using it too. 

When it comes to products like clothing or jewelry, we even imagine what we'd look like wearing those things.   

Seeing a model use a product generates a sense of trust in the eyes and minds of the consumer. 

And that's precisely why, these days, the best advertisers are not selling aggressively or being overly sales-y in their ads. 

They often focus on showing a live model, typically a rather attractive looking male or female, using the product. 

And to most consumers, this doesn't look like it's an ad at all.

Using A Model In Your Story Ads To Boost Ad Performance

In conclusion, we found that using models in your story ads can boost performance anywhere from 46% to 102%. 

So in most cases, using a model is well worth the initial investment. 

But that's not always the case. 

If you’re promoting a sale, using even the prettiest models won’t help you ad perform any better than its competitors. 

However, for other types of ads, especially for new product launches, live models can make a world of difference! 

Therefore, using models in story ads can be one of the most effective ways to create better engagement with potential customers and ultimately increase sales.

More to read

This article focuses purely on the impact of models in story ads. Keep optimizing with the following insights:

Data section

The dataset from this article is based on 1B impressions and 13.6K 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. The content was filtered based on the aspect ratio to separate 1:1 (square) from 9:16 (vertical).

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.