Should You Discount to Close a Deal?

Should You Discount to Close a Deal?

Rarely, and almost never without getting something back. A discount is the fastest concession you own and the most expensive one, because it repeats every renewal and it teaches the buyer what your price really is. Trade it for term, volume, a reference or a case study, or do not trade it at all.

We are not a sales team, but we see the consequences on every pricing page we build. A company asks us to publish prices and then admits half their customers pay something different, which means the page cannot say anything true.

So this is worth arguing out properly. Here is the case on both sides and where we land.

What Does a Discount Actually Cost?

Much more than the percentage. Start with the arithmetic, because it is worse than intuition suggests. A 20% discount means you need 25% more volume to hold the same revenue, since one divided by 0.8 is 1.25.

Now apply it to margin. On a product with an 80% gross margin, a list price of 100 leaves 80 of gross profit. Discount the price to 80 and the cost does not move, so gross profit falls to 60. A 20% price cut became a 25% cut in gross profit.

Then remember it annuitises. In subscription software the discount is not a one off, it is the new baseline for every renewal unless you have written an escalation into the contract, which most companies do not.

Is There Evidence That Discounted Customers Behave Differently?

There is, though it should be read with its source in mind. Paddle published an analysis in July 2022, drawn from its own ProfitWell metrics platform, comparing 55 companies doing minimal discounting with 33 doing aggressive discounting. It reported the average difference in lifetime value was upwards of 32.41% less for the discounted cohorts.

Paddle also reported that discounted customers showed lower willingness to pay, higher price sensitivity and higher churn than the core group. That is a vendor's own analysis of its own customer base rather than independent research, and we would not treat the exact figure as a law of nature.

But the direction matches what we hear from every sales leader we talk to. Customers acquired on price are the ones most available to be acquired on price by somebody else.

Why Does the Discount Reach Beyond the Deal?

Because prices leak. Procurement teams talk, industry peers compare notes, and review sites and communities circulate what people actually pay. The discount you gave quietly in March becomes the anchor a different buyer opens with in September.

It also reaches inside your own company. Once a rep has closed a deal at 30% off, that becomes the reference for the next negotiation, and the next. Discounting is a culture more than a decision, and it sets in faster than anyone expects.

And it undermines the pricing page you spent months getting right, which is the point at which it becomes our problem too. A public price nobody pays is worse than no public price, because it costs you credibility as well as revenue. That tension sits underneath our piece on pricing and packaging tiers.

When Is Discounting Genuinely the Right Call?

Three situations, in our view. When you are buying something specific and you name it. When the customer's buying process genuinely requires a negotiated outcome. And when you are pricing for a future you can see, such as a land and expand deal where the first contract is deliberately small.

The first is the important one. An annual prepayment is worth a discount because it improves your cash position and removes twelve renewal decisions. A multi year term is worth a discount because it removes churn risk. A named reference, a case study or a logo you can use is worth a discount because it lowers the cost of the next deal.

What all of these share is that the buyer gives something. A discount given for nothing is not a negotiation, it is a price correction you have not admitted to.

What Should You Offer Instead?

Scope, not price. Reduce what they get rather than what they pay, because that preserves the relationship between value and cost that your whole pricing model depends on.

Fewer seats. A lower usage tier. A shorter initial term. Onboarding removed or self serve rather than assisted. Slower support response. Each of these lands as a real concession to the buyer and costs you far less than an equivalent discount, and each leaves a clean path to expansion later.

Time limited pricing is the other honest option: a genuinely temporary ramp for the first year, written into the contract with the step up stated. That works only if you actually enforce the step up, which is a conversation better had in advance than at renewal, and it is close cousin to raising prices on existing customers.

Does End of Quarter Discounting Work?

It works on the quarter and it damages the year. Buyers with any experience know when your quarter ends, and once they learn you will move on price in the final week, they will wait for it every time.

You have not accelerated the deal. You have trained the customer to delay, and you have given away margin for a timing benefit that repeats itself away within a couple of cycles.

If your forecast depends on quarter end concessions, the real problem is pipeline coverage rather than pricing, and discounting is treating the symptom at considerable cost.

How Does Expansion Change the Calculation?

It is the strongest argument for a small first deal and the weakest argument for a cheap one. If your customers reliably grow, the initial contract value matters less than getting the right customer in the door with the right expectations.

ICONIQ's 2025 State of Software reporting describes net dollar retention settling into a healthy range of roughly 110 to 120%. Public results show the same shape at scale: Datadog reported revenue of 1.12 billion dollars for the quarter ended 30 June 2026, up 36% year over year, while its count of customers with annual recurring revenue of 100,000 dollars or more grew 23%, from about 3,850 to about 4,720.

Revenue growing faster than the count of large customers is expansion inside existing accounts. If that is your model, start small deliberately, at full price, and grow. Starting cheap is different from starting small, and only one of them is recoverable.

Who Should Be Allowed to Approve One?

Someone whose incentives are not fixed by the current quarter, and the threshold should be written down. In practice that means a rep can concede a defined small band alone, anything beyond it needs a second signature, and anything unusual needs a reason recorded in the system.

The recording matters as much as the approval. If you cannot report on your own discounting by segment, rep and quarter, you cannot tell whether you have a pricing problem, a positioning problem or a sales process problem. Most companies discover they have all three.

Also check what a discount does to any usage based mechanics you run. Stripe's documentation notes that when it evaluates a monetary billing threshold, the value used excludes taxes but includes discounts and billing credits, so a discount can change when an invoice fires. These details surprise finance teams at exactly the wrong moment.

What Is the Honest Counter Argument?

That pricing discipline is a luxury of companies with leverage. A young company competing against an incumbent with a procurement relationship sometimes has price as its only available lever, and telling it to hold firm is advice from a comfortable position.

We accept that. The qualification is that early discounting should be explicitly framed as buying information and references, with an end date, rather than as a pricing strategy. Discount the first ten customers if you must, know why, and stop.

The failure is not discounting. It is discounting without a rule, then discovering two years later that your list price is fiction and your best customers are subsidising your worst negotiators. That is often bundled up with the pattern we describe in chasing enterprise deals too early.

So What Is the Rule?

Never concede price without naming what you get for it, never let the quarter decide, and write the approval threshold down. If you have given a discount, the contract should say what changed in return.

Then look at your own data once a quarter. If the average discount is drifting up, that is a positioning problem showing up in the pricing column, and no amount of approval process will fix it.

If your published pricing no longer matches what customers pay and you are not sure what the page should say any more, we are happy to help you work out what is true and how to present it. You can find us at phoenix.studio.