How to Handle a Client Asking for a Discount

The Discount Request That Costs You More Than the Discount

Every price, percentage and payment term below is an illustrative scenario, not a measured result. Swap in your own numbers before you use any of this.

Picture the end of a good project. The work is done, the client loved it, and the invoice goes out:

“Hi — great work, really. Could you do us a favour and knock 20% off? We’ve got a budget cut coming and it would really help us say yes.”

The project is already finished. The work is already good. If you say yes, you have taught this client something important about how you price.

Here is the version that shows up before you sign, and it is more common. You quoted $1,600 for a five-page site. Three days later an email arrives from someone you have never met:

“Hi — thanks for the proposal. Before we go ahead, could you send a breakdown of your costs by role and your hourly rates? And our policy is that every supplier we hire takes 20% off the first engagement. We can also settle 90 days after delivery if that works.”

That email is not a request. It is a price comparison in disguise. The moment you answer it with an hourly breakdown, you have entered a bidding contest with every freelancer in your city — and you have handed over the only reason you were worth more than the cheapest quote.

If you do not understand the rules of this conversation, you will watch your margin get taken apart on every job you touch. The good news is that they are simple, and you can learn them in a week.

---

1. Why the Person Asking Wants a Discount

Never take the discount request personally.

That person is usually not being difficult. They are responding to their own incentive:

• Their measurable result is money saved, or money never spent.
• If you quote $1,600 and they negotiate you down to $1,280, they register a $320 saving they can show whoever signs off.

The person asking for the discount is often not the person who wants the work. They are a proxy for a budget, a manager, or a partner who never opens the site. Keep that in mind before you get defensive.

The rule that follows from it is simple: give the client a real victory, but never take it out of your own ability to do the work properly.

---

2. The Four Rules That Protect Your Margin

How do small studios and independent practitioners hold their price when a client pushes back?

---

3. Rule 1: Do Not Sell Hours

The buyer’s favorite tool is the cost breakdown spreadsheet:

• Frontend developer: $185/hr
• Project manager: $140/hr
• QA: $95/hr

The moment you populate that sheet, you have signed your own death warrant. The person negotiating will benchmark your rates against every other supplier they have heard of and ask you to bring them down to $65/hr. And then you are arguing about hours, the one subject where you are guaranteed to lose.

The response that works

“We quote a fixed price for a defined result, and that price covers the whole thing: the design, the build, the content, the launch and the handover. We don’t bill by the hour, so there is no rate card for you to compare. If the budget needs to come down, the scope changes — and I will show you exactly what comes out.”

By refusing to break your fee into hours, you keep yourself off the list of interchangeable suppliers.

---

4. Rule 2: Be Specific Enough That Nobody Can Compare You

Once the client has decided they want you specifically, you are no longer one line on a comparison spreadsheet. Getting there takes three things:

1. Name what you build. Not “websites” — say that you build static sites deployed to an edge network, where every file is served from a global CDN and there is no database to patch. Cloudflare’s Pages documentation describes exactly that setup: each file of a Pages site is uploaded to Cloudflare’s globally distributed network, and the Free plan is metered in builds per month and files per site.
2. Write your method down. A named diagnostic step — the performance baseline you run before any build, say — that a competitor cannot claim to do because they have never written it down.
3. Put a measurable target in the contract, with the measurement named and a remedy attached: a Lighthouse mobile performance score of 95 or above — our own target on the 0–100 scale Chrome’s Lighthouse documentation describes, where anything from 90 up is rated “Good” — with the remediation work included in the contract if the score falls short. Do not write a guarantee of a perfect score into the contract. Chrome’s documentation states that a score of 100 is extremely challenging to achieve and not expected, and a clause you cannot honour is worse than no clause at all.

When the person who owns the project says that only you have the method and the measurement to hit the outcome, the discount request loses most of its power.

---

5. Rule 3: The Iron Law of Give-to-Get

Here is the hard one. The client comes back and says: “Our budget has a ceiling of $1,280. We cannot approve $1,600.”

The reflex is to cave: “Okay, we can do $1,280 for this project.”

That is fatal. It proves the $1,600 was fake.

The disciplined move is to take something out instead:

“We understand your budget. We don’t discount the work, but we can meet $1,280 by taking two things out of phase one: the interactive scope-modelling tool and the multi-region deployment. That brings us to $1,250, which is under your number.”

Why this works:

• They get the paper victory: total spend went down.
• Your margin on the work that remains is untouched.
• The person who actually wants the site — and who wanted those two features — often steps in and asks for the full $1,600 back.

Keep the arithmetic honest and keep the saving bigger than the discount. Whatever the fee gives up, the scope should give up more. Scope is cheap to cut. Margin is not.

---

6. Rule 4: Get the Owner in the Room

Never let the negotiator be the only voice you hear.

When the negotiation stalls, or they ask for payment 90 days after delivery, write to the person who asked for the site in the first place:

“Hi [Name], we would like to get your site launched for your window. The terms on the table do not work for us — mainly the 90-day payment, which means we carry the project for three months. If we cannot settle standard terms this week, we have to move our next job up the schedule and your launch slips. How would you like to do this?”

Often the owner replies the same afternoon and asks their own team to sign.

---

Conclusion: Hold the Price, Change the Scope

Negotiation is a question of what you are prepared to take off the table.

Know what your fee buys. Sell the result, not the hours. When a discount comes, take something out rather than money off the top, and let the client be the one who chooses.

Hold your price, change your scope, and never give a concession that costs you nothing.

STRUCTURED PROTOCOL // FAQS

Frequently Addressed Technical Inquiries

Why does my client keep pushing for a bigger discount? [+]

Usually the person negotiating was given a budget to hit or a savings number to report, and a discount is the fastest way to show it. That does not make the request unreasonable, but it does mean the conversation is about their goal rather than your cost, so you have to trade something for every concession.

How do I stop a client from treating my work as interchangeable? [+]

Stop selling hours. A price per hour is directly comparable to every other hour on the market, and you lose that comparison the moment you publish it. Quote a fixed price for a defined result, describe the method that gets you there, and let the client compare you on the outcome instead of on an hourly rate.

What is the 'Give-to-Get' rule for discounts? [+]

It means never lowering your price without removing something of equal value in the same move: a feature, a page, a revision round, a faster timeline, or payment terms. A discount given with nothing removed in return signals that your original price was inflated, and it teaches the client to ask for the same cut again.

What if the client simply says their budget is a hard limit? [+]

Accept that the conversation is about their ceiling, not your value, and trade rather than concede. Name the exact pieces you would remove to land under the limit, price that reduced version honestly, and let the person who actually wants the work decide whether the reduced scope is still worth having.

Sources

1. [1] Lighthouse performance scoring | Chrome for Developers Verifies the measurable target this article tells you to put in a contract: Chrome's documentation defines the Lighthouse performance score as a 0–100 scale, rates 90 to 100 as 'Good', and states that a score of 100 is 'extremely challenging to achieve and not expected'.
2. [2] Limits | Cloudflare Pages Verifies the technical claim behind 'static sites deployed to an edge network': the documentation states that Pages uploads each file of a site to Cloudflare's globally distributed network to deliver low-latency pages, and sets the Free plan's limits as 500 builds per month, 20,000 files per site and 100 custom domains per site.