Stop Discounting Your Ad Space

A discount is permanent and a bonus is not. Cutting your rate by 20 per cent resets the price that buyer expects for ever, while handing them extra impressions in inventory you were not selling anyway costs you a few dollars and leaves your rate card exactly where it was.

Start by not taking the request personally. A media buyer asking for money off is doing their job, and in a lot of companies they are measured on what they negotiated rather than what they bought. They need a win to take back. Your only decision is which win you hand over.

What does a discount actually cost?

Take a site with a $12.00 CPM on its best placement. An advertiser wants 100,000 impressions a month, which is $1,200, and asks for 20 per cent off.

You can give them 20 per cent off the price, or you can give them 25 per cent more impressions for the same money. Both land the buyer on an effective $9.60 per thousand, so from where they sit the two offers are worth the same. From where you sit they are not remotely the same.

	Bonus credit	Cash discount
What the advertiser pays	$1,200	$960
What the advertiser receives	125,000 impressions	100,000 impressions
Effective price per thousand	$9.60	$9.60
Cash in your account this month	$1,200	$960
What the concession costs you	About $12.50 of fallback earnings	$240 in cash
Your rate card afterwards	Still $12.00	Now $9.60 for this buyer, permanently
What they expect at renewal	The same bonus, which you can withdraw	The same discount, which you cannot

The fallback figure assumes the 25,000 extra impressions would otherwise have earned about 50 cents per thousand from your network. Put your own number in.

The last two rows are the ones that matter. A bonus is an event. A discount is a new price, and you set it without meaning to.

One advertiser, twelve months, both offers worth the same to them.

Nineteen times the cost for an offer the buyer values identically. That ratio is the entire argument.

Why does a discounted rate never go back up?

Because the first price somebody pays becomes the price in their head, and everything afterwards is measured against it. Ask for $12.00 next year and you are not quoting your rate, you are asking for a 25 per cent increase, and you will be arguing about that rather than about the placement.

It travels, too. Niches are small and buyers talk to each other, particularly agencies handling several clients in the same sector. A rate you gave once to get a deal over the line has a way of becoming the rate the next three people open with. Raising it back later is a separate and much harder job, covered in how to raise your ad rates without losing the advertiser.

What should you give instead?

Anything that costs you inventory rather than cash, and anything you can withdraw at the end of the term without a conversation.

What do you actually say in the moment?

The request usually arrives as one line in an email, and the answer is one line back. Something close to this: "I hold the rate card at $12.00 for everyone, so I cannot take 20 per cent off. What I can do is add 25,000 impressions to the month at no extra cost, which puts your effective rate at $9.60. Shall I set it up that way?"

Three things are doing work in that reply. It says no to the rate and yes to the buyer in the same breath, so nobody has to lose. It shows the effective number they were asking for, which is the figure they will report internally. And it ends with a question that is easy to say yes to, rather than leaving the ball in a place where a week of silence can happen. Most buyers take it, because the win they needed was a number, not specifically a discount.

How does bonus credit work in practice?

It is simpler than it sounds, because credit is denominated in money rather than impressions. The advertiser pays you $1,200. You issue them a voucher worth $1,500 of ad credit. They spend it in your portal at your published $12.00, which buys them 125,000 impressions, and your rate card has not moved by a cent.

This is also why the concession is so cheap. The extra 25,000 impressions come out of inventory that was going to fall through to your network anyway, and network fill on most small sites is worth cents rather than dollars per thousand. You are giving away the difference between a few cents and nothing, and calling it a 25 per cent bonus, which it honestly is. The mechanics of issuing credit are on the payments page.

When is the bonus the wrong answer?

When the placement is selling out. If that slot is already delivering every impression it has to paying advertisers, the bonus is not free any more, because every extra impression displaces one somebody else would have bought. At that point a bonus costs you full rate and you have talked yourself into the worse of the two deals.

So check first. If your best placement is running at high fill and you are turning work away, the correct response to a request for money off is not a bonus and not a discount. It is a polite no, and probably a price rise at the next renewal.

When is a discount the right answer?

Three times, and they have something in common: each one buys you something specific that you could not otherwise get.

• An annual prepay. Twelve months of cash up front genuinely removes risk and admin for you. Paying something for that is rational, as long as the discount is written as a prepayment term rather than a new rate.
• The first advertiser in a placement nobody has bought yet. You are buying a reference, a case study and a screenshot to show the next buyer. Say that out loud, agree what you get in return, and put an end date on it.
• Inventory that is about to expire. A slot in the last week of a month is a perishable good. Selling it at half rate beats serving network fill, and no buyer expects a clearance price to be the standing one.

What all three have in common is a reason and a deadline. A discount with neither is not a negotiation, it is a price change you did not decide to make.

And if you find yourself discounting on every single deal, the problem is not your negotiating. Your published rate is fiction, and the fix is a lower rate you can defend without flinching rather than a private exception for whoever asks loudest. That is a pricing job, and it starts with working out what your slots are actually worth instead of what you hoped.