AD testing structure
The speaker explains an ad testing structure where each ad group represents a completely new creative concept, containing one original ad and three variations (one major, one medium, and one minor) to test which will perform best. The approach mimics a 'shotgun method' to discover winning creatives, as the original concept is never guaranteed to be the winner.
Summary
The speaker describes a structured approach to ad testing organized at the ad group level, where each ad group is treated as a completely new and untested creative concept. Within each ad group, there are four ads: one original creative concept that has never been tested in an advertising account, and three variations of that original concept. These variations differ in their degree of change from the original. The speaker emphasizes that this original is not a proven winner—it's simply a new idea being tested for the first time. The three variations serve different purposes: one is a major variation that changes the entire structure of the creative while maintaining the original concept, one is a medium variation that represents a significant but not structural change, and one is a minor variation that might only involve changing the hook or headline. The speaker uses the analogy of a 'shotgun method' to explain the rationale: by testing multiple variations simultaneously, they can identify which creative—not necessarily the original—will ultimately emerge as the winning ad in that group. The speaker stresses that there's no way to predict in advance whether the original concept or one of its variations will perform best, making this multi-variation testing approach essential for discovering top performers.
Key Insights
- The speaker argues that each ad group level is defined as a completely new concept that has never been tested in an advertising account before, and the original creative is not a proven winner but rather a new untested idea
- The speaker claims that only one creative typically emerges as the winner from a new ad group, but there is no way to predict in advance whether it will be the original or one of the variations
- The speaker identifies three types of variations: minor variations (changing just the hook), medium variations, and major variations (changing the entire structure while maintaining the original concept)
- The speaker compares the testing structure to a shotgun method where multiple variations of a new concept are tested simultaneously because it's unknown which variation will win
- The speaker emphasizes that all four ads in an ad group are based on one original new concept, with the key difference being that some are the original idea while others are variations, but all are part of testing that single new concept
Topics
Transcript
[0:00] Every time I launch a new ad group, it's a completely new concept. The ad group level is synonymous with a completely new concept. And inside this ad group, you can have four ads. One is an original new concept. One is a big variation on this concept. So technically it's a variation, but not a winning variation, because this whole ad group is a completely new concept. And another great variation on the same original new concept. And then you can make a slight variation on the same original concept. They are all [0:31] part of one original new concept. The only difference is that this is an original idea for the creative, and the rest are just…
Full transcript available for MurmurCast members
Sign Up to AccessMore from Mark Builds Brands
If Your Website Looks Amazing but Nobody Buys, Watch This
The speaker argues that aesthetic design quality does not directly correlate with conversion rates. While visual appeal matters for brand building, the primary focus for conversion optimization should be on metrics like Customer Acquisition Cost (CAC) rather than making a website visually stunning.
The Free Gift Objection That's Killing Your Conversions
The speaker argues that most marketers use generic offer packages (1-2-3 bundles, BOGO deals) that don't effectively overcome customer objections. Free gifts only drive conversions when they specifically address a distinct customer problem that the main product doesn't solve.
Take bigger swings
The speaker advocates for taking bolder, more creative risks in advertising and landing page testing, especially when operating at smaller scales. Rather than making incremental variations on existing winners, marketers should test diverse, extreme concepts to discover new winners and build account stability.
Never rely on one of anything
The speaker argues that relying on a single approach, platform, account, or creative format in e-commerce is inherently risky and creates vulnerability. By diversifying across advertising accounts, payment processors, business structures, and creative formats, businesses can build stability and avoid starting from scratch when one method fails.
The Real Bottleneck in Every Ad Account
The speaker argues that poor creative quality is the primary reason advertising accounts fail to generate profit. They provide specific KPI benchmarks for link CTR (minimum 3%), CPC for US traffic (around $1.50-$2), and native static CTR (5-6%), emphasizing that creatives are the real bottleneck in ad performance.