How To Create an Iterative Creative Process

Digital marketers buying ads on Meta & TikTok know that the algorithms are in charge. But that doesn't mean Creative is any less important. Here's how to feed those machine learning systems the best possible creative.

𝐂𝐫𝐞𝐚𝐭𝐢𝐧𝐠 𝐚𝐧 𝐈𝐭𝐞𝐫𝐚𝐭𝐢𝐯𝐞 𝐂𝐫𝐞𝐚𝐭𝐢𝐯𝐞 𝐏𝐫𝐨𝐜𝐞𝐬𝐬

Ultimately, you want to build an iterative creative process that optimizes Ad Diversity and Team Learnings.

𝐀𝐝 𝐃𝐢𝐯𝐞𝐫𝐬𝐢𝐭𝐲: Your creative process should result in diverse creatives. This variety provides the ad algorithms the widest range of choices, increasing the probability of finding a winner.

𝐓𝐞𝐚𝐦 𝐋𝐞𝐚𝐫𝐧𝐢𝐧𝐠𝐬: Your creative process should be strategic and intentional, ensuring that your team is always learning. Every ad you launch should reveal a clear winner and the reason for its success, allowing you to double down with each new generation of creative assets.

𝐁𝐞𝐭𝐭𝐞𝐫 𝐂𝐫𝐞𝐚𝐭𝐢𝐯𝐞 𝐃𝐫𝐢𝐯𝐞𝐬 𝐁𝐞𝐭𝐭𝐞𝐫 𝐀𝐮𝐜𝐭𝐢𝐨𝐧 𝐏𝐞𝐫𝐟𝐨𝐫𝐦𝐚𝐧𝐜𝐞
Creative is one of 3 core pillars of success in the auction. Your auction bids are multiplied by your quality score. To lift quality scores, you need:

▶ Strong creative (to drives high click-through rates)
▶ Strong landing pages (to drive high conversion rates)
▶ Strong data (to capture 100% of conversions)

𝐌𝐞𝐚𝐬𝐮𝐫𝐢𝐧𝐠 𝐂𝐫𝐞𝐚𝐭𝐢𝐯𝐞 𝐒𝐮𝐜𝐜𝐞𝐬𝐬
As you build creative, monitor ad metrics that indicate high quality scores. Use clicks, click-through rate, and engagement scores to see how Meta perceives your creative.

𝐎𝐩𝐭𝐢𝐦𝐚𝐥 𝐂𝐫𝐞𝐚𝐭𝐢𝐯𝐞 𝐅𝐨𝐫𝐦𝐚𝐭𝐬
Creative format is crucial for automated campaigns that show your creative in different placements with varying formats and configurations.

Ensure that every creative concept has different versions supporting 9:16, 1:1, and 16:9 aspect ratios to perform well across all placements.

Consider creating both static images and videos to reach different audiences based on their preferences.

𝐁𝐮𝐢𝐥𝐝 𝐚𝐧 𝐈𝐭𝐞𝐫𝐚𝐭𝐢𝐯𝐞 𝐏𝐫𝐨𝐜𝐞𝐬𝐬
Develop an iterative process that runs a single cycle of learning and optimization. I recommend this process:

▶ Plan: Have a strategic discussion about the creative concept, what makes it distinctive, and what you'd learn if it becomes a winner.

▶ Execute: Build the concept and extend it to all formats (image vs. video) and various aspect ratios.

▶ Measure: Launch the campaigns and track key click and engagement metrics. Understand which creatives were preferred by the algorithm and why.

▶ Adapt: Document your learnings and incorporate them into the next cycle, starting with a new planning phase.

If you're a small, resource-constrained brand, you might run a creative cycle once per month. A larger brand with more resources might run a cycle once per week. 

𝐒𝐮𝐦𝐦𝐚𝐫𝐲
Be intentional with your creative tests. Build a repeatable, iterative process. Run creative cycles at your own pace. Always be learning.