There's a moment that happens in nearly every small business at some point. Things slow down a bit, the calendar looks thin, the inbox is quieter than you'd like. So you do what feels obvious. You run a discount.
Twenty percent off this week. Buy one get one free. A flash sale. A first-time customer offer.
It works, at first. The phone rings. The orders come in. You feel the relief of seeing some movement again.
Then you do the math a few weeks later, and something doesn't quite add up. The volume was up. The revenue was kind of up. But the bank account isn't really fatter. And the busiest day of the week, the day you'd have been busy anyway, is suddenly the day that's giving away the most margin.
This is the quiet truth about discounts. They're one of the most powerful tools in a small business, and one of the most misused.
TL;DR: Most small business discounts lose money because they're not designed to do a specific job. The math is harsher than it looks (a20% discount on a 30% margin cuts profit by two thirds, not by 20%). The fix isa five-step framework: pick the customer first, design the offer for their reality, put a soft deadline on it, track who comes back at full price, and don't run discounts every month. Done right, a single targeted discount is the cheapest customer acquisition cost you'll ever pay.
The Math Behind Why Most Discounts Lose Money
Here's a number that's worth sitting with.
If you're running on a 30% gross margin (typical for a lot of small businesses), and you offer a 20% discount, you don't lose 20% of your profit. You lose much more than that.
Quick example. You sell something for $100 that costs you $70.You make $30 profit. Now you discount to $80. The cost is still $70. You make$10. You haven't lost a fifth of your margin. You've lost two thirds of it.
To get back to where you were, in actual dollars of profit, you need to sell three times as many units. Not 20% more. Three times.
Most discount campaigns don't come close to that lift.
This isn't to say discounts are bad. They're not. It's to say that running them without doing the math is how good businesses quietly bleed.
The Two Jobs a Discount Can Do
Discounts have one of two jobs. Either they're winning a customer you wouldn't otherwise have got, or they're rewarding a customer you've already got. Both are valid. They're just different jobs and they need different setups.
The mistake most small businesses make is running a discount without being clear which job they're trying to do.
A "20% off everything" sale advertised on Instagram tends to do neither well. It pulls in some new people. It also gives the discount to all your regulars who'd have happily paid full price next Tuesday.The result is busier than usual, with thinner margins, and no clear data on whether you actually grew the business.
You might be thinking, "but the regulars will appreciate the discount." Sure. But the discount you give them isn't doing acquisition work. It's a thank-you gift you didn't quite plan as a thank-you gift. There are cheaper ways to thank a regular than handing them 20% off something they'd have bought anyway.
A Discount That Works vs. A Discount That Doesn't
The difference between the two often comes down to six things. Read across the rows, and you can usually tell which side of the table a campaign is sitting on before you launch it:
You can usually tell which side you're running by asking one question. Who exactly is this offer for, and what's the moment they're in?
How to Run a Discount That Brings Customers Back
A discount is doing its job when the person you offered it to comes back, at full price, more than once. That's the whole game. Get them in, give them a reason to like you, then let the relationship pay you back over the following months.
Get this right and a 20% first-time discount is the cheapest acquisition cost you'll ever pay. Get it wrong and it's a one-off transaction at a thin margin that you'll never see again.
Here's a five-step framework that works for most small businesses.
Step 1: Pick the customer, not the offer
Most owners start with the offer. "20% off this week." Then they advertise it as broadly as possible.
Try flipping it.
Start with the customer you actually want. Someone new to the area? Someone who used you once and never came back? Someone whose friend already buys from you? The offer that gets a new neighbour through the door is different from the one that wakes up a lapsed customer. And both are different from the one that thanks a regular.
Three customer groups, three discounts. Not one blanket offer that hits everyone.
Step 2: Make the offer specific to that customer's reality
A new customer doesn't know what you're like yet. The discount is paying for the trial. Something like "first visit, 20% off" or"first booking, free [add-on]" makes the path to trying you for the first time as low risk as possible.
A lapsed customer already knows what you're like. They just haven't been back. The discount here is paying for the reminder, not the trial."We've not seen you in a while, here's $10 off your next order if you book by Friday" works because the friction is small and the deadline is real.
A regular doesn't really need a discount, they need a thank you. A surprise upgrade, a small extra in the order, an early heads up about something new. Something that says "we noticed." Money off feels weirdly impersonal here. A small gift feels right.
Step 3: Put a deadline on it, but make it gentle
A discount with no deadline gets ignored. A discount with a deadline that feels manipulative (BUY NOW, ONLY 17 SECONDS LEFT) puts people off.
The middle ground is a soft deadline. "Valid until end of the month." "Until Sunday at five." "First ten bookings only." Real, but not aggressive.
Step 4: Track whether it actually worked
This is the bit nine out of ten small businesses skip, and it's where the real learning happens.
You don't need fancy analytics. You need to know two things:
• How many people used the offer
• How many of those people came back, at full price, in the following few months
That's it. If you ran a "first booking, 20% off"campaign and got 30 takers, and four of them came back at full price within ninety days, you can do the math on whether the discount paid for itself. If none of them came back, that's important information too.
A simple spreadsheet is enough. Date, offer, number of takers, number of repeat customers. After three or four campaigns, you'll have a much better feel for what works.
Step 5: Don't run them all the time
The fastest way to train customers to wait for discounts is to run discounts every month.
Once they know you'll go on sale every six weeks, they'll happily sit on their hands and wait. The full price you charge in between starts to feel a bit theoretical.
Two or three a year is plenty for most small businesses. One around a launch or a season. One for lapsed customers. One thank-you to regulars.
What This Looks Like in Practice
Here's the kind of story that plays out when an owner makes the shift, in composite form:
Jasmine runs a small boutique in Austin. For three years she'd been running "20% off everything" sales every six to eight weeks.Each sale brought in volume. The phone rang. The store got busier. But her bank account never quite reflected how much she felt like she was selling.
She tried something different last spring. Instead of one blanket sale, she split her customer list into three groups. New visitors who'd been in once but never bought. Lapsed customers who hadn't shopped in three or more months. Regulars who came in monthly. She sent each group a different offer: a first-purchase incentive for the visitors, a "we've missed you" $15-off note for the lapsed, and a small free-gift thank-you for the regulars. Then she tracked who came back at full price over the next quarter.
The results changed how she thought about discounts. The lapsed-customer email earned its keep within a few weeks. The first-purchase offer was slower to pay back, but it produced two new repeat customers who became monthlies. The regulars who got a free gift sent more referrals than usual.
She stopped running blanket sales after that. Two or three targeted campaigns a year. More margin per dollar of revenue, the same grow that the top line.
This is a composite based on common patterns we see when owners shift from blanket discounts to targeted ones, not a specific named customer.
A Few Traps to Avoid
Don't discount your top sellers. The thing customers already love at full price doesn't need help.
Don't compete on price as a default position. There's almost always a bigger competitor who can go lower than you. Compete on something else (service, specialty, speed, story), then use the occasional sharp offer to pull people in.
Don't bundle a discount with a confusing voucher code. The simpler the offer, the higher the take-up. "Show this text at the counter" beats "Use code SPRING25 at checkout" almost every time.
Don't discount when you're feeling panicked. The discount you run from anxiety almost always loses money. If the calendar's thin, run a targeted lapsed-customer campaign instead of an everyone-gets-20%-off sale.
Where Mighty Fits In
One last thing before you go and rework your next sale.Discounts are a margin question dressed up as a marketing question. How much you can afford to give away depends on how much margin you've got in the first place. If your costs are quietly tighter than they need to be, your discount math gets thin in a hurry. If your costs are sharper, the same headline offer earns its keep.
That's where Mighty comes in. Mighty connects 120,000+ small businesses with exclusive deals and tools from a network of trusted B2B partners, including Amazon Business, across every category your business actually spends in. Members have saved over $36M between them so far. And every member gets an Authorized Mighty Rep, a real person who comes to your business in person to help you set everything up.
Sharper costs on the supply side give you real flexibility on the discount side. You can afford to acquire customers, reward regulars, and fill quiet days, without your margin getting thin.
Frequently Asked Questions
How much of a discount can a small business actually afford to offer?
It depends on your gross margin. On a 30% gross margin, a 20%discount cuts your profit by roughly two thirds, not by 20%. The general rule:work out your profit before the discount, then work out your profit after, and only run the offer if the volume lift gets you to where you'd have been, or better. Most blanket discounts don't.
When does a discount make sense for a small business?
Discounts work best in specific situations: acquiring a first-time customer where the lifetime value covers the cost, waking up a lapsed customer who knows you but hasn't been back, filling a slow shift where the marginal cost is near zero, or running a one-off bundle that adds value without cutting the headline price. They work poorly as a blanket "feeling quiet, run a sale" response.
How do I run a discount that brings customers back?
Pick the customer first (new, lapsed, or regular), design the offer for their specific situation, put a gentle but real deadline on it, track how many of them return at full price within ninety days, and don't run them so often that customers start waiting for the next one. Three campaigns a year is plenty for most small businesses.
Should I discount my best-selling products?
No. Your best-selling products are already converting at full price. Discounting them gives margin away on revenue you'd have earned anyway.Use discounts to introduce slower-moving items, win first-time customers, or wake up lapsed ones, not to push what's already pushing itself.
How often should a small business run discounts?
Two to three times a year is plenty for most small businesses.Once around a launch or a season. Once for lapsed customers. Once as a thank-you to regulars. Running them more often than that quietly trains customers to wait for the next one, and the full price you charge in between starts to feel theoretical.
What's the best way to measure if a discount worked?
Two numbers in a simple spreadsheet: how many people used the offer, and how many of those came back at full price within ninety days. If ameaningful share of takers (say 10-30%) become repeat customers, the discount paid for itself. If almost nobody came back, you just gave away margin topeople who weren't going to stick around anyway.
How do I avoid training customers to wait for sales?
Run discounts rarely, target them at specific customer groups rather than blanket-advertising them, and don't put them on a predictable cadence (e.g. "the first Friday of every month"). The aim is for customers to assume your full price is the price, with occasional moments where they qualify for something specific.
What's the best discount for new customers?
A first-purchase incentive that's generous enough to make trying you for the first time feel low-risk, but specific enough that you cantrack who used it. "First visit, 20% off" or "First booking, free [add-on]" both work. The goal is the second visit at full price, not the first one at a discount.
What's the best discount for lapsed customers?
A specific dollar amount off, not a percentage. "$10 off your next order if you book by Friday" feels more concrete and time-bound than "20% off your next order." Pair it with a one-line "we've missed you" personal note rather than a marketing-style email. The friction is small, the deadline is real, the message is human.
Is it better to discount or to add a free bonus to the order?
Often the bonus, especially with new customers. A bundle or free add-on lets you protect the headline price (so future regulars still seethe full number on their receipts) while giving the customer a clear win."Free [item] with your first order" tends to land better than"20% off your first order" for the same margin cost to you.
The next time business feels quiet, don't reach for a sale on instinct. Ask yourself who the offer is for. Make sure the offer fits thatperson's reality. Put a soft deadline on it. Track who comes back. Then goagain, three or four months later, with a different customer in mind.
That's the whole framework. Boring, repeatable, and quietly more profitable than the version most businesses run.