The comparison that fools most teams
Pull win rate by discount band from any CRM and the discounted deals usually look better. That result is manufactured by which deals get discounted in the first place. Reps offer concessions on deals that reached a decision point with a live competitor and an engaged approver, which is the profile that converts anyway. Deals that stalled in discovery never get a discount, and they sit in the undiscounted group depressing its rate.
How to run the test honestly
| Control for | Why it matters |
|---|---|
| Stage reached | Discounts cluster in late stage, so unmatched groups compare maturity rather than pricing |
| Deal size band | Concession size scales with contract value and so does deal complexity |
| Competitor present | A named competitor changes both the odds and the likelihood of a discount |
| Created cohort | Both groups need equal time to resolve or the faster group looks stronger |
The cost that never appears in the win rate
The visible cost is realization. ORM's data shows most deals close for less than the value recorded against them in the CRM. A pipeline averaging $80,000 in average deal size against $40,000 on closed-won deals is the shape of that gap. That gap flows straight into every forecast built from CRM amounts.
The hidden cost is the calendar. ORM decomposes a quarter into carry-over deals, deals created and closed inside the period, and deals pulled forward from future periods, with pull-forward frequently bought through discounting. Each pull-forward saves the current number by removing revenue from the next one, which is why a team can hit three quarters in a row and then miss badly with no change in selling performance.
Track discount depth against realization and against next quarter's opening pipeline before concluding that concessions are working. See forecast accuracy for the measure that degrades when pulled-forward revenue is treated as growth, and win rate for the segmentation that makes the matched comparison possible.
Frequently Asked Questions
Why does the data usually show discounted deals winning more often?
Selection, not causation. Reps discount where they believe a win is available, on late stage deals with an engaged approver and a live competitor. Deals that never reached that point are never discounted, so they sit in the undiscounted group and drag its rate down. The comparison is between deals at different maturities rather than between two pricing strategies.
How do you test the effect properly?
Match the groups before you compare them. Hold segment, deal size band, competitor presence, and stage reached constant, then compare win rate across discount bands inside each matched cell. If the discounted and undiscounted cells converge once maturity is held constant, the discount was buying speed rather than outcomes.
What does a discount reliably buy?
Timing. ORM treats pull-forward deals as one of the three sources of a quarter's revenue, and notes that they often close early through discounting or a tradeoff against a future quarter. Teams routinely understate what that costs. The deal that closes in March at a discount is the deal that was going to close in April at list.
Which objection does a discount actually address?
Only the budget objection, and only when the budget holder is engaged. A deal stalling on unclear value, a missing approver, or an unresolved security review does not move because the price dropped. Discounting into those conditions converts a possible win into a smaller possible win.
Put these metrics to work
ORM builds custom revenue forecast models that turn concepts like does discounting improve win rate? into prescriptive action for your team.
Schedule a Demo