Focus groups lie. People say they like "professional, descriptive names" - then choose Apple over International Business Machines. Here are 4 testing methods that predict actual customer behaviour, not stated preferences.
Method 1: Implicit Association Test
Measures subconscious reactions. Flash name on screen, measure response time to positive/negative words.
Method 2: A/B Traffic Testing
Create two identical landing pages. Different names. See which converts better. Actual behaviour trumps opinions.
Method 3: Recall Testing
Show name for 5 seconds. Test recall after 24 hours. Names below 60% recall fail in real markets.
Method 4: Semantic Differential Scales
Rate name on opposite pairs: modern/traditional, friendly/formal, expensive/affordable. Maps brand positioning.
Sample Size Matters
50 minimum for directional data. 200+ for statistical significance. 10 friends = worthless.
Warning: Asking "which name do you prefer?" gets false positives. Test behaviour, not preferences.
Generic Names Can't Be Trademarked
If you want legal protection and a name competitors can't copy, make it distinctive from day one. Testing helps identify distinctive names early.