Advertisers are constantly looking for new creative ideas to apply to Facebook Ads. And there’s a good reason for that: if people see your ads too many times, they’ll develop ad fatigue, resulting in higher ad costs.
If you haven’t yet heard of the latest Facebook news, here’s something you should know: Facebook now allows the use of GIFs in video ads.
We were curious to find out whether animated images really outperform still images in terms of click-through rate and cost-per-click.
So, we set up a Facebook Ads A/B test to find out what works best: ads with still or animated images?
And we discovered that… Well, now we can only tell you that the experiment’s outcome was surprising.
To find out all the dirty details, read the full article and learn the secret!
Hypothesis and Experiment Set Up
We hypothesized that using animated Facebook Ads images would result in higher click-through rates and a lower CPC. We thought that moving images would get noticed in the News Feed and attract more curiosity.
There’s plenty of research to support our hypothesis. A report by Kinetic Social showed that video ads have the lowest eCPC compared to photos and link posts.
But you can never be 100% sure. And here in AdEspresso, we love A/B testing. So we decided to run a test with our two Lead Nurturing campaigns promoting our eBook bundle. The goal of our campaigns was to get new leads in exchange of our Facebook marketing eBooks.
Campaign setup
We set up a 21-day Facebook Ads experiment to see which ad design works best: a still image or an animated picture.
To get actionable results, we resisted the urge to test too many things at once, and only kept four different ad groups:
- Ad with still image targeting a warm audience
- Ad with animated image targeting a warm audience
- Ad with still image targeting a cold audience
- Ad with animated image targeting a cold audience
As you can see, we wanted to test two things: the ad image and cold vs. warm audience.
We used a Facebook Lead Ads campaign as our goal was to get new leads and enter them to our email campaign.
If you’re curious whether to use lead ads or drive people to your landing page, read about one of our previous experiments: Landing Pages vs Lead Ads: The $2,000 Facebook Experiment.
Here’s what our lead ad with still image looked like:
Here’s an example of our lead ad with the animated image:
Do you want our most popular social marketing eBooks all in one package?
Posted by AdEspresso on Friday, January 6, 2017
As people saw our Lead ad on Facebook, they could download the eBook bundle without leaving the platform.
All they needed to do was enter their name and email address.
To set up the campaign, we used – you guessed it! – AdEspresso. It took about 30 minutes to have our campaigns up and running.
We decided to set up each ad variation as an independent campaign for unmixed campaign results and accurate reporting.
As you can see, we uploaded the animated campaign image as a video and promoted it as a video Lead Ad.
Ad placement
We only used Mobile ad placement as it has shown great results before.
Budget
Each of our four Lead Ads campaigns had the total budget of $500 and ran for 21 days. We set the ad campaign budgets to $15 per day for consistent ad delivery throughout the campaign period.
Bidding
We used automatic bidding and optimized our ads for lead generation, which means you’re paying for impressions. We decided against custom bidding as we didn’t want the bids to be affecting our A/B test’s delivery or results.
Audience
As we mentioned before, one of the campaign elements we wanted to test with this A/B test were cold vs. warm audiences.
Warm audiences include people who have previously visited your website and are familiar with your product and offer. A cold audience consists of people who haven’t interacted with your brand before.
To target warm audiences, we created Facebook Custom Audiences to deliver our lead ads to people who had shown interest in our product but hadn’t yet created a free trial.
We targeted people who had visited our website in the past 7 days, and excluded the Custom Audiences of people who had already downloaded our eBooks or started a trial.
We didn’t add any other specific targeting options to avoid narrowing the audience down too much.
As a result, our campaigns with warm audience targeting had the potential reach of 52,000 people.
For cold audiences, we targeted people living in Europe, North America, Australia, and New Zealand. As our goal was to reach marketing professionals, we targeted people interested in marketing-related Facebook Pages and brands.
We also excluded from the cold audiences the people who had already downloaded our eBooks or started a trial.
Results
Drumroll, please!
As we hinted before, the experiments’ results didn’t match our expectations and created mixed feelings.
Are you ready to see the campaign results?
Here’s the comparison of our two campaigns’ results, both targeting a warm audience:
Ads with still images had the average CPC of$0.52.
Ads with an animated image had the CPC of $0.78.
That’s a huge difference: the ad campaign with still images got ⅓ more clicks at the same budget.
In terms of CPC, ads with the still image outperformed ads with animated image by 49%.
That was not what we’d expected. We hypothesized that animated images would have higher click-through rates and lower CPC, but none were the case.
But there are more interesting stats to look at. After all, the goal of our Facebook ad campaign was to generate leads.
Ads with still images had the average cost-per-lead of $1.028.
Ads with an animated image had the cost-per-lead of $1.665.
In terms of cost-per-lead, ads with the still image outperformed ads with animated image by 38.2%.
Warm audience vs. cold audience
Another hypothesis we had when setting up this Facebook Ads experiment was that warm audiences outperform cold audiences.
Well, this hypothesis wasn’t completely right either…
The Lead Ads campaign targeting a cold audience and using a still image had the lowest cost-per-lead of all four campaigns.
Here’s the overview of all four campaigns’ cost-per-lead:
- Ad with still image targeting a warm audience: $1.028
- Ad with animated image targeting a warm audience: $1.665
- Ad with still image targeting a cold audience: $0.749
- Ad with animated image targeting a cold audience: $1.60
Regarding the cost-per-lead, the ad campaigns targeting cold audiences outperformed warm audiences by 37.2%.
However, as later we analyzed how many of these leads created free trial accounts and set up a payment account, warm audiences delivered the best results.
Additional stats:
Here are some more statistics that you might find helpful when analyzing and optimizing your Facebook campaigns:
For the total ad budget of $1,800, our ads gathered close to 250,000 ad impressions and generated over 1,500 new leads.
The average ad frequency when targeting warm audiences was 2.7. When targeting cold audiences, the average ad frequency was 2.1. All campaigns generated leads and delivered good results through the entire lifecycle.
Conclusion
Our Facebook Ads experiment clearly showed that regular still images worked better than animated images. In terms of cost-per-lead, ads with the still image outperformed ads with animated image by 38.2%.
We also discovered that it’s well worth testing both warm and cold audiences when promoting low-threat offers such as an eBook.
However, keep in mind that it was just one specific experiment.
To know what works for your company’s Facebook campaigns, why not replicate the A/B test and see the outcome for yourself?
The test is cool and I thank you for the insight, however I think that the Title is a bit misleading. Since there are plenty of goals on facebook ads I think the title should mention the goal.
I have found gifs to be very good at transmitting emotion, especially when they are in a meme form (a gif of movie character well known by the audience). I usually use them on cold audiences to make them a bit warmer :).
In your case though, although the animation is nice, it doesn’t send much emotion, furthermore I think that the reason the ctr is lower is because the animation takes the focus away from the download button.
I would bet one of my kidneys that the results would be positive (atleast to warm audiences) if your gif would be funny. I imagine your character taking Mario’s place in this video (https://www.youtube.com/watch?v=2iNKPkTOr5k ) and instead of the super mushroom he eats your social media bundle. The caption could be something like “Afraid of Goombas? Grow your social marketing skills with our Social Marketing Bundle” (maybe a bit too specific but you get the point).
Anyhow, huge comment tldr.
Hey Andrei,
The comment was long, but totally worth reading! 🙌
I agree with your point about using more engaging animations, that’s definitely something to experiment in the future.
I’m not 100% sure about the cold audience though, Facebook + cold audiences (usually) = 💘. I take it would make more sense to advertise to someone who’s been to the website at least once.
One of the goals of this test was to test whether a moving image, no matter how “boring”, would catch people’s attention and lead to higher impression-to-click rates. We were quite surprised it didn’t work better than static images. Maybe you’re right about the fact that the animation stole attention from the CTA button.
Love that you called these questions to our attention. Great discussion!
Have you tested more traditional video against still images? I agree, the creative on the animation may have caused the difference… fascinating though. Sometimes things on FB seem like they will work and the oddest thing does way better.
Yes!! to the part of the oddest thing working best.
We tested a 1.5-minute product video for our SaaS startup Scoro, and these ads converted quite well.
Hey Karola!
Thanks for the reply! I wanted to clarify the cold audience part so here it goes:
There are some businesses that are on the opposite side of a love product, lets call them hate services XD. I’m talking here about an insurance broker, people hate the commercials that they see on TV, always about dying or accidents, there areplenty of cold calls still in this sector (offering you life insurance while they make you think about your possible death). Tldr people hate being sold insurances but THEY ALWAYS need insurances. For this reason I’ve started to make some brand awareness (engagement) ads with funny gifs that regard specific insurances, people like them and share them and remember the brand. Since you cant really make demand generation for insurance products I find that you must have a nice brand with much awareness so when they need the services they will think about you. For the insurance sector a cold audience is ice cold, people don’t empathise, you make the audience a bit warmer with subject related funny gifs and then you comeback with some articles and send them to your website.
I totally agree with you Andrei.
I’d even say this kind of cold-to-warm marketing funnel makes sense for both “loved” and “hated” services. 😉
The video ad added nothing to the conversation, it was just annoying. A bit like early websites with their slot machine type graphics.
When creating any kind of animation, you need to consider – what is the intention? And this has to go beyond “catching attention”.
Like Andrei said, if you are conveying emotion, or delivering an explanation, it would likely convert much better.
Having said that, great test to bust the myth that “animation converts better”
Thanks for your input, Kim!
I agree that the video wasn’t super fun an engaging. Nevertheless, the point of this experiment was to test whether animation works better IF it’s similar to the static ad.
Had we used a GIF or some other kind of fun video, the test results wouldn’t have reflected on the “animation vs. still image” question. They’d shown whether “fun video ads vs. traditional static ads” work best.
Thanks everyone for this conversation here. It actually gave us a new idea to experiment with more engaging videos in the future. 🙌
In my opinion the animated video production took more cost than still images but they are more effective and give more good CPC rate
Thx for weighing in your opinion Jeff!
I completely agree with you, especially if the animation is engaging and interesting to the target audience.
As there are a number of social media channels, therefore, there are really maximum chances of your explainer video to get viral in a more lesser time than ever.
This is interesting. I’m working a lot with photography, and there too I’m observing similar effects.
For one campaign I had played with animations from Animated Traffic * and ran a comparison with some photos from Unsplash **. Although the sample was small, the animations drew more attention and converted better.
* https://animatedtraffic.tumblr.com/
** https://unsplash.com/
Nice Blog! Thanks for sharing very useful post. keep it up.
Very interesting study, please do a study comparing the Videos (autoplay) vs. Static images
Karola—
Do you want to venture a guess as to WHY still is better than animated? I have my theory, but I would like to hear yours first.
Thanks for the article.
This was an interesting test, to be sure. I actually think that the age of the target demographic comes into play. From a Baby Boomer perspective, I find animated gifs and video autoplaying on my screen is the fastest way to get me to leave and stay away from that site. I just find them obnoxious, useless, and annoying.
X-geners and Millenials who find them entertaining would probably be okay with them, as they are just used to tuning them out.
I have never (and will never) clicked on a moving image for any reason. In fact if a site has video or animation on the web page, I will leave immediately unless I really want to see the information I came to see. I will, however, scroll away and hide the obnoxious crap to get to the information I came to site to see. If I can’t get away from it, or it’s just too annoying, I leave and put it on my blacklisted sites listing.