Somewhere right now, a marketer has just been asked to put together a promo for this quarter. With no brief, no target segment, and a deadline of Friday. The result is almost always the same: a discount code sent to the whole list, a shy bump in orders, and a revenue number that looks fine until someone starts asking questions.
This guide is for that person. It's not a catalog of promotion formats, there are plenty of those. Instead, each example below shows what the promotion actually was, why the economics made sense, and what result it produced.
A sales promotion is a time-limited incentive designed to drive a specific customer action, like a first purchase, a repeat order, a referral, a reactivation. The incentive can be a coupon, a free gift, loyalty points, or anything else that makes the action more attractive.
What separates a good sales promotion from a bad is whether the incentive went to the right person, at the right moment, with the right constraints to protect margin. Every example below did at least two of those three things well.
After 10+ years working with world-leadinf brands on incentives, we have a fair share of examples of the best-performing incenitve sout there, here are some of the best:
What it is: A first-purchase incentive sent to newly registered users who haven't yet bought. Common formats include a percentage off, free shipping, or bonus points on the first order.
Why the economics work: Registration is the highest-intent moment in the acquisition funnel. The customer has already done the research and made a decision to engage. The welcome offer reduces the final friction of handing over money to a brand they don't fully trust yet. The key is that you're not discounting customers who'd buy anyway, but converting customers who are genuinely on the fence.
Real example: Whoppah, a Dutch marketplace for second-hand furniture, launched a targeted welcome promotion aimed specifically at recently registered users who hadn't purchased. They saw a 35% conversion lift within 14 days.
What to do differently: Don't send welcome offers to your whole email list. Send them to the segment who registered in the last 14 days and haven't converted. The redemption rate will be lower in absolute terms and much higher where it matters.
What it is: A triggered incentive sent hours after a customer abandons a cart without purchasing. Usually a small discount or free shipping offer.
Why the economics work: Cart abandonment rates sit around 70% across ecommerce. Most of those customers walked away because of uncertainty, distraction, or friction. A cart abandonment offer gives them a concrete reason to finish a decision they'd already mostly made. The timing gap matters: send it immediately and you've trained customers to abandon carts on purpose. Send it after a few hours and it reads as a genuine second chance.
Real example: ecoATM, a kiosk-based electronics buyback network, ran a cart abandonment promotion and recorded a 20% conversion uplift. For a business with 7,000+ kiosk integrations, that's a meaningful revenue line from a campaign that costs nothing to run once it's configured.
What to do differently: Test 2-4 hour delays before sending. If you send immediately, you're subsidising intentional abandonment, not recovering genuine hesitation.
What it is: A targeted incentive triggered by a negative customer experience: a late delivery, a support complaint, a failed transaction. The offer is a response to something that went wrong, not a campaign sent to everyone.
Why the economics work: The timing is everything here. A customer who just had a bad experience is actively reconsidering whether to stay. The cost of winning them back at that exact moment is a fraction of what it costs to acquire a replacement customer. A recovery offer sent three weeks later is just a discount. Sent within 24 hours of the incident, it's a signal that the brand is paying attention.
Real example: Apology discounts tend to be a drag on support teams. Altitude Sports, a Canadian outdoor retailer, automated their recovery workflow through Voucherify so that specific circumstances fire a personalised offer automatically, with minimum support agent involvement, achieving 75% faster sorry coupon generation enabled by Vincent.
What to do differently: Before you can automate recovery offers, you need the triggers wired up (support ticket status, delivery failure events, NPS response). That's the ops work that makes the whole thing possible. Start with one trigger (late delivery is usually the easiest) and get the workflow clean before adding more.
What it is: Single-use codes generated per customer or per segment, as opposed to one public code broadcast everywhere.
Why the economics work: A public code like SAVE20 shared on Reddit reaches everyone who wants a discount, including customers who'd have paid full price and competitors who want to monitor your pricing. Unique codes flip this: each code has a specific owner, redemption rules, and can expire. The discount only reaches who you intended.
Real example: Subito, an Italian online marketplace, tested promo codes across different customer segments and found that unique codes increased the conversion rate on their paid visibility options by 37%.
What to do differently: If you're still using public codes for campaigns that should be targeted, switch. The cost of generating unique codes is negligible, while the cost of broadcasting a public discount to people who didn't need it compounds every campaign.
What it is: A time-limited discount with a genuinely hard end date, usually 24-72 hours.
Why the economics work: Urgency changes buyer behavior in a specific way: it forces a decision that would otherwise be deferred indefinitely. "I'll come back to this" is the default for most browsing behavior. A flash sale with a real deadline removes the option to defer. The critical word is "real". A sale that resets every 72 hours trains customers to ignore the timer. The urgency only works if it's genuine.
What to do differently: Run flash sales selectively and infrequently. The scarcity signal degrades fast if customers learn to expect them on a cycle.
What it is: A promotion restricted to a specific group: healthcare workers, students, loyalty members, trade buyers, employees of a partner company. Access is gated by eligibility criteria rather than a public discount.
Why the economics work: Exclusivity changes how a discount feels to the customer. Being recognised as part of a specific group (NHS workers, students, members) is a different emotional experience than receiving a wide offer. The targeting also means margin stays tighter: you're not funding discounts for customers who don't qualify. The challenge is verification, which is why gating by email domain, loyalty tier, or location tends to work better than an honour system.
Real example: Beryl, a UK bike-sharing company, used to offer free rides to NHS workers and key workers during the pandemic, gating access by email domain and location. They also ran a 25% student discounts gated by student email domains.
.webp)
What to do differently: Gated promos don't need to be one-size-fits-all within the eligible group.
What it is: Buy-one-get-one or discounted bundles of complementary products. The customer gets more for their money, while the brand moves more volume.
Why the economics work: BOGO and bundling shift the customer's reference point from "how much does this cost?" to "how much am I getting?" That's a meaningful difference. A 50% discount on a single item and a BOGO deal on the same item cost the same in margin terms, but BOGO drives significantly higher perceived value. Bundling works on the same principle: the bundle feels like a deal the customer couldn't assemble themselves at the same price, which makes the total spend feel justified even as it increases.
What to do differently: The product pairing logic matters more than folks realise. BOGO on complementary products reads as a benefit, while BOGO on unrelated things reads as clearance, desperate one at that.
What it is: Free shipping unlocked above a minimum order value. Usually displayed as a progress bar.
Why the economics work: Shipping costs are one of the top reasons customers abandon checkout, because delivery feels like a penalty for buying. A free shipping promotion removes that feeling while also increasing AOV: customers who are €8 away from free shipping will often add something to the cart rather than pay the fee. The threshold also gives you a way to increase AOV without a blanket discount.
What to do differently: Test the threshold against your actual AOV distribution. If most orders are already above the threshold, it's not driving incremental behaviour, it's just a cost. The threshold needs to be slightly above the natural order size to change behaviour.
What it is: A promotion where customers get a percentage of their spend returned as store credit, cash, or a future voucher. The reward is deferred: you buy now, you earn back later.
Why the economics work: Cashback sits in an interesting position compared to a straight discount. A 10% upfront discount and a 10% cashback offer cost the same on paper, but they behave differently in practice. The deferred reward means not all customers redeem, some forget, some churn before using it, which lowers the actual cost below the headline rate. More importantly, cashback tied to a future purchase creates a reason to come back that a straight discount doesn't. The customer has store credit sitting there and walking away means leaving cash behind and nobody likes that.
What to do differently: Set cashback expiry dates short enough that customers act on the reward rather than forgetting it. 30 days converts better than 90 days, even though 90 days sounds more generous. A cashback offer that expires before it's used helps no one.
What it is: A targeted offer sent to customers who haven't purchased in a defined window. Usually personalised to their previous purchase behaviour.
Why the economics work: Dormant customers already know your brand and have purchased before. The cost of reactivating them is lower than acquiring a new customer from scratch, but only if the offer is relevant. A generic "we miss you, here's 15% off" email competes with every other win-back email in the inbox. An offer tied to the category a customer last bought from or the brand they've purchased most converts at a higher rate because it's answering the right question.
Real example: Blacklane, a global chauffeur service, faced a specific version of this problem. Every trip is a standalone decision as there's no natural purchase cadence forcing a next booking. So they built a consecutive journey mechanic: guests who completed one leg of a trip unlocked an incentive for the next, pulling them through a multi-step sequence rather than hoping a single discount would stick. The result was a double-digit increase in average revenue per guest.
.webp)
Ioana Pascaru, their Senior CRM Manager, put it well: "A single voucher is a transaction. A sequenced journey is a relationship. That shift changed how we think about the entire role of rewards at Blacklane."
What to do differently: Segment your dormant list before you send. Customers dormant for 60 days need a different message than customers dormant for 180 days. The longer the gap, the stronger the reason to return needs to be.
What it is: Discount tiers that unlock at higher spend levels. Spend €50 and get 10% off. Spend €100 and get 15% off. Spend €150 and get 20% off. And so on.
Why the economics work: Tiered discounts change the cart abandonment calculation in a useful way. A customer with €85 in their cart, looking at a 15% discount threshold at €100, will often add €15 more rather than leave. The additional revenue tends to outweigh the additional discount cost, especially if the threshold is set just above the natural AOV. It also avoids discounting customers who'd have spent the lower amount anyway.
Real example: Munhowen, a beverages retailer, saw a 70% lift in AOVs after implementing cart-level promotion mechanics.
What to do differently: Run an analysis on your AOV before setting thresholds. The tiers should sit at points where a meaningful number of customers are just below the threshold, otherwise you're giving a discount to people who'd have hit the number anyway.
What it is: A win-back campaign where the offer is tied to each customer's specific history rather than a flat discount across the segment.
Why the economics work: Generic win-back offers (10% off, come back) work to some degree. Personalised ones work better, and at lower cost. If you know a customer bought cycling gear three times but never bought apparel, an offer tied to cycling gear converts at a higher rate than a general 10% off. More importantly, personalisation lets you calibrate the offer size: customers who only buy on discount need a different incentive size than customers who stopped buying for other reasons.
Real example: CarParts.com used personalised win-back incentives, improving customer engagement by 35%.
What to do differently: Before running a win-back campaign, segment by likely reason for churn. Price-sensitive customers need a discount. Customers who left due to a bad experience need an apology and a reason to trust you again. Mixing them into one campaign wastes budget on the wrong offer for each group.
What it is: A promotion built around a game mechanic (spin-to-win, mystery discount, scratch card, challenge completion). The incentive isn't just the reward, it's the act of earning it.
Why the economics work: Gamification works because it adds engagement to what would otherwise be a passive transaction. A customer who spins a wheel has invested attention in the promotion, they're more likely to redeem the reward they won than a flat discount they received by email. Mystery discounts exploit curiosity: customers open to find out what they got, which drives higher email open rates and redemption rates than known-value offers.
What to do differently: Control the probability distribution. If your highest-value prize (a free product, a 50% discount) is too likely, the campaign destroys margin. Set the odds per prize tier before launch, not after the first redemption wave.
What it is: An automatically triggered incentive sent on a customer's birthday or their anniversary as a customer.
Why the economics work: Birthday offers have the highest open and redemption rates of almost any promotional email format, and the reason is straightforward: it's the one communication that's clearly about the customer rather than about the brand's revenue targets. The personalisation cost is zero once the trigger is set up. The trick is timing: send too early and it feels like a data grab, too late and the moment is gone. Most brands send a few days before the birthday with a 7-day redemption window.
What to do differently: Add an anniversary trigger alongside the birthday. A customer who's been with you for a year is a more valuable customer to celebrate than one who just gave you their birthday in a sign-up form. The message lands differently: "you've been with us a year" signals that you're paying attention in a way that "happy birthday" (which every brand sends) doesn't.
What it is: A targeted incentive offered to existing customers to drive a purchase in a category they haven't tried yet. The trigger is a completed purchase in one category, the offer is a reason to try another.
Why the economics work: This is one of the better-kept secrets in promotional economics. A cross-sell offer looks like a discount, but it's actually an acquisition campaign: you're acquiring a customer for a new product line. The difference is the starting point. A cold acquisition campaign has to clear every barrier: awareness, trust, intent. A cross-sell offer starts with a customer who already trusts you, already has payment details on file, and has already demonstrated intent to spend. The incremental CAC for a cross-sell is usually a fraction of what you'd pay to acquire the same customer from scratch via paid channels.
Real example: ekar, a car-sharing service in the UAE, ran cross-category promotions targeting customers who had used one service type to drive bookings in others. They recorded a 15% cross-category booking boost alongside a 40% reduction in overall customer acquisition costs. The mechanics: offer eligibility tied to prior category behaviour, delivered through Braze at the right moment in the customer journey were what made it measurable and repeatable.
What to do differently: Cross-sell promotions fail when the offer is generic. They work when the offer is triggered by a specific prior behaviour and tied to a logical next step. A customer who just booked an airport transfer is a good candidate for an in-city ride offer.
Choosing the right promotion type is simple – match the promotion mechanic to the specific behaviour you want to drive.
The question to ask before any campaign: would this customer have bought anyway? If yes, the promotion is a margin cost, not an acquisition cost. The best promotions reach customers who genuinely needed a reason to act.
.svg){kind=icon}
