How do I run a real pricing experiment to test a higher price without tanking my signups?
The short answer
Test the new price on new traffic only, hold your existing customers at their current rate, and watch conversion and revenue-per-visitor together, not signups alone. A drop in signups with higher total revenue means the price is working, not failing. Give it enough volume and time to be more than noise before you trust the result.
Go deeper, your way
3 hand-picked resources, 3 link-checked. Pick how you want to dig in.
📄 Article
✓ Link checkedFreeIntermediate
Why we picked it
This is the vendor-neutral walkthrough of how to actually structure a pricing test so you can trust the result. It starts with a falsifiable hypothesis, forces you to change one variable, gets sample size right, and reads conversion, ARPU, and retention together instead of celebrating a single number. It also tells you plainly to keep existing customers on their current rate and think through what happens when people compare prices, which is the part most founders skip.
Why we picked it
Most pricing advice stays abstract, so here is a real writeup of an experiment with the numbers left in. They raised the monthly plan from 25 dollars to 33 and 41 dollars, watched monthly signups drop, and still ended up 16 percent higher on revenue per visitor at 99 percent significance because quarterly plans jumped. It is a clean picture of the tradeoff a higher price actually creates, and a reminder to wait for the cohort data before calling it a win.
Why we picked it
Before you touch live traffic, the cheapest way to read price sensitivity is to just ask people, and the Van Westendorp method is the four-question survey that does it. OpinionX is a real tool you can run this on for free, and this guide walks through each question and how to plot the answers into a price corridor. Treat the output as a direction to test, not a final number, then confirm it on real signups.
