A/B testing
A/B testing in outreach means sending two versions of a message to similar audiences and comparing results, to learn which wording, offer, or timing performs better.
You change one element, such as the opening line or the call to action, and send version A to one group and version B to another. Then you compare a metric like reply rate or acceptance rate.
For the result to mean anything, the groups must be similar, the sample large enough to be stable, and only one thing should differ. A handful of sends can mislead, and changing several things at once hides what actually caused the difference.
Keep a record of what you tested and what you learned. Over time this builds a baseline that is specific to your audience, which is more useful than benchmarks from other companies.
Example
Example with made-up numbers: version A sends a connection request with no note and version B sends a short specific note. If A is accepted 28 times out of 100 and B 33 times out of 100, B looks better, but a larger sample is needed before you rely on a gap this small.
Common questions
How many sends do you need for an A/B test?
Enough that results are stable across batches. Small samples can mislead, so keep testing until the numbers settle.
What should you test first?
Start with the element likely to matter most, such as the opening line, the size of the ask, or whether to include a note.
Related terms
See FollowArc run your outreach
An autonomous AI sales agent for LinkedIn + email — you stay in control.
Start your 7-day free trial7-day free trial. Add a card to start, $0 today, cancel before day 7 and pay nothing.