Sales reps and pricing teams generally agree on one thing: discounts win deals. The instinct is intuitive, the conversations support it, and the rep with the looser hand on price often does pull in more revenue. What the conventional wisdom misses is that discounts only matter where price is actually the deciding factor — and across most ranges of most B2B businesses, it isn't.
For a building-materials manufacturer we worked with, this assumption was costing real money. The team's discounting policy gave reps wide latitude, with the implicit theory that bigger discounts close more deals. The data — quote-by-quote close rates, mapped against discount levels — told a very different story.
The Empirical Question
We assembled the firm's quote-level data for a multi-year window: every quote, the discount applied, and whether the deal closed. The exercise wasn't to look for the relationship between discount and closing — it was to test whether one existed in the first place, and across what ranges. Some discount levels are real concessions; others are just rounding errors that fall well within the customer's tolerance for paying.
The pricing question that matters isn't "what discount should I offer?" — it's "what's the smallest discount that wouldn't change the customer's decision?" Anything beyond that is margin you've handed over voluntarily.
What We Found
Across a wide middle range of discount levels, close rate was effectively flat. A 22% discount was no more likely to win than a 12% discount. The customer had decided whether they wanted the product before discount levels mattered, and the small variations within that range weren't moving the decision. At very low discounts (under about 5%), close rate ticked down — those quotes were often submitted without real concession, sometimes accompanied by other friction. At very high discounts (above about 25%), close rate did rise meaningfully — those concessions were buying real wins. But the middle, where most quotes lived, was elasticity-free.
where discounts didn't move close rate
inside that flat zone
tightened discount band
What Changed
The team revised the discounting policy to constrain the band more tightly, aligned with the empirical evidence. Reps still had latitude — judgment still mattered for the deals at the edges where price was actually a factor — but the default discount range narrowed substantially. The result was a meaningful margin recovery on deals that were going to close anyway, with no measurable impact on win rate.
The Broader Point
Most pricing decisions get made without empirical data. Reps believe what their customers tell them, leadership trusts the reps, and the assumption that "we had to discount to win that deal" goes untested. It's testable — every business that quotes prices and records outcomes has the data sitting in its CRM — and the answer is rarely what intuition suggests. Sometimes elasticity is real. Often, in the middle of the price range where most deals live, it isn't.
The same analysis applies anywhere the relationship between price concession and close rate matters: subscription renewals, professional services, durable goods, software contracts. The first question isn't "what discount level wins more deals?" — it's "where does discounting stop being elastic and start being a gift?" The data is usually willing to answer that.
Wondering whether your discounting is buying you anything? Say hello.