Here's the problem most PMMs face: you're about to brief your marketing team on a product you don't fully understand yet. No customer research backing it. No positioning validated. Just a product that exists and a launch date that's locked.
That's exactly where Shannon found herself a few years back. Her hand was forced. So she did something unconventional—she decided to test the messaging in market before committing the budget. She worked with her acquisition team, set up rapid tests, and learned more in two weeks than she would have in months of gut feel.
That forced experiment became a framework. And she's now refined it across 15+ launches for D2C, PLG, and early-stage companies.
What is A/B message testing for product launches?
According to Shannon, it's the fastest way to move from ideas, hypotheses, and gut feel into actual data-backed things that work—and then scale them. It's not a replacement for research. It's a feedback loop during your launch that lets you test multiple angles in market, learn what resonates in real time, and bring those insights back to the entire organization so everyone's learning together as you roll out.
Here's how Shannon approaches doing message testing for product launches:
Step 1: Start with performance context — Pull existing data that informs how tests should run: old campaign data, results from your last message test, whatever you have that shows what's worked before. Work with your acquisition team to see what they want to measure.
Step 2: Design your test structure — Run a couple rounds of concept tests with a specific budget allocation. Build 3-4 ad variants per concept (not just one). This gives you valid output. Shannon's seen organizations run $500 per concept over a 5-day sprint for enough data to make a decision.
Step 3: Define your primary test metric — How are you actually measuring success? Short sales cycle D2C? Maybe it's return on ad spend. Maybe you need to move up market and just look at cost per click. These are directional results—you're not trying to do the acquisition team's job, you're getting a read on the market.
Step 4: Build hypotheses from real inputs — Find recurring themes in Reddit threads, support tickets, competitor ads, customer conversations, G2 reviews. Turn those themes into testable predictions. Then structure each hypothesis around the three levers: emotional messaging (aspiration, loss framing), functional messaging (what the product does), or social messaging (who they want to be seen as).
Step 5: Hand off to creative and acquisition, then iterate — Write out your hypothesis, the insight backing it, the behavioral driver you're testing, and any creative guardrails. Let them run the test. Track directional results (not necessarily stat sig). Bring learnings back to the org. Rinse and repeat.
And a special shout out to this season's sponsor, PMM Camp - where product marketers build leadership range so they can move beyond execution to influence decisions, shape strategy, and build a trusted team.
Podden och tillhörande omslagsbild på den här sidan tillhör
An Alley-Oop Podcast. Innehållet i podden är skapat av An Alley-Oop Podcast och inte av,
eller tillsammans med, Poddtoppen.
