Welcome back!

We hope everyone had an awesome weekend! This week, we’re continuing October’s theme of creative testing and zooming into creative imagery testing specifically. We’ve learned that creative testing more generally involves experimenting with different content variations to find out which performs best with your target audience, so as it says in the name, creative imagery testing involves experimenting with the visual aspects.

Let’s dive in! 🏃‍♂️🚀

News

Tools + Productivity

Strategy

The step-by-step

Before getting into the thick of creative experimentation, start with an audit of current ad/content performance with a gap analysis.

This just means identifying areas for improvement based on key performance indicators (KPIs) such as click-through rate (CTR), conversion rate, and return on ad spend (ROAS). Look at how different audience segments interact with your ads to see which visual and copy elements are performing well and which aren’t.

For example, a fitness app might discover that its ads have a low click-through rate among the 35-44 age group— indicating a potential gap in appealing to this demographic.

Continuing with this example, the next step would be to define a goal— such as, “increase click-through rate by 20% for the 35-44 age group,” and form a hypothesis about potential improvements (like, “using images of people in this age range exercising will resonate better with the target audience”).

Then it’s game time. Develop multiple versions of your ad focusing on specific elements you want to test— changing one element at a time for A/B testing or multiple elements at a time for multivariate testing.

When choosing a testing method, consider the following:

When to use A/B Testing:

1. Clear, Specific Hypotheses: When you have a specific element you want to test (like the headline “Buy Now” vs “Get Started,” or in this case, variations of certain visual element like subject, composition, colors, etc).

2. Limited Traffic or Budget: A/B tests don’t need huge sample sizes to achieve statistical significance.

3. Quick Results Needed: A/B tests can often produce actionable results faster than multivariate tests.

4. Simple Changes: When you're testing straightforward modifications rather than complex combinations of elements.

5. Early Stages of Optimization: If you're just starting with creative testing or optimizing a new campaign.

When to use Multivariate Testing:

1. Complex Hypotheses: When you want to understand how multiple elements interact with each other.

2. High Traffic Volume: Multivariate tests require larger sample sizes to achieve statistical significance across all variations.

3. Substantial Budget: More variations mean higher costs to ensure each combination gets enough exposure.

4. Longer Testing Period Available: Multivariate tests often take longer to reach conclusive results.

5. Advanced Optimization: When you've already done basic A/B testing and want to fine-tune multiple elements simultaneously.

6. Comprehensive Insights Needed: If you want to understand the impact of various elements and their interactions in depth.

Let’s look at how we would implement this using our example (for simplicity and illustrative purposes, we’ll look at an A/B test):

Example A/B Test:

• Version A: Current ad with a young adult exercising.

• Version B: Same ad layout but with a person aged 35-44 exercising.

To make implementing the test easier, use platform-specific tools like Facebook’s experiment tool or Google Optimize, making sure to budget fairly across all variants to ensure each has an equal chance of success.

The duration of the test should be long enough to gather meaningful data, usually at least 1-2 weeks.

Let’s say these are the test results for our example:

• Version A (original): CTR 1.2%, Conversion Rate 2.5%

• Version B (new image): CTR 1.8%, Conversion Rate 3.2%

In this case, Version B shows a 50% increase in CTR and a 28% increase in conversion rate, indicating a successful test— this means future campaigns would implement the image style in Version B.

What kinds of variations are there?

So what does image testing look like in the real world, and in what ways can you change up images?

An obvious example is Netflix’s thumbnail testing. The company constantly tests different thumbnail images for shows and movies to see which ones receive the most clicks.

Airbnb does the same with its’ properties, testing professional, staged photos as well as authentic, user-generated photos. Interestingly, they found that while professional photos initially performed better, authentic photos led to more bookings in the long run.

A CRM software company might test content featuring a smiling customer service representative and compare it with a snapshot of the product interface to see which image performs better.

Another variation is the composition and alignment of image ads on their pages. Dell used A/B testing to place product images on their laptop category page between left-aligned and center-aligned images. They found that center-aligned images increased click-through rates by 36%.

You can also switch up the size— Fab.com, an e-commerce site, tested smaller, uniform-sized images as well as larger, variable-sized images (multivariate testing for size and variability!), and found that the latter increased click-through rates and engagement.

Looking Ahead

The examples above really go to show that oftentimes, seemingly minute differences in your ad campaign can have a major impact on their effectiveness, as seen in click-through rates and other KPIs.

That’s why testing is necessary to find out whether tweaks in visuals make a difference in impact (and based on that, maximize ROI).

If you’re interested in automating your marketing processes to grow faster, click here to learn more about how Loopgenius can help you.

Stay tuned for next week, where we’ll dig into using testing for audience targeting & segmentation!

Creative Testing

Week 1: Message testing
Week 2: Creative imagery testing (this week!)
Week 3: Audience Targeting & Segmentation
Week 4: Landing Page Optimization

REFERRAL PROGRAM

Hey! Don’t keep us a secret! Refer your friends and start earning rewards! Share this link with your friends: https://beyond9to5.beehiiv.com/subscribe?ref=PLACEHOLDER

If you get just 3 friends to subscribe you’ll get our exclusive sticker pack OR refer more and earn a trip to Vegas!

Your Progress