SMB Pricing Problems
How Do I Run a Price Test?
Short answer
The simplest price test for a small brand: run a time-boxed promotion at the new price point for 2-4 weeks, measure the impact on units sold and revenue, then decide whether to make the change permanent. You don't need A/B testing infrastructure — a before/after comparison with a clear test window works for most early-stage pricing decisions.
The full answer
Price testing sounds like something that requires a data science team and sophisticated A/B testing infrastructure. It doesn't. Most DTC brands and small retailers can run meaningful price tests with tools they already have. The key is structure: a clear hypothesis, a defined test window, and a pre-committed decision rule.
Method one: time-boxed promotion test. This is the simplest and most common approach. If you think a product might be overpriced, run a 2-4 week promotion at the lower price point. Measure units sold, revenue, and contribution margin during the test period vs. the same period before (or the same period last year, if your business is seasonal). If the volume lift at the lower price generates more total contribution margin, make the change permanent. If not, revert. The promotion framing gives you a clean exit — customers expect promos to end.
Method two: channel-based testing. If you sell through multiple channels (your own website, Amazon, a retail partner), test a different price on one channel. This isn't a perfect A/B test because the customer populations differ, but it gives you directional data. If you raise the price on your DTC site and sales hold steady while Amazon (at the old price) sees no change, that's evidence the higher price is viable.
Method three: cohort-based A/B testing. If you have enough traffic (roughly 1,000+ transactions per week on the tested product), you can run a true A/B test — show different prices to different visitors and compare conversion rates. Shopify apps like Intelligems or custom implementations can do this. The statistical rigor is higher, but the setup cost is too. For most brands under $5M in revenue, methods one or two provide enough signal.
Regardless of method, commit to your decision rule before the test starts. 'If contribution margin is higher at the new price, we switch. If not, we revert.' Write it down. Without a pre-committed rule, you'll rationalize whatever the data shows, and the test was wasted.
One critical mistake to avoid: testing too many changes at once. If you change the price and the packaging and the ad creative in the same week, you can't attribute the result to any single variable. Test one thing at a time. Pricing changes are high-leverage enough to deserve their own clean test window.
Related questions
How long should a price test run?
2-4 weeks minimum. Less than 2 weeks and you're likely seeing noise, not signal — a random good or bad week can dominate a short test. 4 weeks gives you enough data to account for weekly purchasing patterns. If your product has a longer purchase cycle (e.g., monthly replenishment), extend the test to cover at least two purchase cycles.
Should I tell customers I'm testing prices?
No. Price testing works best when it's invisible to the customer. If you're using the promotion method, frame it as a standard promotion. If you're A/B testing, the customer simply sees a price — they don't need to know another cohort sees a different one. Transparency about testing can bias the results.
What if my test shows higher prices don't hurt volume?
That's the most common finding — and the most profitable one. Most brands underestimate their pricing power. If volume holds steady at a higher price, raise it permanently. Then test the next increment. You'll eventually find the ceiling, and every step along the way is incremental margin.
Before running a price test, run a PricePilot report to identify which SKUs have the most pricing headroom — so you test where it matters most. Get your ranked report for $39.