Klaviyo's A/B testing tools are significantly more powerful than most brands use them. Most stop at subject line testing. The brands compounding email revenue year over year test email content, from names, send timing, flow logic, and offer structures systematically. Here is the complete Klaviyo A/B testing playbook.
What Klaviyo Lets You Test
Campaigns: subject line (most commonly tested), preview text, from name, send time, and full email content variants. Flows: Klaviyo allows flow splits where a percentage of subscribers see version A and a percentage see version B. You can test different email content, different offer amounts, different send timing, and different flow branching logic.
Flow A/B testing is significantly more powerful than campaign A/B testing because flows run continuously, accumulating data over time, and directly impact automated revenue. A flow split test comparing a 10 percent discount to a free gift in your abandoned cart email runs until you have statistical significance, potentially over several weeks, and directly affects every abandoned cart your store generates during that period.
Subject Line Testing: The Basics Done Correctly
Klaviyo's campaign A/B test sends variant A to 20 percent, variant B to 20 percent, waits your chosen duration (4 to 8 hours recommended), and sends the winner to the remaining 60 percent. The winning metric can be open rate, click rate, or revenue per recipient. Use revenue per recipient for commercial campaigns. Use open rate for non-promotional content-first emails.
Subject line testing principles: test one variable at a time. "50% off your favourite products" versus "Your exclusive offer expires tonight" tests two variables simultaneously (offer specificity and urgency framing). Run them against a common control instead. True A/B discipline, one variable per test, produces learning that accumulates into a meaningful understanding of what your specific audience responds to.
Flow Testing: Higher Stakes, Higher ROI
Setting up a flow A/B test in Klaviyo: in any flow, add a "Conditional Split" block that routes 50 percent of recipients down path A and 50 percent down path B. Each path has its own email sequence. Monitor the split in Klaviyo's flow analytics under the A/B test section. Statistical significance in flow testing requires patience: you need sufficient order volume through the flow to reach significance.
High-value flow tests: abandoned cart discount amount (10 percent versus free shipping versus no discount), post-purchase cross-sell timing (day 14 versus day 21), welcome flow discount expiry timing (7 days versus 14 days), and win-back offer structure (store credit versus percentage discount). Each of these tests can produce 5 to 20 percent improvement in flow revenue, which compounds across all future sends in that flow indefinitely once you implement the winner.