The discount breakeven table every salesperson should carry
Most discounting arguments are arguments about a number nobody in the room has calculated. One line of arithmetic ends them: the extra volume a discount demands just to stand still. Here is the table, the give-get rule that follows from it, and the four-row matrix that governs who may spend your margin.
Summary
- Required volume increase = d ÷ (CM − d). At 30% contribution margin a 20% discount needs you to triple your volume simply to stand still.
- The table is not a weapon for refusing discounts. It converts a feeling into a number, reframes the question to what could we get instead, and surfaces the alternatives worth asking for.
- Never concede price without receiving something: a discount granted for nothing teaches the buyer that your price is fiction, and they will test it again at every renewal.
The salesperson says the deal needs 20% off or it walks. The founder says that feels like a lot. Neither of them can say what 20% off actually costs, so the conversation resolves on conviction, urgency, and who is more tired.
There is a single line of arithmetic that ends that conversation permanently. It fits on a card.
The identity
When you cut price by d and your is CM, the extra volume you need simply to end up where you started is:
Required volume increase = d ÷ (CM − d)
That is the whole thing. What it produces, though, is genuinely surprising to most people the first time they see it.
If your contribution margin is 50%
| Discount | Extra volume just to stand still |
|---|---|
| 5% | 11% |
| 10% | 25% |
| 15% | 43% |
| 20% | 67% |
| 25% | 100% |
| 30% | 150% |
If your contribution margin is 30%
| Discount | Extra volume just to stand still |
|---|---|
| 5% | 20% |
| 10% | 50% |
| 15% | 100% |
| 20% | 200% |
| 25% | 500% |
| 30% | Infinite: you are at zero contribution |
And if your contribution margin is 20%, which plenty of Indian manufacturing, distribution and services businesses are:
If your contribution margin is 20%
| Discount | Extra volume just to stand still |
|---|---|
| 5% | 33% |
| 10% | 100% |
| 15% | 300% |
| 20% | Infinite |
That last table is worth sitting with. At 20% contribution margin, a 10% discount requires you to double your business to be no better and no worse off than before you gave it.
Build your own version, not this one
The tables above are illustrations. The only ones that matter are the ones built on your actual contribution margin, which means every cost that varies with serving a customer, including implementation, support, customer success, payment processing and the financing cost of your receivables.
Compute it per segment. Enterprise and SMB will not have the same floor, and the segment you priced lowest is very often the one that costs the most to serve.
Then print it. Genuinely print it. A table that lives in a spreadsheet somebody owns is not carried into a negotiation; a card in a laptop bag is.
How it actually gets used
The table is not a weapon for refusing discounts. It is a tool for having a better conversation, and it does three specific things.
It converts a feeling into a number. "20% is a lot" is an opinion the salesperson can argue with. "20% means we need two-thirds more volume from this account to be where we are today" is a fact you can both look at. The argument stops being about conviction.
It reframes the question. Once the cost of the discount is visible, the natural next question is not should we give 20%? but what could we get instead that is worth more than 67% more volume? That is a far more productive place for a sales conversation to land.
It surfaces the alternatives. Almost always, there is something the customer will give you that costs them less than the discount costs you. A longer term. Payment in advance. A reference call. A case study. Reduced scope. A commitment to a second product. The table is what makes those alternatives visibly worth pursuing.
The give-get rule
This follows directly, and it should become non-negotiable in your sales culture.
Never concede price without receiving something.
Not because concessions are wrong: discounting is a legitimate commercial tool. But because a discount granted for nothing teaches the buyer that your price is fiction. And a buyer who learns that will test it again at renewal, and at every renewal after that.
Things worth asking for, roughly in order of how easily buyers agree to them:
- A longer contract term
- Payment in advance or quarterly rather than in arrears
- A named case study or logo rights
- Two reference calls in the next six months
- Reduced scope: fewer modules, fewer sites, a smaller pilot
- An annual escalator clause written into the contract
- Commitment to a second product or a defined expansion trigger
That last one, the escalator, deserves attention. A 6% annual uplift written into a three-year contract is worth over 12% of cumulative contract value, requires no renegotiation, and is far easier to agree at signature than to impose at renewal. Most Indian founders have never asked for one.
The four-row matrix that does the rest
The table tells you what a discount costs. The matrix decides who may spend it.
The discount authority matrix: four rows and an absolute
| Discount depth | Who approves | What is required |
|---|---|---|
| 0 to 10% | The salesperson | A note in the CRM |
| 10 to 20% | Sales lead | Give-get documented |
| 20 to 30% | Revenue or finance head | Written case, contribution margin checked |
| 30%+ | Founder | Strategic rationale, visible to the board |
| Below the contribution margin floor | Nobody | Not permitted |
Four rows and an absolute. It takes an afternoon to build and it is the highest-return artifact in this entire hub.
Three things make it work:
The friction is the feature. If a 25% discount requires a written case, fewer will be requested, and the ones that are will be better ones.
Track the distribution, not the average. The average discount depth hides the tail, and the tail is where the damage lives. Look at the histogram and examine the bottom decile by name.
Measure it by salesperson. Wide variance between reps on comparable deals is a coaching signal, not a market signal. In a large corporate this is exactly the work a VP of Pricing owns.
What you will find when you run this
Two things, reliably. The first is that your realized price is lower than you think. Extract the actual discount granted across your last twelve months of contracts and compare it to what you believe your policy is. In our experience the gap surprises almost every founder, because the policy was never written down and so was never violated.
The second is that some of your largest customers by revenue sit at the bottom of your contribution margin ranking. Usually they are the early logos, won on deep discounts when you badly needed proof, and never repriced since.
Neither of those is a reporting failure. They are the findings. And you cannot see either of them without the arithmetic. Both are among the leaks that never show up in your reporting.
Your salespeople are making pricing decisions every week with more conviction than information. Give them the table.
Frequently asked questions
How do I calculate discount breakeven?
Required volume increase = discount ÷ (contribution margin − discount), with both expressed as fractions of price. At 40% contribution margin, a 10% discount needs 10 ÷ (40 − 10) = 33% more volume to break even.
What discount is too much?
There is no universal answer: it depends entirely on your contribution margin. What is universal is the floor. Any discount that takes a deal below contribution margin destroys value on every unit sold, and no volume will fix it.
Should we ever discount at all?
Yes. Discounting is a legitimate tool for winning volume, entering a segment, or securing a strategic logo. What makes it damaging is doing it without a rule, without a floor, and without receiving something in return.
How useful was this article?
One tap. It tells us what to write more of.
The next one
Get what we publish next, by email.
Working notes on raising, borrowing, protecting, growing and structuring capital in India. One email a week at most, and you can leave any time.
We use your address only to send this. See our privacy policy.
We store your address to send you these emails and nothing else. See our privacy policy.
Related reading
