Price testing is the practice of gathering evidence on how buyers respond to a price before committing the whole book to it. The point is to reduce the size of the bet, not to eliminate it.

Methods split by what they cost and what they prove. Survey-based work, Van Westendorp, Gabor-Granger, conjoint, captures stated intent from people spending nothing. Live testing captures revealed behavior from people spending real money. Stated intent is cheap, fast, and systematically optimistic. Revealed behavior is expensive, slow, and true.

B2B rarely permits the clean version of the second. Deal counts are too low for significance, buyers talk to each other, most-favored-nation clauses make differential pricing a contractual breach, and the sales rep is a variable no test controls for. Consumer businesses A/B test at scale; a manufacturer closing forty deals a quarter cannot, and pretending otherwise produces confident conclusions from noise.

What works instead is bounded exposure. New logos only, so no existing account sees a change. One region or segment where the offer is already fenced. A new SKU carrying the new logic. A staged rollout at renewal, cohort by cohort. None is a controlled experiment, and each one puts a real number in front of a real buyer, which is more than any survey delivers.

Reading the result is where most tests fail. Win rate alone tells you nothing, because the price that wins everything was too low and the loss data is where the signal lives. What matters is whether realized price held, or whether the number survived on paper while discounting quietly absorbed the increase. Reach out to Acustrategy to design a test your deal flow can actually support.