Most founders build a discount approval process, then never check if it's doing anything. The four metrics that actually tell you: discount leakage rate, approval turnaround time, win rate by discount tier, and renewal discount creep. Track those four and you'll know within a quarter whether your process is protecting margin or just adding a step reps route around.
I built a three-tier approval chain the day I noticed two reps had quietly settled into offering 20% off as their default opening move. The tiers stopped the bleeding on paper. What I didn't have for the first two months was any way to prove it. I was assuming the process worked because it existed. That's a different thing from knowing.
Why "we have an approval process" isn't a metric
A process with no measurement is a policy, not a control. You can require manager sign-off on anything above 15% and still have no idea whether reps are gaming the threshold, whether the sign-off is a rubber stamp, or whether the customers who get approved actually needed the discount to close.
The gap shows up in the data reps don't see: aggregate discount rates by rep, by deal size, by month. If you're not pulling that report, the approval process is running on trust, and trust doesn't show up on a margin line.
Discount leakage rate is the number that matters most
Discount leakage rate is the percentage difference between your list price and your average realized price across all closed-won deals in a period. If your list price implies $500,000 in bookings and you actually booked $410,000, your leakage rate is 18%.
Track it monthly, and segment it by rep. A company-wide leakage rate can hide a single rep discounting at 35% while the rest of the team holds at 8%. That rep isn't a rounding error. Over a year, on meaningful deal volume, the difference compounds into six figures of margin nobody approved.
Approval turnaround time tells you if the process is actually being used
A discount approval process that takes three days to get a signature will get bypassed. Reps find a workaround: quoting the discount before approval, splitting a deal into a "trial" period, or getting a verbal yes from a manager that never gets logged.
Measure the median time between a discount request being submitted and a decision being returned. Anything over 24 hours on a deal under $50,000 in annual contract value is slow enough that reps will start routing around you, and once that habit forms it's hard to undo. Fast approval is what keeps the process the path of least resistance instead of the thing everyone avoids.
Win rate by discount tier shows you if discounting is even working
This is the metric most companies skip, and it's the one that answers the actual question: is this discount making you more likely to close the deal, or are you giving away margin on deals that would have closed anyway?
Pull win rate for deals with no discount, deals with a 1-10% discount, and deals above 10%. In a lot of B2B SaaS pipelines, the win rate difference between 0% and 10% off is small, and the difference between 10% and 25% off is smaller still.
Ebsta and Pavilion's 2025 GTM benchmark data puts average B2B win rates around 19%, down from roughly 29% a year earlier. If your discounted deals aren't clearing that baseline by a meaningful margin, the discount isn't buying you a higher win rate. It's just a lower price.
Renewal discount creep is the one everyone forgets to check
The discount you approve at signing rarely stays a one-time exception. Reps and customer success teams tend to extend it at renewal by default, because raising a price a customer already has feels like a fight nobody wants to pick.
Pull your renewal cohort from 12 months ago and compare the discount rate at signing to the discount rate at the most recent renewal. If it went up, or even held flat while your list price increased, you're accumulating discount debt that compounds every cycle. This is the metric that turns a one-time exception into a permanent haircut on every account it touched.
What to do with 30 days and a spreadsheet
You don't need a BI tool to start. Pull your last two quarters of closed-won deals into a spreadsheet with rep, deal size, list price, close price, and approval timestamp. Calculate leakage rate and turnaround time first, since those need the least data to be useful.
Add win-rate-by-tier once you have 20 or more deals in each bucket. Revisit renewal creep at your next renewal cohort review. The founders who fix their discounting problem are rarely the ones with the strictest approval rules. They're the ones who can see, in one report, exactly where the exceptions are happening and whether they're buying anything.
Frequently asked questions
How often should I review discount approval metrics? Monthly for leakage rate and turnaround time. Quarterly for win rate by tier and renewal creep, since those need more deal volume to be meaningful.
What's a healthy discount leakage rate for early-stage B2B SaaS? There's no universal number, but most healthy pipelines run under 15%. Above 25% usually means either your list price is fictional or your approval process isn't being enforced.
Do I need special software to track this? No. A spreadsheet with rep, deal size, list price, close price, and approval timestamp covers the first two metrics. CRM reporting or a simple pivot table covers the rest.
What if my team resists reporting on this? Frame it as protecting commission, not policing reps. A clear discount-to-win-rate picture usually shows reps they're giving away margin on deals that would have closed anyway, which is in their interest to know.
A discount approval process without these four numbers is a policy you're hoping works. With them, it's a system you can actually manage, and adjust before a habit like reflexive 20% discounting quietly becomes your default price.