A/B Test SaaS Pricing: Boost Revenue Without Risk

Worried that tweaking your SaaS pricing will scare away customers and hurt revenue? You can A/B test SaaS pricing safely—using only new traffic—to see real impact without risking existing income.

Over on My Blog I’ve shared a few pricing missteps, and today I’ll walk you through the simple A/B test I ran that kept my revenue steady while delivering actionable data.

Why Founders Fear Pricing Tests

When I first thought about running an A/B test on my pricing, my mind jumped to worst‑case scenarios: higher prices scaring loyal users, lower prices making the brand look cheap, angry support emails, and a sudden dip in monthly recurring revenue. Those fears aren’t irrational; they come from real stories I’ve heard from other founders who saw overnight drops after bold price changes.

I also worried about the technical side—setting up a proper split test felt like it needed a dev team, fancy analytics tools, and weeks of preparation. As a solo founder juggling support, marketing, and product work, I didn’t have bandwidth for a complex experiment. The thought of contaminating the data made me hesitate even more.

Deep down, though, I knew I needed evidence. Guesswork wasn’t getting me anywhere, and I was leaving money on the table simply because I was afraid to try. So I looked for a low‑risk way to test a pricing change that wouldn’t jeopardize my current income.

A/B Test SaaS Pricing: My Low‑Risk Experiment

The test I chose was incredibly straightforward: I duplicated my existing pricing page, changed just one element—the price of the middle tier—and sent only a small slice of new traffic to that version. Existing customers kept seeing the original pricing, so any revenue impact came only from fresh sign‑ups.

I used a simple URL parameter to split the traffic. Half of the visitors who landed on my pricing page via a specific ad campaign saw the original price, and the other half saw the new price I wanted to test. I let the test run for two weeks, which gave me enough conversions to see a clear pattern without dragging things out forever.

During those two weeks I monitored two key metrics: conversion rate for each version and average revenue per user. Surprisingly, the higher price didn’t scare people away. The conversion rate dropped only a few percentage points, while the average revenue per user rose enough to offset that small loss. Overall revenue from the test group stayed flat, meaning I wasn’t losing money while gathering data.

A few factors made this approach feel safe and easy to implement. First, I limited the experiment to new traffic only, protecting my existing base. Second, I kept the change minimal—just one price point—so I could attribute any difference directly to that tweak. Third, I used a basic analytics dashboard I already had; no extra tools were needed. If you’re thinking about trying something similar, start with a tiny traffic slice, test one variable at a time, and watch both conversion rate and average revenue per user.

I’ve written a short follow‑up on My Blog that walks through the exact steps I took to set up the URL split, the metrics I tracked, and how I decided when to roll the change out to all traffic. If you’re curious about the nuts and bolts, head over there for a quick read.

Wrap Up & Thoughts

Testing your SaaS prices doesn’t have to be a scary, all‑or‑nothing gamble. By isolating the experiment to new visitors and keeping the change small, you can learn what works without putting your current revenue at risk. I hope this story shows that a cautious, data‑driven approach is totally doable, even if you’re flying solo.

If you found this useful, consider signing up for the newsletter over at My Blog—I share more practical tips like this every week, straight to your inbox. And if you know a founder who’s been stuck on pricing anxiety, feel free to pass this along. No pressure, just a friendly heads‑up.