Should I Use Percentage Off or Dollar Off?

Twenty percent off. $20 off. Buy one, get one half off.

They're all roughly the same discount on a $100 purchase. So why might one outperform the others? And more importantly, how do you know which one will work for your specific products and customers?

Most brands default to percentage discounts because that's what everyone does. But what if your customers respond better to dollar amounts? Or free gifts? The only way to know is to test, and the results might surprise you.

Why Do Most Brands Default to Percentage Off Discounts?

It's the path of least resistance, not necessarily the best choice

Here's how most promotional decisions go: "What discount should we offer?" The team debates 15% versus 20% versus 25%.

Nobody asks the more fundamental question: "What's the best way to present value?"

This is like deciding how much to spend on a gift without considering what to buy. How you present value matters as much as the amount of value you offer.

Think about it: Would you rather save $50 or get 20% off? Depends on the purchase price, right? Your customers are making the same calculation, but you're not giving them options.

What Are Your Options Beyond Percentage Off?

Most brands use one of four common approaches, each can trigger different responses

Before testing, understand what you can test. Here are four common ways brands present value, though creative brands sometimes combine these or invent variations:

Percentage Off

"Save 20%" is the classic, scalable approach that can feel fair across price points. But 20% off a $25 item ($5) barely covers shipping, while 20% off a $500 item ($100) might give away too much margin. Percentage discounts can lose impact at lower price points.

Dollar Amount Off

"Save $25" is the concrete value statement that can remove mental math and create clarity. "Save $25" can feel more tangible than "Save 17.3%" even when they're equivalent. But they don't scale as easily: $10 off a $30 purchase feels substantial, while $10 off $300 might not. Fixed dollar amounts tend to work well when your products cluster around similar price points, or when you have a small, consistent catalog.

Gift With Purchase

"Free tote with orders over $75" is the added value approach where a gift costing you $5 might have a perceived value of $25. This arbitrage can make gift-with-purchase incredibly powerful when done right with branded accessories, samples, or exclusive items. But if the gift feels cheap or irrelevant, it can hurt your brand instead of helping it.

Why Does Testing Different Discount Types Give You Certainty?

Same value, different psychology, measurable impact on profit

Testing reveals how YOUR customers perceive and respond to value. Something you can't learn by testing 15% versus 20%.

What testing reveals:

Percentage off might feel "fair" but lack urgency. Dollar amounts can create concrete, immediate value. 20%Gifts can create surprise and delight without training customers to wait for discounts.

When you test different discount types with A/B testing, you're holding the value constant and changing only the presentation. Same Monday morning. Same products. Same discount depth. Only the type differs.

This isolates the presentation as the variable. When profit per visitor changes, you know the discount type caused it.

What's the Simplest Test I Can Run?

Test your current promotion against its opposite

Step 1: Establish Your Baseline

Run your current promotion and measure conversion rate, average order value, and profit per visitor.

Step 2: Test the Opposite

Try the most different type from your current approach. If you use percentages, test dollar amounts. If you discount prices, test gifts.

If you currently offer 20% off:

• Control: 20% off

• Variant A: $20 off (if average order is around $100)

• Variant B: Free $20 gift with purchase

Step 3: Run Until You're Confident

The progression matters more than the timeline. High-traffic brands might see clear patterns in weeks. Lower-traffic brands might need more time. Let your data confidence guide your pace.

Step 4: Measure Profit Per Visitor

Focus on profit per visitor. This is your North Star metric.

A discount type that increases conversion 10% but tanks your average order value 20% isn't a win. Sometimes a discount type that converts lower drives more profit because customers spend significantly more.

Different situations may require different goals. If you're clearing inventory, volume might matter more than profit. But for most tests, profit per visitor tells you what's actually working.

Common Mistakes When Testing Discount Types

Avoid these testing pitfalls

• Testing during different time periods: Seasonality and competitor actions muddy your results. Test simultaneously with A/B testing, not sequentially.

• Testing too many options at once: Start with your current promotion plus one or two alternatives. Too many variants split traffic thin and take forever to reach confidence.

• Not matching discount type to behavior: BOGO on high-ticket items customers buy once doesn't match purchase patterns. Match your test to how customers actually shop.

• Stopping tests early: Run tests until you reach statistical confidence. Early results can mislead.

Stop Guessing. Start Knowing.

Every time you default to your usual discount type, you're making an untested assumption. That assumption might be costing you 20-30% in profit per visitor.

Test alternative discount types with the same value and measure profit per visitor. The data will show you which discount type tends to work best for YOUR customers.

Don't guess which discount type works best. Know.

Ready to discover which discount type can drive the most profit? When you're ready to know for sure, let's get you testing beyond what's typical.