Now scale it up with an example. Say you do $50M a year in direct sales, and your promotions give away 2% more of that than they needed to. That's $1M of margin, assuming the deeper offer didn't bring in enough extra orders to cover it. For a brand that size, that's real budget, the kind that could fund a senior hire or a product launch, handed to people who were already in the cart.
Deep discounts land differently on new customers and loyal ones
So who actually needs the deep offer? This is the part of the research I find most useful, because it gets more specific than "discounts are bad."
Eric Anderson and Duncan Simester ran field tests with a US catalog retailer, sending some customers deeper discounts than others. For first-time customers, deeper discounts increased what they bought later. For established customers, deeper discounts did the opposite and reduced their future purchases. The authors also found that short-run numbers overstated the real gain from discounting to established customers [3].
Think about what that means for a sitewide code. The same 20% that might help win a brand-new buyer is also going to your most loyal customers, and for them it can cost you twice. You give up margin on the order, and you may get less from them afterward.
Customers who might have bought from us earlier, or even at full price, learned to hold off until the next event came around.
I've lived this one. When I was on the brand side, we ran deep promotions around the big sale events every year, and what we ended up with was a behavior we didn't want. Customers who might have bought from us earlier, or even at full price, learned to hold off until the next event came around. It was almost like they were running their own holdout group on us. And that's the cost the research describes. The discount didn't just eat margin on the order, it changed how our best customers bought from us the rest of the year.
That lines up with what we hear from brands, too. In our own conversations with brands this year, one of the margin complaints we heard most was paying for purchases that would have happened anyway.
So something to consider is splitting your offer by customer type. If a deeper offer is going to earn its keep anywhere, it's more likely to be with people who have never bought from you. Your established customers probably need a reason to feel recognized more than they need a bigger percentage.
Points can bleed margin too, so work out what a point costs before you hand them out
This is where a lot of Black Friday loyalty advice goes sideways. The common playbook is to swap the code for double or triple points. That can be smarter, but a point is still a cost. It's a discount you pay later, and your finance team carries it as a liability, deferred revenue, until a member redeems it or it expires.
So do the math on the point first. Here's a simple version. Say a member earns one point per dollar, and 100 points gets them $5 off. That's 5 cents back on every dollar they spend, once they redeem. Run double points over Black Friday and it becomes 10 cents. On a product with a 50% margin, that's a fifth of your gross profit on that sale, paid out later.
That can still be a good trade. A point gives a member a reason to come back and use it, and a code is gone the moment they check out. But it's a cost you chose, and you should know that number, along with its forecasted impact, before you promote it.
You might also set earn rates by product margin, so your thinnest-margin products aren't your most rewarding ones. And please don't build the plan around points going unspent. A program that only works when members forget their balance won't hold up the first time someone in finance asks why redemptions doubled in December.
Four things to offer members besides a deeper discount
-
Give them a bonus they can only use in January. A reward earned over Black Friday that only becomes usable after the holidays gives a member a reason to come back in the quietest stretch of your year, and you only pay it out when they come back.
-
Add value instead of taking money off. A gift with purchase or a bundle costs you only what the item costs you to make. A $10 add-on that costs you $3 can feel bigger to the shopper than a slightly deeper discount, and it protects your price.
-
Run a member-only drop. A limited bundle or product that only members can buy gives people a reason to come to you directly instead of waiting for the sale.
-
Make earned benefits visible while they're still comparing. Show members their balance, their status and what they're close to earning on the product page and in the cart, so the relationship shows up before you ask for another purchase.
If you want to see how these pieces come together in one program, take a look at Rewards. And if you want the longer version of why discounts don't build loyalty on their own, I wrote about that here.
Now, this part gets hard, and you should plan for it
You can't switch discounts off overnight. Your customers expect a Black Friday offer, and your competitors will run one. Pulling yours completely is a big decision, and it's yours to make.
If you sell through retailers, they'll run their own promotions on your product, and you don't set that shelf price. Everything in this piece is about the part you do own, which is your site, your members and the offer you send them.
And you won't know whether member benefits beat the code unless you test it. Try the member offer on part of your audience first, compare gross profit dollars against the people who got your standard promotion, and keep measuring for 90 days after Cyber Monday so you can see what happens once the weekend is over. I go deeper on how to set that up, and how to bring the results to finance, in this post on proving your loyalty program is working.
Before you sign off on this year's promo plan, run one check
Pull last year's Black Friday and Cyber Monday orders that used a code. Then check how many of those customers had already bought from you at full price in the 90 days before.
If that number is high, you paid a discount to people who were likely already going to buy from you, and you should know that before you do it again.
Then bring that number to the meeting where this year's promotion plan gets signed off. Asking who actually needs an offer is a much better question to be working through in October than how deep the discount should go. I put together a BFCM readiness checklist that walks through this step by step, from looking back at last year's data to setting up your test before the weekend. If you'd like a copy, reach out to our team and we'll send it over.
Frequently Asked Questions
How much more do you need to sell to make up for a 20% discount?
It depends on your gross margin. At a 50% margin, a 20% discount means you need to sell 67% more units to earn the same gross profit dollars. At a 40% margin, you need to sell twice as many units just to stay even.
Are loyalty points cheaper than discount codes?
A point isn't automatically cheaper. It's a discount you pay later, and it sits on your books as a liability, deferred revenue, until a member redeems it or it expires. If a member earns one point per dollar and 100 points is worth $5, that's 5 cents on every dollar, and double points makes it 10 cents. Points can still be a better trade because they bring the member back to redeem.
Should new and existing customers get the same Black Friday discount?
The research suggests they respond differently. In field tests with a US catalog retailer, deeper discounts increased future purchases from first-time customers but reduced future purchases from established customers. One option is saving deeper offers for new buyers and giving loyal customers recognition and member benefits instead.
How do you know if a Black Friday offer actually worked?
Test it before you roll it out. Send the offer to part of your audience, compare gross profit dollars against the people who got your standard promotion, and keep measuring for 90 days after Cyber Monday.
Sources
- Adobe, 2025 holiday shopping season results. US online spending and peak discounts off list price by category, November 1 to December 31, 2025 (January 7, 2026).
- Salesforce, Shopping Index, Cyber Week 2025. US online sales and order volume, November 25 to December 1, 2025.
- Eric T. Anderson and Duncan I. Simester, "Long-Run Effects of Promotion Depth on New Versus Established Customers: Three Field Studies," Marketing Science, 2004. Field tests with a US catalog retailer, 1999 to 2001.
About the author
Amanda Boshell is a Product Marketing Manager at TrueLoyal. She spent twelve years as a loyalty practitioner before moving to the platform side, across agency, tech, and retail, in roles spanning program operations, strategy, and marketing. She has sat on both sides of the table, as the operator running the program and as the buyer evaluating the platforms.
