Two offers. Three items at $49 each. One says "33% off". The other says "Buy 3 for $99".
The first costs you $48.51. The second costs you $48.00. Functionally identical, and the second is fractionally cheaper for the store.
They do not perform the same. Not because one is a better deal, but because they ask the shopper to do completely different things. One asks for arithmetic. The other asks for a decision.
Most discount conversations are about depth. How much off. The more useful conversation is about shape.
Quick List – What This Covers
1. Why bundle pricing outperforms an equivalent percentage
2. The Rule of 100 – when to show a rate, when to show an amount
3. The zero-price effect and why "free" isn't just a big discount
4. The goal gradient and how thresholds create effort
5. Anchoring, and who sets the reference price
6. Why a capped discount can feel more credible, not less
7. Which structure to reach for at which price point
Same Economics, Different Decision
Look at what each offer actually demands of a shopper standing on a product page with eleven other tabs open.
"33% off" requires them to work out what 33% of $147 is, judge whether that result is good, and then hold it in their head while comparing against a competitor advertising 35%. Three cognitive steps before they can even evaluate the offer.
"Buy 3 for $99" requires one decision: do I want three of these? The price is already computed. There's no rate to benchmark and nothing to convert. The offer has also quietly done something the percentage never does – it named the basket size.
That's the part worth sitting with. A percentage discount applies to whatever the customer was going to buy anyway. A bundle price tells them what to buy. You've moved from discounting demand to shaping it.
Six Effects Doing the Work
Behavioural research on pricing is deep, and the findings below are among the more robust. Effect sizes vary enormously by category, price point and audience, so treat these as directions to test rather than settled numbers.
The zero-price effect. Work by Dan Ariely and colleagues found that dropping a price to zero produces a jump in demand far larger than the equivalent price cut down to a very small amount. "Free" appears to be processed as a category rather than a point on a scale. Practically: a free gift often outperforms a discount of the same cash value, and usually costs you less, because you give up COGS rather than a slice of the whole basket.
The goal gradient effect. People accelerate toward a reward as it comes into view – documented in loyalty-card research by Kivetz and colleagues. This is the entire engine behind "spend $150, get $25 off". The threshold isn't just a qualifying rule; it's a target, and shoppers work harder as they approach it. It's also why the number you choose matters more than the reward attached to it.
Anchoring. Any price is judged relative to a reference, and whoever supplies the reference controls the judgement. A compare-at price does this. So does a bundle: "3 for $99" implies a per-unit price of $33 and quietly makes $49 the anchor. Choose your anchor deliberately, because if you don't set one, your competitor's price becomes it.
The left-digit effect. $99 is read as "ninety-something" and $100 as "a hundred". A single cent moves the price into a different mental bracket. This is why bundle totals cluster at 99 rather than 100 – and why "Buy 3 for $99" reads as double-digit spending while "3 for $100" reads as triple.
Mental accounting. Richard Thaler's work suggests gains are often valued more highly when separated than when combined. Two distinct wins – a discount and a free gift – can register more strongly than a single larger discount of equal value. Worth testing before assuming one big number is the strongest play.
Metric intuition. People evaluate units more comfortably than rates. Three items for $99 is concrete. 33% off $147 is a calculation. Every ounce of arithmetic you remove from the decision is friction you stop charging the customer.
The Rule of 100
A useful heuristic, popularised by Jonah Berger: below $100, percentages tend to look larger; above $100, absolute amounts do. Shoppers read the digits before they do the maths.
On a $40 product, "25% off" and "$10 off" are the same saving, but 25 is a bigger number than 10. On a $1,200 product, the same 25% becomes "$300 off" – and now the absolute figure is the one that looks substantial.
It's a rule of thumb rather than a law, and it's worth testing on your own catalogue. But when you're setting a campaign with no prior data, it's a far better default than picking whichever framing you used last year.
Why a Cap Can Make an Offer Stronger
Merchants usually treat discount caps as a defensive measure – a way to stop a 20% offer from costing $600 on a high-ticket order. That's true, and reason enough on its own.
There's a second effect that gets missed. "20% off, up to $50" reads as a considered, specific offer. Unqualified deep discounts on expensive goods can trigger suspicion – shoppers wonder what's wrong with the stock, or whether the original price was ever real. A stated limit signals that the number was arrived at deliberately.
Specificity does similar work elsewhere. "Buy 3 for $99" feels like a considered offer. "Buy 3 for $100" feels like a round number someone guessed at.
Matching Structure to Price Point
There's no universal best structure. There is a reasonable default for each type of product, and these are the ones worth testing first.
Two things to hold onto when applying this. First, the structure should match how the product is actually bought – bundles work where multiples are natural and fall flat where nobody wants three. Second, your thresholds should come from your own order data, not from round numbers. Pull ninety days of orders, take the median value, and set the first threshold roughly 20–25% above it so it's reachable by adding one more item.
Where This Breaks Down
Structured offers aren't a free win, and it's worth being honest about the failure modes.
A bundle on a considered purchase is noise – nobody buys three mattresses. A threshold set too far above your median order value reads as unreachable, and shoppers disengage rather than stretch. Complexity has a cost too: an offer that needs a paragraph to explain will underperform a blunt percentage, however elegant the logic behind it. And repeated deep discounting of any shape trains customers to wait, which no amount of framing fixes.
The test is simple. If a shopper can't understand the offer from the banner alone, restructure it.
Building These on Shopify
Here's the practical catch. Shopify's native discount engine handles percentages, fixed amounts and simple BOGO. Almost every structure discussed above sits outside it – fixed-price bundles, spend-threshold tiers, capped percentages, Nth-item patterns, cross-category pairing.
That used to mean Shopify Plus and Scripts. Since Shopify Functions became available across all plans, it doesn't.
Seventh Triangle built Every Possible Discount to cover exactly this range: bundle pricing, tiered thresholds, capped percentage discounts, volume breaks, customer-tag gating and market-specific offers, on any Shopify plan at a flat $19/month. For a wider view of the category, including where other tools fit better, see our breakdown of the Top 7 Best Shopify Discount Apps. And if you're planning a peak-season campaign around these structures, the BFCM 2026 Promotion Playbook covers how to protect margin while you do it.
|
Explore App - Every Possible Discount |
Final Thoughts
Depth is what most brands argue about internally. Shape is what actually determines whether a discount builds anything.
The same money can buy you a shrug or a larger basket depending entirely on how it's framed. "33% off" spends your margin on a customer who had already decided. "Buy 3 for $99" spends slightly less and tells them what to put in the cart.
Before your next campaign, take whatever percentage you were planning and write down three other ways to express the same cost. Then pick the one that asks the least of the shopper.
Frequently Asked Questions
1. Is "Buy 3 for $99" better than "33% off"?
For multi-unit products, bundle pricing usually performs better than an equivalent percentage because it removes the arithmetic and specifies the basket size. The shopper only has to decide whether they want three, rather than calculate a saving and benchmark it. Bundles are a poor fit for considered single purchases, where nobody wants multiples.
2. Should I advertise a percentage or a dollar amount off?
Use the framing that produces the larger number. Below roughly $100, the percentage usually looks bigger – 25% off beats $10 off on a $40 item. Above $100, the absolute amount looks bigger – $300 off beats 25% off on a $1,200 item. This is a heuristic known as the Rule of 100 and is worth testing on your own catalogue.
3. Why does a free gift work better than a discount of the same value?
Research on the zero-price effect suggests demand jumps disproportionately when a price reaches zero, beyond what an equivalent price reduction produces. "Free" appears to be processed as a category rather than a point on a price scale. A free gift also costs the merchant its COGS rather than a percentage of the entire basket, which is typically cheaper.
4. Where should I set a spend threshold?
Set the first threshold roughly 20–25% above your median order value from the last ninety days. Use the median rather than the average so a handful of large orders don't distort it. The threshold should be reachable by adding one more item; if reaching it requires three, most shoppers disengage rather than stretch.
5. Does ending a price in 99 actually make a difference?
The left-digit effect means $99 is processed as "ninety-something" while $100 is processed as "a hundred", placing them in different mental brackets despite a one-cent gap. The effect is well documented in pricing research, though its size varies by category and is generally weaker for premium positioning, where round numbers can signal quality.
6. Can Shopify create fixed-price bundles natively?
Shopify's native discount engine supports percentages, fixed amounts and basic BOGO, but not fixed-price bundles such as "buy 3 for $99". That structure requires a discount app built on Shopify Functions. Since Functions is available on every Shopify plan, this no longer requires Shopify Plus as it did under the older Scripts system.
7. Can I put a maximum on a percentage discount?
Not with Shopify's built-in discounts, which apply a percentage with no ceiling. Capping a percentage – for example "20% off, up to $50" – requires a Functions-based discount app. Beyond protecting margin on high-ticket orders, a stated cap tends to read as a deliberate, credible offer rather than an arbitrary markdown.
