

Why most message testing is complete BS (and how to fix it)
10 snips Mar 27, 2025
Tired of ineffective message testing? This discussion uncovers why many product marketers are missing the mark, with only 30% ever testing messages correctly. A/B testing is deemed a cop-out, while testing multiple variables is likened to chaos in the kitchen. Discover the crucial distinction between message testing and A/B testing, and learn how to craft a framework that gets buy-in from leadership. Plus, enjoy hilarious analogies about cooking that parallel the frustrations of marketing clarity. Get ready to refine your messaging strategy!
AI Snips
Chapters
Transcript
Episode notes
A/B Tests Are Misapplied
- A/B tests are often misused as a default when teams avoid committing to a direction.
- Small sample sizes (e.g., 150 people) and low traffic usually make A/B results meaningless.
Use A/B Tests To Scale, Not Discover
- Use A/B testing to scale and optimize existing, high-traffic pages rather than to discover positioning.
- Use message testing to validate positioning before launching broad campaigns.
Feature Launch A/B Test With 150 People
- Gab recalled a feature launch where a senior insisted on A/B testing subject lines for 150 recipients.
- She pushed to commit because that sample size would not produce useful insight.