Pricing changes can have a swift impact on your business’s profit. Raise prices too high and sales might drop; set them too low and profit margins shrink. And the pressure to get pricing right only increases when the cost of doing business goes up: about 84% of global small and medium-sized enterprises cited rising costs as their top concern in 2024.

To figure out how to maximize revenue, companies can run experiments that test prices in the market to see how customers respond. Below, we’ll explain how to design, run, and improve pricing experiments for revenue growth.

What’s in this article?

• What is a pricing experiment?
  
• What are the goals of pricing experiments?
  
• What types of pricing experiments can businesses do?
  
• How do you design an effective pricing experiment?
  
• How do you implement and manage pricing tests?
  
• How do you analyze pricing experiment results?
  
• How Stripe Billing can help
  

What is a pricing experiment?

A pricing experiment is a structured way to test what customers do when you change the price of a product or service. You create a controlled setup where all elements (e.g., product, features, checkout flow) except for price remain static. One group sees the current price while another sees a variant, and you measure the differences in sign-ups, average revenue per user (ARPU), and retention between these groups.

You must isolate pricing as the only variable. If product changes or marketing campaigns overlap, you won’t know what really drove the outcome.

What are the goals of pricing experiments?

The intent of pricing experiments is to find out how customers behave when you change your prices. Here’s what a pricing experiment can help you do.

Find an effective price

Undercharging or overcharging are common worries for companies. A pricing experiment can show you the price points at which customers will buy your products and those at which they won’t.

See what pricing changes do to revenue

Sometimes, if the conversion rate holds, you can increase revenue by raising the price. Other times, lowering the price can boost revenue by attracting a rush of new customers.

Gauge segment sensitivity

In a pricing experiment, you can watch demand rise and fall in real time. You’ll see who still pays when you increase the price and who leaves. This will help you understand which customer segments you can charge more and which you can’t.

Try new models

According to a 2024 Stripe survey, 69% of subscription business leaders globally said they planned to launch new pricing models in the next year, motivated in part by concerns about churn. If you’re considering shifting to bundles, subscriptions, or usage-based billing, a pricing experiment can make the shift less challenging by letting you try it on a small group first.

Monitor customer retention

Usage-based billing plans can help reduce churn and make customers stick around longer. Monitoring retention during an experiment shows you whether the change in pricing model built staying power or just pulled in deal seekers.

When done well, pricing experiments outline the relationship between price and demand in the real market—and help you set prices that balance customer value with sustainable revenue.

What types of pricing experiments can businesses do?

The type of pricing experiment you should do depends on what you’re trying to learn. Here are the most common formats:

• A/B price tests: Split customers randomly into two groups and show each a different price point (e.g., $50 for one group, $45 for another). This simple format shows how a price directly affects conversion and revenue.

• Tier or multiple-choice tests: Test entire product or service menus. For example, a software-as-a-service (SaaS) company might run an experiment offering multiple plans: Basic at $19, Pro at $49, and Enterprise at $99. This will gauge customers’ willingness to pay for different features, and it will reveal whether having choices pushes customers toward a midtier option.

• Bundles vs. à la carte: Try combining products or services into a package (e.g., “X and Y together for $99”) as well as selling them individually to see which pricing model lifts overall spending. This can help you discover whether customers prefer a package deal or per-item pricing.

• Discounts and promos: Give a coupon or discount (e.g., “20% off this month”) to one group, but not to another. By comparing results from the two groups, you’ll learn whether promotions increase sales volume enough to outweigh the thinner profit margins on products with reduced prices. You can also learn how different types of discounts (e.g., slashed prices, free shipping, free trial) affect conversion.

• Psychological pricing: Test psychological cues such as charm pricing (e.g., $49.99 vs. $50), anchor pricing (e.g., “was $129, now $99”), and framing (e.g., presenting the price as a limited time offer). Psychological cues shape behavior in subtle ways and can be tested like any other variable.

• Dynamic or usage-based trials: Raise prices when demand peaks, or charge per transaction rather than flat fees. These are bigger pricing changes, but even small tests can show whether customers embrace flexibility.

• Segment-specific tests: Test a lower student rate or a regional discount and compare results against those at the standard price. This can help you gauge how much certain groups are willing to pay.

How do you design an effective pricing experiment?

If the design of your pricing experiment is messy, you won’t know whether the results are reliable. Here’s how to design it so you learn something useful.

Frame the question

Rather than simply testing a higher price, start with a falsifiable hypothesis (e.g., “Dropping prices from $50 to $45 will lift conversion enough to raise monthly revenue by 10%”). This forces you to name both the variable you’re changing and the metric you’ll judge it on.

Pick one variable

Price points, billing cycles, packaging, and discounts are all testable, but not simultaneously. If you change both the price and the feature set, you won’t be able to determine which change caused the experiment’s outcome.

Set up control and test groups

The control group holds steady at the current price; the test group gets the variant price. Random assignment to groups is typically the best approach. Lower-volume businesses sometimes base group assignments by region or time of year, but this can expose them to outside factors such as seasonality and competitor moves.

Define your metrics up front

Choose the metrics that relate to your goal: conversion, average order value, net revenue, churn, customer lifetime value, etc. Establish criteria for success (e.g., “Variant wins if conversion rises 15% without decreasing ARPU”). Set limits that tell you when to stop the test early, if revenue collapses.

Use a power calculator

An experiment that generates a handful of sign-ups won’t prove anything. Use a power calculator: this statistical tool helps you figure out how many participants or data points you need for the test results to be trustworthy, as well as the suggested duration for the experiment. Many pricing tests need weeks to produce results of significance.

Control the variables

Avoid launching new features or promos in the middle of your test. Keep the environment steady so you can credibly say price drove the outcome.

Check your infrastructure

Legacy billing systems can’t always handle parallel prices. Stripe Billing was built with experimentation in mind: you can create multiple price IDs for the same product, route customers randomly, and keep reporting separate. Whatever billing stack you’re on, ensure a customer sticks with the price they were assigned, from landing page to invoice.

How do you implement and manage pricing tests?

Once you’ve designed your experiment, you need to implement and manage the test in a way that doesn’t create chaos for your customers or your team. Here’s how to get it right:

• Get everyone in sync: Inform anyone who interacts with customers or revenue (e.g., sales, support, marketing, finance) when an experiment is live. Your internal teams need to be ready if questions come up.

• Roll it out gradually: Start with a small share of traffic to check for bugs or billing errors, then scale up. That way, if something breaks or the test wildly underperforms, you can contain the damage.

• Watch in real time, but decide later: Dashboards should track conversion, revenue, and guardrail metrics as data comes in. If revenue collapses, stop the experiment. Otherwise, resist concluding the experiment early. Early results don’t always show the full picture.

• Keep the test clean: Customers in one group shouldn’t drift into another (e.g., by sharing links or switching devices). Don’t overlap major experiments that could interfere with each other.

• Protect the customer experience: Pricing is visible, and people talk. If you’re testing different prices in the same market, think through how you’ll handle customers’ price comparisons. Some companies frame tests as limited-time offers to soften the perception of unfairness.

• Use the right infrastructure: Running two prices in parallel can stress billing stacks. Stripe Billing makes it easier to run different prices and keeps the process in one place.

How do you analyze pricing experiment results?

When the test ends, you’ll have to sort through the raw data produced. Here’s how to analyze the results of your pricing experiment.

Start with significance

Check that the results are statistically valid. A 2% lift in conversion means little if the sample is tiny. Use figures such as P values and confidence intervals, or use your experimentation platform to confirm the results are statistically valid.

Look at the core metrics together

Put the results of the control and test group side by side: conversion rate, average order value, ARPU, and gross profit. One metric in isolation doesn’t provide the full picture. A higher price might lower conversion but still grow revenue. A discount could sharply increase sign-ups but sharply decrease margin. Examine all the metrics together for a complete answer.

Segment the data

Separate the results into cohorts: for instance, new vs. returning; enterprise vs. small and medium-sized businesses; or Region A vs. Region B. Sometimes, a change can lift one segment and depress another. Comparing the results from one segment to another can tell you whether to roll your price out universally or target more precisely.

Consider the long term

Short-term wins can mask long-term setbacks. Determine whether discount-acquired customers leave faster or whether higher-paying cohorts engage more. If you’ve tracked retention, net promoter score, refund rates, or support volume, factor those in. A pricing win that eventually results in dissatisfaction isn’t really a win.

Translate results into business impact

Calculate the long-term impact of the lift or loss: for example, “This price point increased revenue per visitor by 5%, which equals $X million annually if rolled out.” Explain the outcome in terms that executives and finance teams care about.

Decide, then document

Adopt the winning price formula if it meets your success criteria. If it doesn’t, keep the current pricing or design the next test. Whatever the results, write down what you learned: the variable, effect, and numbers. Pricing experiments build on each other, and a clear record helps you refine them over time.

The goal is to build a trustworthy dataset about how your market reacts to price so you have empirical evidence when the next assessment of pricing starts.

How Stripe Billing can help

Stripe Billing lets you bill and manage customers however you want—from simple recurring billing to usage-based billing and sales-negotiated contracts. Start accepting recurring payments globally in minutes—no code required—or build a custom integration using the API.

Stripe Billing can help you:

• Offer flexible pricing: Respond to user demand faster with flexible pricing models, including usage-based, tiered, flat-fee plus overage, and more. Support for coupons, free trials, proration, and add-ons is built-in.

• Expand globally: Increase conversion by offering customers’ preferred payment methods. Stripe supports 125+ local payment methods and 130+ currencies.

• Increase revenue and reduce churn: Improve revenue capture and reduce involuntary churn with Smart Retries and recovery workflow automations. Stripe recovery tools helped users recover over $6.5 billion in revenue in 2024.

• Boost efficiency: Use Stripe’s modular tax, revenue reporting, and data tools to consolidate multiple revenue systems into one. Easily integrate with third-party software.

Learn more about Stripe Billing, or get started today.