Which Instagram Ads Are Better? Results From Our Experiment

If you’re an experienced social media marketer, you’ve probably already asked yourself this very question many times. But few have tried and tested every option with the goal to uncover the answer.

And that’s likely the reason why you’re reading this!

Recently, I decided to run an experiment and see which Instagram ads available perform better. I utilized all of the ad format options available on Instagram to test their efficiency and performance. Today, I’m here to share the results so that you can choose your ads wisely in the future.

So without further ado, allow me to tell you which Instagram ads are better, and why.

If you’ve already taken the plunge and run ads for your business on social, you’ll know that it can be a bit of a minefield. Especially on Facebook and Instagram.

Why? Because there are so many options available to us social media marketers, that it’s difficult to know which way is up when designing an ad campaign.

So, it’s definitely a good idea to understand the potential of each type of ad, rather than blindly throwing money at your computer and hoping for a good result.

Why ads are important

Let’s start at the beginning. Are Instagram ads even THAT important?

In the great scheme of things: no. Of course not.

But if you’re a small business owner hoping to grow your brand visibility: absolutely, yes.

In fact, they’re essential.

Paying for your content to get seen by the audience of your choice is not to be forsaken. OK — you have to pay for it. But the results you get will be of a much higher quality than those you get organically. That’s pretty much guaranteed.

Here’s a list of some of the main perks of paying for advertising on Instagram:

• Targeted audience: you only reach the people you want to reach
• Designated budget: you can limit the amount you spend daily, as well as cap your global expenditure
• Minimal pressure: you can pause or cancel a campaign at any time if you notice it’s not getting the results you want
• Quality results: with time and testing your results will be the fruit of your labor!

Being a social media marketer is no walk in the park. I don’t think anyone reading this will disagree with that! Instagram ads especially take up quite a bit of energy, because they need to be thought through individually — from the creative and the copy to the targeting, the objective, and the budget.

And that’s not even taking into account the fact that you’re potentially also advertising on other social platforms, as well as keeping up with all your usual workday tasks. It can be tricky to know if you’re making sound advertising decisions when you’re trying to keep all the plates spinning in other areas too.

So let’s take things slow.

If you’ve never created an Instagram ad before and want a simple guide to help you get to grips with the whole process, you can start off by reading our easy 7-step tutorial.

Our Instagram ad experiment: Which ads are better?

1. Protocol

Before even thinking about who to target or the budget to attribute to this test campaign, I first needed to decide what to promote, how to promote it, and which formats to use.

Here are the main points I had to be sure of in order for the test to be fair and objective:

• All content promoted should be identical
• All copy and creative should be as similar as possible: given that the formats are different, the creative cannot be identical
• The objective for the different ads within the campaign should be coherent: ‘link clicks’ or ‘landing page views’ — not ‘new followers’, for example
• The duration of each ad within the campaign should be equal
• The audience targeted for each ad within the campaign should be identical
• The budget used for each ad within the campaign should be identical

The ad formats tested were the six up for grabs on Instagram:

• Single image ad
• Carousel ad
• Video ad
• Collection ad
• Stories ad
• Boosted post

Another important factor for tracking performance was the use of unique UTM source codes on the CTA link used for each of the different ad formats. Basically, I planned in advance to add for example for the video ad format: ?utm_source=instagram_video_ad in order to easily track its results in Google Analytics.

2. Content and copy:

For the purpose of this test, I felt that the simplest and most coherent type of content would be one of our most recent blog posts at the time. Promoting a blog post is a great idea because it offers the advertiser a lot of creative freedom when it comes to crafting the copy, in order to entice the reader to click on the ad to read more.

The article in question was this one. A click-worthy piece if we do say so ourselves…! 😉

The next step was to write the caption which would send our audience to the landing page. This was a crucial part of the ad creation process, so I knew I had to get it right. I settled on:

“Our research showed that, across all industries, the average Engagement Rate on Instagram is 4.7% — So, how does yours compare?”

Exactly the same copy was used throughout the campaign, on all ads.

3. Creative:

The creative for the campaign was relatively easy to source, as our design team is used to producing images and videos for Instagram on a regular basis.

If you don’t have a dedicated designer, I’d recommend getting creative with a design app such as Canva or similar. The benefit of doing it this way, is that you’d only have to create one or two images and reuse them (with different dimensions) for each format of ad you want to test.

For the Single image ad, I went for this camera lens, spruced up with our Iconosquare colored shapes for brand identity. This image would become the central defining element of this campaign.

This same image would also become the cover for the Carousel ad; the following images being much simpler, with plain backgrounds and layouts to inspire clicks. The figures featured in these images are simply the stats covered within the article I was promoting, so all very relevant to our prospects.

Initially stumped where the Video ad (in-feed) was concerned, I eventually decided that — in order to keep in sync with the protocol, stating that creative should be as similar as possible across all formats — I’d use a gif-style video, using the same 3 images as in the carousel.

Next up was the Collection ad, which was a little more demanding in terms of effort and creative, as it asks for a cover image (or video) and at least 4 other images. To keep in line with the protocol, I settled on these images for the Collection ad:

The Stories ad was also created using the same images, but in a different format, and with the addition of some text to motivate watchers to swipe up for the content. The Story ended up looking like this:

Finally, the Boosted post (which couldn’t be created through Facebook Ads Manager, but directly through the Instagram app) was simply a promotion of the initial Instagram post which was created when the blog article was first published. The imagery was therefore slightly different, but nevertheless in concordance with the protocol. Here’s the creative from the carousel post which was boosted for the campaign:

A lot more information and detail was present in this original carousel post, therefore I was aware that the results may differ more for the Boosted post when compared to the other formats. It was a case of wait and see!

With all of the creative and copy settled upon, I could move onto the next step which was to build a custom audience for targeting the campaign.

4. Audience:

I decided for this test to target the following audience:

So, to recap, that’s 18-65 year olds, located in the US, who are interested in:

• Content marketing
• Digital marketing
• Marketing strategy
• Social marketing
• Social media marketing
• Influencer marketing
• Instagram
• Social media

I felt that with this audience I’d be targeting a qualified bunch of people who’d be interested in the stats I had to share. According to Facebook, the approximate total number of people targeted in this audience was 140 million.

5. Duration & Budget:

I then had the task of defining the global budget for the campaign, and splitting it equally between the Instagram ads to comply with the protocol.

I set a global budget for this campaign, which was distributed equally between ad formats (capped at $10 / day / ad) and run over 10 days.

6. Overview:

Overall, including the creation of images and videos, audience building and uploading everything into Ads Manager, the process of setting up this Instagram ads campaign, across all 6 available formats, took me about half a day.

The only notable obstacle that I encountered during the setup process of this test campaign, was the fact that the Boosted post had to be created via the Instagram app, and therefore couldn’t be built in exactly the same way as the other ad formats we setup for this campaign. It basically just added an extra step to the process, where it could have been smoother.

But that’s life.

On to the results!

Results and interpretation

With the help of insights from Google Analytics, Facebook Ads Manager, Instagram (under Promotions), I was able to uncover the preliminary result for each ad.

Another bonus: if you’re an Iconosquare user, you can track this boosted post on the platform, providing you with a whole other set of analytics to tweak and improve your performance! Create your free account today and discover what Iconosquare has to offer.

Performance according to Facebook Ads Manager:

The table below was exported from Ads Manager. It shows very clearly that the most successful ad in the campaign in terms of number of link clicks was the Single image ad, followed by the Stories ad.

Reach was also pretty good for these two ads, along with that of the Boosted post.

The cost per result is the main metric I needed to look to for confirmation: the Carousel ad and the Video ad were expensive, whereas the Single image ad and the Collection ad were cheaper.

However, the Collection ad and the Carousel ad can’t be counted as a success as they only resulted in one click each (which is all that we paid for, so we’re not too upset!). I should state here that it’s always preferable to keep the cost of clicks as low as possible (duh) but essentially, below $1 per click is the goal.

Overall, it’s clear that the best-performing ad was the Single image ad.

Note: The total spend for each ad is different, because I selected automatic optimization during the setup of the campaign. This means that Facebook took the global budget into account and optimized spend according to the performance of each ad within the ad set. In essence: a global budget dedicated to an ad set containing multiple ads in different formats will be split between ads initially. Then, after the first results come in, Facebook (and Instagram) will automatically continue optimizing the spend based on the success of those first results. And this is a good thing! It means that Facebook isn’t throwing your money at ads which aren’t performing. If you’d prefer to have more control over the amount spent, I recommend selecting the option for manual bidding during the campaign creation process. Perhaps a manual bidding strategy permitting total control over the amount spent per ad would alter the results seen? I can feel another experiment brewing…!

Performance according to Instagram Insights

According to the Insights natively provided by Instagram, the Boosted post (which is the only ad format I created in-app — and therefore the only one which was tracked in Insights) did pretty well.

However, when comparing the performance report with that of Facebook Ads Manager, it’s obvious that I was paying too much for the clicks ($1.52).

Let’s also note that Instagram Insights counted 67 promotion clicks, whereas Facebook only reported 66. Small margins of error like this are quite common (different platforms gather data in different ways which can lead to slight discrepancies) but larger gaps would indicate a problem, so watch out!

Traffic to the blog article, per ad format, according to GA:

Why is it important to look into Google Analytics?

Here we can see that the average blog session duration is much higher for users coming from the Stories ad (contrary to the single image ad which had the best stats elsewhere) which is interesting.

Also, the video ad only generated 2 views of the blog post, but those users stayed longer on the page, along with users coming from the Stories ad. So, perhaps users who took the time to watch a video/read the text on the Story were most qualified prospects for our offer.

Summary

So, what did we learn from this experiment?

The takeaways I’d like to leave you with are threefold:

1. It’s worth testing all ad formats to see what works the best with your target community
2. Single image ads are the most basic, but have proved to be the most effective in this test
3. The simplest option — the Boosted post — won’t necessarily give you the best results

After running this test on Instagram ads, I feel like a great combination to test would be a single image from your feed as a Boosted post. You get the simplicity of creation (in-app) with the performance of the Single image ad. Why not give it a try and let us know how you get on?

‍