4 Reasons Bonuses Work Better Than Discounts

While both bonuses and discounts are effective, as a general rule, bonuses work a little better than discounts in most cases. Let’s explore the four reasons why this is.

1. Consumers frame losses and gains differently

People make decisions not only based on the content of the decisions but by how those decisions are framed. And a product promotion, by definition, is a different way to frame a decision to buy.

One of the easiest ways to frame a decision is by framing it as a loss or a gain. A discount results in a reduced loss in money. A bonus results in the gain of extra product. For example, a 50% bonus and a 33% discount may be economically equivalent, but one is framed as a loss (or a reduced loss) and the other as a gain.

Why does this matter? Because people are more likely to be affected by losses than gains; the desire to avoid a loss is greater than the desire for its equivalent gain. For example, would you take a bet on a coin toss, where you win $10.00 if it lands heads, but lose $10.00 if it lands tails?

At first glance, this is irrational. A coin toss has a 50% chance of landing heads and a 50% chance of landing tails. This means the expected value is the same:

• Heads: $10.00 × 50% = $5.00

• Tails: $10.00 × 50% = $5.00

As you can see, the expected value of both outcomes is equal, yet most people don’t take the bet because $5.00 lost is worth more than $5.00 gained.

Two psychologists, Amos Tversky and Daniel Kahneman, found that it takes a win of $22.50 against a loss of $10 for most people to take the bet.

• Heads: $22.50 × 50% = $11.25

• Loss: $10.00 × 50% = $5.00

This shows us that, for this specific bet, even though $10 is worth less than $22.50, a $10 loss hurts as a much as a $22.50 gain feels good.

This difference is called the loss aversion ratio. It describes the ratio in how people value losses to gains. For this coin toss bet, the loss aversion ratio is 2.25, although for most decisions the loss aversion ratio is roughly 1.31. In other words, a $10 loss is valued equally to a $13.10 gain.

As we’ve seen, discounts and bonuses that may be economically equivalent produce divergent consumer responses:

• Discounts can be viewed as reduced losses in money or gains in money.

• Bonuses can be viewed as gains in product or reduced losses in money.

At the same time, a difference in framing can’t fully explain why people generally prefer a bonus to its economically equivalent discount. It can only explain why framing changes behaviour. In doing so, it just pushes the question further back:

• Why do consumers sometimes experience discounts as reduced losses in money and at other times gains in money?

• Why do consumers sometimes experience bonuses as gains in product and at other times reduced losses in money?

Next, we’ll dig a little deeper by taking a close look at the perception of numbers themselves.

2. Consumers are bad at calculating percentages

Numbers are tricky things.

For example, 50 is a bigger number than 33. Does that mean 50% is bigger than 33%?

You might be inclined to answer yes, but before you do, consider the following.

It’s true that sometimes 50% is bigger than 33%:

• While stuck in traffic, I have two alternate options for getting home: one that takes 50% longer and one that takes only 33%.

• When I see two promotions: one that offers 50% more shampoo and one that offers 33% more shampoo.

But other times, 50% is not bigger than 33%:

• When I choose between 50% more product and 33% off the price.

• When I see an advertisement telling me Brand A is 50% better than Brand B, instead of an advertisement telling me Brand B is 33% worse than Brand A.

That’s the big difference between regular numbers and percentages: bigger numbers are always bigger than smaller numbers, but bigger percentages are not always bigger than smaller percentages.

The reason this gets you confused is that you sometimes compute differences in percentages as if they were differences in absolute magnitude.

Here are two examples:

• When two brands are compared, people like Brand A if it is framed as 25% better than Brand B, as opposed to saying Brand B 20% worse than Brand A, even though both are the same.

• When percentage discounts are stacked, people are more likely to buy. Specifically, they’re more likely to buy something discounted 25% and then marked another 20% off compared to a discount of 40%, even though a stacking discount of 25% and 20% is economically equivalent to a 40% discount.

3. Consumers neglect the base value

We just saw how people calculate percentages as if they were regular numbers, when, in fact, percentages don’t work the same way.

But how, exactly, are percentages not regular numbers?

Here’s how: they require a base value.

If something is 50% more or 50% less, you need to ask less than what? The “than what” is the base value.

In this way, percentages aren’t absolute, they are always relative to a base value. And if there’s one thing you can count on humans to do, it’s that they’ll neglect the base value.

Researchers tested this in an experiment with 120 consumers. They were shown one of two promotions:

1. Regular price of $3.89 for an 8-ounce bottle with a special promoting 50% more free

2. Regular price of $3.89 for an 8-ounce bottle, with a special promoting 35% off the regular price.

If you look closely, you’ll see that the two promotions are economically equivalent. But, as expected, people opted for “50% more free” because it sounds like a greater amount than “save 35%”. They didn’t pay attention to the base value of $3.89.

As a follow-up, researchers gave extra emphasis to the $3.89 number, and then re-ran the experiment. When they did, people were more likely to make a quick calculation—which made them just as likely to choose both the discount or the bonus.

What does this mean? As a default, people are prone to making quick decisions and using mental shortcuts, like “50 is bigger than 35, so 50% must be bigger than 35%.” But when people are prompted to consider their starting point, they’re more likely to do the math—which means they’re equally likely to choose from two economically equivalent options.

4. Consumers have a hard time comparing numbers close together

There’s another wrinkle. People have a hard time comparing numbers that are close together. In fact, it takes a few milliseconds more to compare 25 and 50 than it takes to compare 10 and 11.

The same is true of percentages: when the numerosity of percentages is close, you’re forced to pause for a few extra milliseconds and consider the numbers you’re looking at. As a result, you’ll make a more conscious and effortful comparison between a discount and a promotion, too.

In other words, you’re more likely to choose a bonus pack over an economically equivalent discount when the bonus pack offer is larger. This is because as the promotion gets better, the numerosity of the percentage between a bonus and a discount gets bigger, too, as you can see in the far-right column in this table:

To test this, researchers offered bonus and discount promotions at three levels:

1. In the first scenario, customers were offered either a 50% bonus or a 33% discount (a difference of 17%).

2. In the second scenario, customers were offered either a 100% bonus or a 50% discount (a difference of 50%).

3. In the third scenario, customers were offered either an 11% bonus or a 10% discount (a difference of 1%).

They found that people perceived the difference in percentages to be large in the first two scenarios, but not in the third scenario. They also found that people in the 50% bonus/33% discount condition preferred a bonus to a discount more than people in the 100% bonus/50% discount condition and the 11% bonus/10% discount condition. In fact, people tended to prefer a bonus pack at 100%-50% and 11%-10% at roughly the same rate.

What’s going on here? People prefer the bonus under two conditions: 1) when it’s easy to calculate the difference between the two, and 2) when the numbers are bigger compared to smaller.

The bottom line: customers prefer bonuses to discounts because they’re bad at math, at least most of the time.