On this page
Most promotion roundups show you a screenshot of a campaign someone else ran, and leave you to rebuild the useful part yourself.
This one is organised the other way round. Twenty-five promotion types, grouped by the job they do. Thirteen of them come with an OnVoard template you can open and install rather than recreate. The other twelve are explained but ship no template, and every example says which it is.
The templates come in a version for most categories, so the fashion example is a fashion popup and the coffee one is a coffee popup. Screenshots below are OnVoard templates. None of them are other brands' campaigns, and there are no conversion figures attached to any of them, because a screenshot proves what was on the screen and nothing about what it earned.
What a sales promotion actually is
A sales promotion is a temporary change to the value of an offer, made to shift behaviour now rather than to reposition your prices permanently. The two words doing the work are temporary and behaviour.
Temporary is what separates a promotion from a price cut. If the offer never ends, it is not a promotion, it is your price, and shoppers will treat it that way. Behaviour is what separates a good promotion from an expensive one: every promotion should be trying to cause something specific, whether that is a first order, a bigger basket, a phone number, or a second purchase from someone who bought once and left.
The mistake worth avoiding early is running a discount with no stated behaviour attached. A percentage applied to everyone at all times costs the most and changes the least, because most of it goes to people who had already decided to buy.
How to choose: goal, trigger, ask, incentive
Every promotion in this article is a combination of four decisions. Get these right and the specific mechanic matters less than you would expect.
The goal is the behaviour you are buying. The trigger is the moment you interrupt. The ask is what the visitor has to give. The incentive is what they get. A mismatch between any two of them is the usual reason a promotion underperforms, and it is easier to spot in a table than in a description.
Promotion | Fires | Asks for | Gives | Length |
|---|---|---|---|---|
The first-order discount | On page load | Email and phone | Discount code | 3 steps |
Free shipping instead of a discount | On page load | Email and phone | Discount code | 3 steps |
The minimum-spend threshold | No trigger set | No incentive | 2 steps | |
Early access instead of money off | On page load | Email and phone | Discount code | 3 steps |
The scratch card | On page load | Email and phone | Game prize | 4 steps |
Spin to win | On page load | Email and phone | Game prize | 4 steps |
Pick a gift | On page load | Email and phone | Game prize | 4 steps |
The mystery discount on the way out | On exit intent | Email and phone | Game prize | 4 steps |
The exit-intent save | On exit intent | Email and phone | Discount code | 3 steps |
The flash sale window | On exit intent | Discount code | 3 steps | |
Abandonment recovery on desktop | On exit intent | No incentive | 3 steps | |
The SMS signup incentive | On page load | Email and phone | Discount code | 3 steps |
The birthday club | On page load | Email and phone | No incentive | 4 steps |
Read that table as a set of trades. A bigger ask needs a bigger incentive. A later trigger needs a smaller ask, because a visitor deep in checkout has less patience for a form. An arrival trigger reaches everyone and therefore spends the most, and an exit trigger reaches only sessions that were about to end and therefore spends the least.
If you are running your first promotion, start at the bottom of that list rather than the top. An exit offer cannot annoy anyone who is still shopping, so it is the cheapest way to learn whether your audience responds to a discount at all.
Promotions that win the first order
1. The first-order discount
What it does. A percentage off the first order in exchange for an email address and a phone number, shown on arrival across three steps.
Why the design matters. The three steps matter more than the percentage. Asking for the discount, the email and the phone on one screen is three decisions at once, from a visitor who has not decided they like you yet. Splitting them means the visitor commits to the offer first and fills in contact details second, once the offer is already theirs.
When to use it. Use it when most of your traffic is new and you have nothing else to trade. It is the default opening promotion for a store with a thin list.
When not to. Skip it if a large share of your visitors are returning customers. They will take a first-order discount on an order they were going to place anyway, and you have paid for a sale you already had.
What to change for your store. Change the percentage before you change anything else. If your margin cannot carry 15 percent, a smaller number with a clearer reason reads better than a big number you have to claw back through shipping.
The trade-off. It trains people to wait for the popup. Every store running one has some share of visitors who close it, browse, and come back for it later.
OnVoard template: Supplements Welcome Offer. Install it as it is, or change the numbers to suit your margins.
2. Free shipping instead of a discount
What it does. Covers shipping on the first order rather than cutting the product price, again in exchange for an email or phone number.
Why the design matters. Shipping is the cost shoppers resent most, because it arrives late and feels like a penalty for having decided to buy. Removing it reads as a bigger concession than the same money off the product, and it leaves your price list untouched.
When to use it. Use it when your margin is tight or your prices are already positioned carefully. A brand that never discounts can still absorb postage without contradicting itself.
When not to. Not for heavy or bulky goods where postage is a real fraction of the order, and not where you already offer free shipping over a threshold, because then it promises nothing.
What to change for your store. Decide whether it applies to the first order only or to any order, and say which on the popup. Ambiguity here produces support tickets.
The trade-off. You absorb a real cost per order rather than a percentage, so a small basket hurts more than a large one. It is the opposite risk profile to a percentage discount.
OnVoard template: Outdoor Free Shipping. Install it as it is, or change the numbers to suit your margins.
3. The minimum-spend threshold
What it does. Offers 20 percent off a first purchase above a spend floor, in two steps on a fullscreen layout.
Why the design matters. The threshold changes what the promotion does. A flat discount lowers the price of whatever was already in the basket; a threshold gives the shopper a reason to add one more item to reach it. Same money, different behaviour.
When to use it. Use it when your average order sits just below a natural round number and one more item is a realistic ask.
When not to. Not when you sell one expensive thing at a time. If the typical order is a single item, a threshold above it just reads as a discount nobody qualifies for.
What to change for your store. Set the floor from your actual average order value, not from a round number that looks tidy. A floor far above where people normally land is worse than no promotion, because it advertises a reward the visitor cannot reach.
The trade-off. Fullscreen is the most interruptive surface in the set. It earns attention and it costs goodwill, so it suits arrival better than mid-browse.
OnVoard template: Email Capture Full Screen. Install it as it is, or change the numbers to suit your margins.
4. Early access instead of money off
What it does. Trades a place at the front of the queue for new drops and limited runs, rather than a discount.
Why the design matters. It is the one offer on this list that costs nothing and does not touch your prices. What it trades is sequence, and for a brand whose products sell out, sequence is worth more to the buyer than a few percent.
When to use it. Use it when you genuinely release things that sell out, or run limited quantities. Scarcity has to be real for this to work.
When not to. Do not run it on an evergreen catalogue that never sells out. The promise is checkable and visitors will check it.
What to change for your store. Say what they get early and how often. "Early access" with no cadence is a newsletter signup wearing a better jacket.
The trade-off. Fewer people will take it than would take money off, and I would expect the ones who do to be worth more. Judge it on what the list does later rather than on the popup itself.
OnVoard template: Fashion Early Access. Install it as it is, or change the numbers to suit your margins.
Promotions that turn the discount into a game
5. The scratch card
What it does. A four-step flow where the visitor shares an email or phone number, then scratches a card to reveal the discount they have won.
Why the design matters. The reveal is doing the work. A visitor who has scratched the card has taken an action to get the code, and a code you uncovered feels different from a code that was printed on a banner, even when the number is identical.
When to use it. Use it where the mood of the category can carry it. It fits food, supplements, beauty and gifting. It reads as fun rather than cheap when the product itself is a small pleasure.
When not to. Not for considered, expensive or professional purchases. A game in front of a serious purchase reads as a distraction from a decision the buyer is trying to make carefully.
What to change for your store. Decide the prize range and whether every card wins. Both are legitimate, but a card that mostly loses will be recognised as theatre quickly.
The trade-off. It is four steps rather than three, so more visitors leave partway through than with a plain offer. You are trading completion rate for engagement from the people who finish.
OnVoard template: Supplements Scratch Card. Install it as it is, or change the numbers to suit your margins.
6. Spin to win
What it does. The same four-step trade with a spinning wheel as the reveal instead of a card.
Why the design matters. The wheel shows the whole prize set before the spin. The visitor sees what could be won, which raises the perceived value of taking part, and it makes a modest top prize look generous next to the smaller ones beside it.
When to use it. Use it during a promotional window when a bigger prize is genuinely on the wheel. It is at its best when you have something worth winning.
When not to. Not as an always-on default. A wheel that has been on the homepage for a year stops being a game and becomes furniture.
What to change for your store. Set the segments so the common outcome is one you are happy to give every visitor, because that is what most people will get.
The trade-off. It is the most visually loud item in the catalogue. On a restrained brand it will be the least on-brand thing on the site.
OnVoard template: Supplements Spin Wheel. Install it as it is, or change the numbers to suit your margins.
7. Pick a gift
What it does. The visitor chooses between mystery gifts rather than being assigned a prize.
Why the design matters. Choosing is not the same as receiving. Picking one of three boxes makes the reward feel selected rather than handed out, and it sidesteps the disappointment of a wheel landing on the smallest slice.
When to use it. Use it when your prizes are genuinely different in kind rather than in size. Choice needs something to choose between.
When not to. Not when every option is the same discount wearing a different label. Visitors work that out on the second visit.
What to change for your store. Make the options mean something: a sample, free postage and a discount are three different offers. Three discounts are one offer drawn three times.
The trade-off. Setting it up takes longer, because you need several rewards that are all acceptable to you rather than one.
OnVoard template: Outdoor Pick a Gift. Install it as it is, or change the numbers to suit your margins.
8. The mystery discount on the way out
What it does. A scratch reveal on a fullscreen takeover, fired on exit intent rather than on arrival.
Why the design matters. Curiosity works better on someone leaving than on someone arriving. On arrival a hidden number is one more obstacle; on the way out it is the only reason left to stay, because the alternative is closing the tab.
When to use it. Use it as the last offer on a session, not the first.
When not to. Never alongside an arrival popup in the same session. Two interruptions in one visit is the point where visitors stop reading either of them.
What to change for your store. Keep the reveal fast. An exit-intent visitor has already decided to leave and will not wait through an animation.
The trade-off. Fullscreen on exit is the most aggressive combination available here. It converts, and it is the one most likely to be remembered as an annoyance.
OnVoard template: Mystery Scratch Takeover. Install it as it is, or change the numbers to suit your margins.
Promotions that save a visitor who is leaving
9. The exit-intent save
What it does. Offers 10 percent for an email or phone number when the cursor heads for the browser chrome.
Why the design matters. It only spends money on people who were leaving. An arrival discount is given to everyone including the visitors who would have bought anyway; an exit offer is targeted by behaviour, so the discount goes to the sessions that were otherwise worth nothing.
When to use it. Use it as the safest first promotion on a store that has never run one. It interrupts nobody who is still shopping.
When not to. Exit intent needs a cursor, so treat it as a desktop mechanism. On phones there is no equivalent signal and it will fire at the wrong moment or not at all.
What to change for your store. Keep this discount smaller than your arrival offer if you run both. The visitor who sees the arrival one and leaves anyway should not be rewarded with more.
The trade-off. It reaches a smaller share of your traffic than anything fired on page load. That is the mechanism working, not a fault to tune away.
OnVoard template: Coffee Exit Intent. Install it as it is, or change the numbers to suit your margins.
10. The flash sale window
What it does. A 20 percent limited-time coupon for an email, shown on exit intent.
Why the design matters. The deadline is the offer. A discount with no end date is a price change, and the shopper can take it whenever they like, which usually means later or never. A window converts a decision they were postponing into one they have to make now.
When to use it. Use it for a real window: a weekend, a launch, an end of season. The date has to be true.
When not to. Not on a rolling basis. A flash sale that never ends teaches your list that your deadlines are decorative, and that lesson is expensive to unlearn.
What to change for your store. State when it ends in the popup itself rather than only in the email that follows. The urgency has to be visible at the moment of the ask.
The trade-off. Deadlines pull sales forward from next week as well as creating new ones. Judge the window against a normal period, not against zero.
OnVoard template: Jewelry Flash Sale. Install it as it is, or change the numbers to suit your margins.
11. Abandonment recovery on desktop
What it does. A 20 percent offer for an email, fired on exit, and built for desktop only.
Why the design matters. It is honest about where it works, which is unusual. Rather than degrade quietly on a phone, it declares the surface it is built for, so you can pair it with something else on mobile instead of assuming coverage you do not have.
When to use it. Use it on a store where desktop still carries a meaningful share of sessions, and where you can name what mobile gets instead.
When not to. Not as your only recovery mechanism if most of your traffic is mobile. Check your own split before assuming this covers your audience.
What to change for your store. Pair it with a timed or scroll-triggered offer on mobile so the two together cover the whole audience.
The trade-off. You are running two promotions to cover one intention, and they have to agree with each other on the number.
OnVoard template: Stop Abandonment (Desktop Only). Install it as it is, or change the numbers to suit your margins.
Promotions that ask for more than an email address
12. The SMS signup incentive
What it does. Offers 15 percent for a phone number as well as an email, in exchange for alerts about new arrivals and restocks.
Why the design matters. It names what the messages will be about. A phone number is a bigger ask than an email and the visitor knows it, so the trade has to be specific. "Alerts about new arrivals and restocks" is a promise; "updates" is not.
When to use it. Use it when you actually have restocks or drops to announce. The value of the list is the reason to message it.
When not to. Do not collect phone numbers you have no plan to use. Messaging consent rules are stricter than email and an unused list is a liability rather than an asset.
What to change for your store. Set the expected frequency in the copy. Unsubscribes on messaging come from surprise more than from volume.
The trade-off. Fewer visitors will complete it than an email-only ask, and the ones who do are reachable in a channel almost everyone opens.
OnVoard template: Home Decor SMS Signup. Install it as it is, or change the numbers to suit your margins.
13. The birthday club
What it does. Collects a date alongside contact details in exchange for something on the day, with no discount at signup.
Why the design matters. It buys a reason to message later instead of a sale now. Every other promotion here spends margin at the moment of capture; this one spends nothing at signup and creates a dated, personal reason to appear in the inbox months afterwards.
When to use it. Use it where a birthday is a real occasion for the category: gifts, children, beauty, food.
When not to. Not for categories with no occasion logic, where the date is just an extra field that lowers completion for nothing.
What to change for your store. Decide what actually arrives on the day, and make it worth the wait. A club that sends a standard discount code is a newsletter with a delay.
The trade-off. The payoff is months away, so it will look like your worst-performing promotion for a full quarter before it looks like one of the better ones.
OnVoard template: Baby Birthday Club. Install it as it is, or change the numbers to suit your margins.
Twelve more promotions, with no OnVoard template
These are worth knowing and worth running. They are not on-site messages, so there is no template to install, and saying so is more useful than pretending otherwise.
A fixed amount off
Ten pounds off rather than ten percent off. The fixed version is easier to judge at a glance and it behaves differently across your catalogue: on a cheap item it is a deep discount, on an expensive one it is a rounding error. That makes it a good fit for a narrow price range and a poor one for a catalogue that spans a lot of price points.
Free gift with purchase
A product rather than a price cut, so it costs you unit cost instead of margin percentage and your price list stays intact. The gift has to be something the buyer would plausibly have wanted, otherwise it reads as clearing stock you could not sell, which is often exactly what it is.
A category or collection sale
Discounting one part of the range rather than the whole shop. It concentrates the offer where you have stock or margin to spare, and it protects everything outside the sale from being repriced in the shopper's mind. It also gives you a reason to talk about a specific part of the catalogue rather than announcing a generic sale.
Buy one, get one
Two units move instead of one and the headline price never changes, which protects your list price in a way a percentage does not. It suits consumables and anything people stockpile. It suits nothing that a household only needs one of, where the second unit is not a reward but a storage problem. OnVoard has no template for this one because it is a catalogue and pricing rule rather than an on-site message.
Product bundles
A fixed set sold below the sum of its parts. The useful version solves a job the shopper already has, like everything needed to start, so the discount is the reason to buy today rather than the reason to buy at all. The weak version is three slow sellers taped together, which shoppers recognise immediately.
Spend more, save more tiers
A ladder rather than a single threshold: a little off above one number, more above the next. It raises order value further than a single floor, and it is harder to understand at a glance, so it belongs on a cart page rather than in a popup that has one line to explain itself.
Referral discounts
The existing customer gets something for introducing someone, and the newcomer arrives already recommended. The acquisition cost is genuinely lower than advertising. It needs an existing base of happy customers to work at all, which makes it a second-year promotion for most stores rather than a first-month one.
Loyalty points
Value accrues across orders instead of arriving at one. It shifts the question from this purchase to the next, which is the point, and it is the slowest of everything on this list to show a result. Points also create a liability you owe later, so the maths deserves more care than a one-off code.
Gift cards and store credit
Credit keeps the money in your store rather than refunding it out. As a promotion it works best as compensation or as a returns alternative, and it is weaker as an acquisition offer because it asks a new visitor to commit before they have anything to spend it on.
Free samples with an order
A small extra that costs you unit price rather than margin percentage, and introduces a product the shopper has not tried. It is the most reliable way to move a customer from one product to two. It only works where sampling is meaningful, so mainly food, beauty and supplements.
Clearance and end of line
A real reason for a real discount, which is the part shoppers believe. Because the stock is finite, the deadline is honest and needs no manufactured countdown. Keep it visibly separate from your main range so the discount does not reset expectations for everything else.
Seasonal and holiday sales
The calendar supplies the reason, so you do not have to invent one, and everyone else is doing the same thing at the same time, so attention is expensive. It rewards preparation more than any other promotion here: the list you built in October is what makes November work.
How deep the discount should be, and when not to discount
Discount depth is a margin decision before it is a marketing one. On a 50 percent gross margin, a 20 percent discount removes about 40 percent of the profit on that order, which means the promotion has to bring in substantially more orders than it gives away to break even. The lower your margin, the smaller the number you can afford and the more the promotion has to earn.
That maths is why the offers on this page are not all percentages. Free shipping costs you a fixed amount instead of a proportion, so it hurts less on a large basket. A threshold spends the discount only when the shopper adds more. Early access and a birthday club spend nothing at all at the moment of capture.
When not to discount
There are three situations where a discount is the wrong instrument. When the visitor was going to buy anyway, in which case you have paid for an order you already had. When your problem is that people cannot choose rather than that they will not pay, in which case a recommendation is the fix and a discount is noise. And when you have discounted recently enough that your audience is waiting for the next one, in which case another promotion deepens the habit rather than breaking it.
The honest test is simple. If you cannot say which behaviour the discount is buying, do not run it.
Timing, frequency and who sees it
The mechanic decides what the promotion offers. Timing decides whether anyone is willing to read it.
Trigger
On page load reaches everyone, including people who have been on the site for two seconds and have no idea what you sell yet. On exit intent reaches only sessions that were ending. On add to cart and on checkout intent reach people who have already shown intent, which is why they suit recommendations rather than discounts. OnVoard templates carry a trigger already set for the job they do, so the sensible order is to pick the job and inherit the trigger, rather than picking a design and choosing when to fire it.
Frequency
One interruption per session is the rule worth keeping. Past that, my view is that the second interruption mostly costs you the attention you bought with the first, and it is the self-inflicted problem I would look for before touching anything else. If you run an arrival offer and an exit offer, make them the same campaign with two moments, not two campaigns competing for the same visitor.
Mobile
Exit intent depends on a cursor leaving the viewport, so it is a desktop mechanism. On phones there is no equivalent signal. One template in the catalogue says so in its name rather than degrading silently, which is the behaviour to copy: decide what mobile gets instead, rather than assuming your desktop promotion covers everyone.
Who sees it
The single highest-value targeting decision is separating new visitors from returning customers. A first-order discount shown to a loyal customer is money handed to someone already at the till. If you make one targeting change this quarter, make it that one.
How to run one of these with OnVoard
The templates in this article are ActionPop templates, and they are installed rather than built. Open the template page linked under any example, and it carries the screens of that flow so you can see every step before you commit to it.
Two practical notes from the catalogue itself. Most of the discount offers here mint a real code rather than displaying a static one, which means they need to be connected to the store that issues the code. The recommendation templates need a product recommender behind them, because a panel of suggestions is only as good as the products it can choose from. Both are capabilities to confirm you have before you plan a campaign around them, not afterwards.
Beyond that, the work is copy rather than construction. Every number in the templates is a starting point: the percentage, the spend floor, the prize set and the message frequency are all yours to set, and they should come from your margins and your catalogue rather than from the template defaults.
What to measure once it is running
Most promotion reporting stops at the popup conversion rate, which is the least informative number available. It tells you how many people accepted a free thing. It says nothing about whether the promotion made you any money.
Four things are worth watching instead.
Whether the discounted orders would have happened anyway. Compare the promotion period against a normal one rather than against zero, because some share of every promotion is orders you were getting regardless.
Average order value, particularly on threshold and cross-sell promotions. Those promotions exist to move this number, so if it has not moved, the mechanic is not doing its job whatever the acceptance rate says.
What the captured list does afterwards. An email or phone number is the actual product of most of these promotions, and its value shows up in the second order, not the first.
Redemption against capture. A large gap between people who took the code and people who used it usually means the offer was attractive and the products were not, which is a catalogue problem wearing a promotion problem's clothes.
Give it long enough to read. A promotion aimed at a first order can be judged in a fortnight. A birthday club cannot be judged until the birthdays arrive.
Common questions
What is the difference between a sales promotion and a discount?
A discount is one kind of sales promotion, the kind that lowers the price. A promotion is any temporary change to the value of the offer, which includes free shipping, a free gift, early access, a bundle or a competition. Several of the strongest promotions on this page do not lower a price at all.
What is a flash sale?
A promotion whose defining feature is a short, stated window rather than the size of the offer. The deadline is what does the work: it converts a decision the shopper was postponing into one they have to make now. A flash sale that runs continuously is not a flash sale, and running one teaches your audience that your deadlines are not real.
Should I offer a discount or free shipping?
Free shipping when your margins are tight or your prices are carefully positioned, because it costs a fixed amount rather than a proportion and it leaves your price list alone. A percentage discount when your basket sizes vary a lot, because a fixed shipping cost hurts disproportionately on small orders. If postage is a large fraction of your order value, free shipping is the expensive option rather than the safe one.
How often should I run promotions?
Less often than the mechanics allow. The constraint is not what your software can do, it is what your audience learns. A store that discounts every month has taught its customers to wait, and that habit costs more than any single promotion earns.
Do popups still work?
They work when they interrupt at a moment the visitor can afford, and they fail when they interrupt at a moment they cannot. This is why the trigger matters more than the design, and why an exit-intent offer is a safer first experiment than an arrival takeover.
Which promotion should I start with?
If you have no list, the first-order offer. If you have thin margins, free shipping or early access. If you have never run one, the exit-intent save, because it only reaches sessions that were ending anyway.