Sales Discounting Policy: How to Set B2B Discount Approval Thresholds That Protect Margin
By Rick Elmore ·
Every B2B sales team leaks margin through discounts, and most leaders don't find out until the quarter closes and the blended ASP comes in lower than the forecast. The reason is almost never a greedy rep. It's the absence of a discounting policy that tells people exactly what they can give away on their own, what needs a sign-off, and how fast that sign-off happens.
The short answer: build a tiered discount approval process where small discounts are self-serve, mid-size discounts route to a manager, and anything that touches your margin floor escalates to finance or the VP. Set the thresholds off your gross margin math, not off round numbers, and automate the routing so approvals happen in minutes instead of days. Do that and reps stay fast while your margin stays intact.
Why a discounting policy protects margin better than discipline alone
You can tell a rep to "hold the line on price" all day. Under quota pressure, with a prospect threatening to walk, discipline loses. What wins is a system that makes the right behavior the easy behavior and the wrong behavior require a conversation.
Here's the mechanical problem. A discount isn't a linear hit to revenue. It's a direct hit to profit. If your gross margin is 70% and a rep gives a 15% discount, you didn't lose 15% of the deal. You lost roughly 21% of the profit on that deal, because the discount comes straight off the margin, not off the cost. On lower-margin products the leverage is worse. Reps rarely think in these terms during a negotiation, and they shouldn't have to. The policy should do the thinking for them.
A good discounting policy does three things at once. It sets clear ceilings so reps know their authority without asking. It routes exceptions to the right person automatically. And it creates a paper trail of why discounts were given, which is the only way you ever improve the policy later. When teams skip the policy and rely on a shared sense of "be reasonable," the result is predictable: your best negotiators give away the least and your most pressured reps give away the most, with no connection to deal quality.
How to set discount approval thresholds using your margin floor
Thresholds should start from one number: the lowest margin you're willing to accept on a standard deal. Call it your margin floor. Everything in your discounting policy works backward from there.
Work through it in this order:
- Define your margin floor. This is the gross margin below which a deal stops being worth it. For most B2B software and services businesses that floor sits well above zero, because you still have to fund CAC, onboarding, and support out of what's left. Pick the real number, not the aspirational one.
- Translate the floor into a maximum discount. Given your list price and standard cost, calculate the discount percentage that would drag a deal down to the floor. That's your absolute ceiling—the point where no rep or manager can approve on their own.
- Carve the space below the ceiling into tiers. Split the range between zero and your maximum discount into bands. A common shape: self-serve up to a small percentage, manager approval up to a middle band, and director or finance approval up to the ceiling.
- Adjust tiers by product and segment. A high-margin flagship product can tolerate a wider rep-level band than a thin-margin add-on. Enterprise deals with multi-year commitments justify different rules than one-year SMB deals. Don't apply one flat discount table across products that behave nothing alike.
- Set hard non-negotiables. Some things never get discounted regardless of tier—implementation fees you need to cover real cost, or the first year of a multi-year deal that's already priced to win. Name them explicitly so they don't get bundled into a "total deal discount."
The output of this exercise is a simple grid every rep can read in ten seconds. The grid below shows the shape—plug in your own numbers once you've run the margin math.
What a tiered discount approval process looks like
The table is where abstract policy becomes something a rep can act on mid-call. Keep it small enough to memorize.
| Discount band | Who approves | Target turnaround | What's required |
|---|---|---|---|
| 0–10% | Rep (self-serve) | Immediate | Logged in CRM, no approval needed |
| 11–20% | Sales manager | Under 1 hour | Short justification: competitor, volume, or timing |
| 21–30% | Sales director / RevOps | Same day | Deal context, margin impact, path to close |
| 31% to margin floor | VP Sales + Finance | 24 hours | Full business case, strategic rationale |
| Below margin floor | Not approvable | N/A | Requires a scope or term change, not a discount |
Two design principles make this work. First, each tier up should require materially more justification, so the friction scales with the margin at risk. Second, the turnaround times are promises, not suggestions. The whole policy collapses if a 15% approval sits in a manager's inbox for two days while the deal cools. Fast approval is what buys you the right to have approval at all.
Notice the bottom row. Anything below the margin floor isn't a bigger discount—it's a different deal. If a prospect genuinely can't afford the floor price, the answer is a smaller scope, a longer term, or a different package, not a sign-off that erases your profit. This is the single rule that separates a discounting policy from a rubber stamp.
When exceptions should escalate—and when to just say no
Every discount policy needs an exception path, because real deals don't fit grids. The mistake is treating every exception as a yes waiting to happen. Escalation exists to apply judgment, not to find a way to approve.
Legitimate reasons to escalate past the normal tiers are specific and rare. A genuinely strategic logo that opens a new segment. A multi-year commitment with real expansion built in. A competitive displacement where winning the account changes your position in the market. These justify going deeper because the lifetime value, not the first deal, is what you're pricing against.
The reasons that should get a fast no are just as specific. "The prospect says a competitor is cheaper" with no proof. "We need this to hit the number" at quarter-end. "They'll sign today if we drop another 10%." These aren't strategic exceptions. They're negotiation pressure doing its job, and the only correct response is to let the policy hold. A rep who hears "no, that's below our floor—here's what we can restructure instead" learns something. A rep who hears "fine, approved" learns to open lower next time.
One rule keeps the exception path honest: whoever approves an exception owns the margin consequence in their own reporting. When a VP's approved exceptions show up in the VP's blended margin, exceptions stay rare. When they vanish into a general pool, they multiply.
How to automate discount routing so speed doesn't suffer
A tiered policy that lives in a PDF nobody reads is worse than no policy, because it creates the illusion of governance without the reality. The policy only works when it's wired into the tools reps already use.
The automation layer should handle four jobs:
- Enforce the thresholds automatically. When a rep sets a price in the CRM or quoting tool, the system knows the discount percentage and the margin impact instantly. Within the self-serve band it just logs. Above it, the quote can't advance to a signed state until the right approval clears.
- Route to the correct approver without a human deciding who. The tier determines the approver. A 14% discount goes to the manager; a 24% discount goes to the director. No rep has to ask "who do I send this to," and no approval lands on the wrong desk.
- Enforce the turnaround with reminders and fallbacks. If an approver doesn't respond inside the target window, the request escalates or pings them again. Speed is a feature of the system, not a hope.
- Capture the reason every time. The justification field isn't bureaucracy—it's the data you use to refine the policy. After a quarter you can see which tiers get used most, which reasons recur, and whether a band is set too tight or too loose.
This is where discount governance connects to the rest of your revenue engine. The approval rules sit alongside your quoting logic, your CRM stage gates, and your reporting. When they're built as one connected system rather than bolted together, a rep gets an answer in minutes and leadership gets clean margin data without chasing anyone. That integration is exactly the kind of RevOps plumbing we build into our done-for-you packages, because the policy and the automation have to ship together to actually hold.
A warning on tooling: automation amplifies whatever policy you feed it. Automating a bad discount table just means you leak margin faster and more consistently. Get the thresholds right on paper first, test them against a quarter of closed deals to see what would have changed, then wire them in.
Where this fits
A discounting policy is one layer of a larger revenue operating system. It sits next to your quoting rules, your deal stages, and your forecasting, and it only delivers when those pieces talk to each other. Set thresholds off your real margin floor, route exceptions to people who own the consequence, keep approvals fast enough that reps don't route around them, and let the automation enforce it all quietly in the background. Done right, you stop choosing between speed and margin—you get both, because the system makes the margin-safe move the fast one.
If you're not sure where your deals are leaking margin or how tight your thresholds should be, we can map it with you. Book a Revenue Systems Audit and we'll pressure-test your discount approval process against your actual closed-won data.