Shopify Price Testing: How to Run It Without Angering Buyers
You suspect you're underpriced. A price test can prove it, or it can put two prices in front of one buyer and torch the trust you spent years building. Here's how to get the proof without the blowup.
I'll say the unpopular thing first. Most founders who think they're underpriced are right, and most of them never find out, because they're scared of the test.
Here's the short answer. Shopify price testing works when you follow 4 rules:
- Test one product, with a control and one or two prices 5 to 10% apart.
- Give every buyer one price, from first visit through checkout.
- Run at least two weeks and about 300 orders per price.
- Crown the winner on revenue per visitor and gross profit, never conversion rate.
Break rule 2 and you're the store a customer screenshots. Break rule 4 and you'll throw away your best price because it "converted worse."
Price moves revenue per visitor faster than any button color ever will. That's exactly why it deserves the most careful test you'll ever run.
Why does price testing make buyers so angry?
Because the buyer doesn't see a test. They see you charging them more than the person next to them.
This has blown up on giants. Back in September 2000, Amazon customers found the same Planet of the Apes DVD listed at $64.99 for one shopper and $74.99 for another on the same day. Computerworld's coverage quotes Amazon saying some customers would pay more and some would pay less. Buyers didn't care that it was a test. They cared that they were the "pay more" group.
Twenty-five years later, same story. In December 2025, a study by Groundwork Collaborative and Consumer Reports had nearly 200 shoppers buy the same groceries on Instacart at the same time. ABC30 reported that about three in four items showed up at more than one price, with one box of crackers at $3.99 for some shoppers and $4.89 for others. Instacart called them randomized A/B tests. It still ended the program on December 22, saying the tests "missed the mark for some customers."
And you don't need a newspaper to get burned. One merchant on the Shopify Community forum described customers seeing $110 on a product priced at $74.97, with one shopper watching the price jump $35 after a refresh. That wasn't even a test. It was a glitch. The buyer can't tell the difference.
So the anger has nothing to do with testing. It comes from inconsistency. One person, two prices. That's the thing you engineer out.
How do you set up Shopify price testing without two prices for one person?
You buy the plumbing instead of faking it.
The old hack was duplicating a product at a new price and splitting traffic with a redirect. Don't. Duplicate URLs leak into Google, emails and shared links, and that's how a buyer sees both numbers.
Here's how the two apps most stores use handle it, checked on their own pages today:
| Intelligems | Shoplift | |
|---|---|---|
| Price testing | Built in: "Find the right price for every product" (Intelligems on the App Store) | Advanced plan, $399/mo on the Shopify App Store |
| How one buyer keeps one price | Sets Shopify to the highest test price, cart and checkout show each group's own price | Server-side, assigned price renders before the page loads, same URL |
| Google Shopping | Feed shows the highest price, so a shopper may land on a lower one | Single URL, no duplicate listings |
| AI features | "Use AI to discover ideas, build tests, and analyze results in Claude & Slack" | Lift Assist builds page variants (Advanced and up) |
The Intelligems detail matters most. Its price testing FAQ says it sets your Shopify price to the highest test price so Google Shopping, Facebook and Shopify Markets stay accurate. If anyone sees a mismatch, it breaks in the buyer's favor: they expected $59 and found $54. Nobody writes an angry email about that. One honest gap: the same FAQ says Shop Pay installments still show the catalog price.
Shoplift's price testing guide takes the server-side route: one stable URL, no flicker, one price per visitor through checkout.
Want the full head-to-head? Read Shoplift vs Intelligems, or the Intelligems alternatives list if neither fits your budget.
What prices should you actually test?
Fewer than you think, closer together than you want.
Intelligems' own price test walkthrough recommends prices 5 to 10% apart. Under 5% and you can't tell the groups apart. Over 15% and you're testing a fantasy. It suggests a "straddle": your control, one price below, one above.
Now watch what happens on a hypothetical store. A $54 magnesium sleep powder, one jar per order to keep the math clean, $14 product cost per jar. Three groups of 14,000 visitors over three weeks:
- $49 (down 9%): conversion rate 2.6%, average order value $49. Revenue per visitor $1.27. That's 364 orders, $17,836 in revenue, $12,740 gross profit (364 x $35).
- $54 (control): conversion rate 2.4%, average order value $54. Revenue per visitor $1.30. That's 336 orders, $18,144 in revenue, $13,440 gross profit (336 x $40).
- $59 (up 9%): conversion rate 2.3%, average order value $59. Revenue per visitor $1.36. That's 322 orders, $18,998 in revenue, $14,490 gross profit (322 x $45).
14,000 visitors per group. Conversion rate x average order value: 2.6% x $49, 2.4% x $54, 2.3% x $59.
A conversion-only dashboard crowns $49. It "won" at 2.6%. And it's the worst result on the board.
Here's the math that matters. On 60,000 monthly visitors, the control earns $77,760 (60,000 x $1.296). The $59 price earns $81,420 (60,000 x $1.357). That's $3,660 more revenue a month, and $4,500 more gross profit, from changing one number on one page.
The cheapest price in a test almost always wins on conversion rate. That's the trap. You're not paid in conversions.
How do you read a price elasticity test?
Price elasticity sounds like an economics exam. It's one division.
Take the percent change in conversion rate and divide it by the percent change in price. In the hypothetical test, raising the price from $54 to $59 is a 9.3% increase. Conversion fell from 2.4% to 2.3%, a 4.2% drop. That's an elasticity of about -0.45.
Anything between 0 and -1 means buyers barely flinched. The higher price more than pays for the few you lost. Below -1, the price is pushing people away faster than it earns, and you go back to the control.
But here's the thing. With about 330 orders per group, a gap of 2.4% versus 2.3% sits inside the noise. Intelligems' significance docs set 300+ orders per group and 7+ days as the floor, and say hitting that still doesn't guarantee significance. So read this test honestly: it proves the $59 price didn't scare buyers off. It doesn't prove the exact conversion rate. That's enough to roll out $59 and keep watching.
I'll admit something. The first price test I ever ran, I called it on day four. The higher price was "losing." By day 14 it was ahead. Every founder I know has a version of that story. Set the end date before you start and don't touch it.
How do you keep buyers happy while the test runs?
This is the part every tutorial skips, and it's where the screenshots come from.
Intelligems publishes 5 customer experience practices for price tests, and I'd run all of them:
- Keep it short. They recommend 6 weeks or fewer. Longer, and buyers who switch phones land in a new group.
- Brief support before launch. Give the team a code to honor the lower price. A complaint becomes a "we've got you" moment.
- Audit every price touchpoint. Abandoned cart emails in Klaviyo, affiliate posts, ads that print the price. Any of them can show the wrong number.
- Exclude anchored traffic. Leave out repeat customers, or a creator's audience that already saw a price in a video.
- Keep ads price-free. Their walkthrough also says to keep ad copy off specific prices and skip tests during a sale.
And here's my own rule. Never run a price test during Black Friday or any sitewide promo. You'll be testing the discount, not the price.
Why does the page decide whether a higher price wins?
Because buyers don't judge a price. They judge a price against what the page told them.
A spec sheet at $59 loses to the tab next door at $39. A page that answers "why is this one worth more" makes $59 look fair. That's why the price anchoring work goes on the page before the price test goes live. You're giving the higher price something to stand on.
Look at what happened to the Cooling Bamboo Sheets page on one of our client stores. Real client numbers, not typical results, and not a promise of what your store will do. Before: conversion rate 1.0%, average order value $125, revenue per visitor $1.25. After the page rebuild: conversion rate 4.3%, average order value $254, revenue per visitor $10.92. On 10,000 visitors that's $12,500 against $109,200. The full before and after numbers are on our results page.
Read that again. Order value doubled and conversion rate went up at the same time. A page that sells the value lets buyers spend more without flinching. A weak page turns every price test into a test of how weak the page is.
That's also where AI earns its keep today. Paste your last 200 reviews into Claude and ask which complaints are about price and which are about doubt. Doubt is a page problem. Fix that before you test a cent. If you've never run a clean split test before, start with how to A/B test a Shopify product page, then come back to price.
What to do next
Pick your best seller. Write down its conversion rate, its average order value and last month's visitors. Then write the one sentence your page uses to justify the price. If you can't find that sentence, rebuild the page first. If you can, set up a three-price straddle 5 to 10% apart and pick the end date today.
For the full testing playbook, sample sizes included, read the store owner's guide to ecommerce A/B testing. And judge every test on revenue per visitor instead of conversion rate. To pick which product to test first, rank your catalog by profit per product, not revenue.
Book Your Profit Audit
A higher price only wins when the page gives buyers a reason to pay it, and most pages never do. Book a profit audit and we'll show you where your revenue per visitor is leaking, then how to rebuild a high-converting product sales page in less than 15 minutes.
Or go here to check it out → revenueflows.ai
P.S. Discounting asks buyers to forgive a weak page. A price test asks a strong page to prove its worth. Only one of those makes you money.
Frequently asked questions
Is Shopify price testing legal?
Randomized price tests are common in ecommerce, but rules vary by country and state, so check with your own lawyer before you run one. The safe pattern is random assignment, one consistent price per buyer through checkout, and a support team ready to honor the lower price if anyone asks.
How do I A/B test prices on Shopify?
Use a price testing app such as Intelligems or Shoplift's Advanced plan, pick one product, test a control plus one or two prices 5 to 10% apart, keep each buyer on one price from product page to checkout, and judge the winner on revenue per visitor and gross profit, not conversion rate.
How long should a Shopify price test run?
At least two full weeks, ideally three, and long enough to reach about 300 orders per price group. Intelligems recommends keeping price tests to 6 weeks or fewer, because longer tests let buyers switch devices and land in a different group.
Will customers notice if I test prices?
A few might, usually through an abandoned cart email, an ad or a friend's screenshot that shows a different number. Audit every place your price appears before launch and give support a code to honor the lower price, and a noticed test turns into a good service moment.
What is a price elasticity test?
It's a price test read as a ratio: the percent change in conversion rate divided by the percent change in price. If you raise the price 9% and conversion drops 4%, demand is inelastic and the higher price earns more per visitor.