Our verdict: MailerLite offers the most test types per dollar
If testing matters to you, MailerLite covers the most ground: subject line, sender, content and send time in standard split tests, a choice of clicks or opens as the winning metric, and multivariate tests on paid plans [1] [2]. Kit's tests are simpler and fixed to 15% test groups, with content testing only on Pro [5] [6] [7]. Moosend documents subject, body, image, CTA and sender tests [8], and its plan comparison marks A/B testing on the trial and every paid plan [11]. beehiiv lists A/B testing on Lite and Pro but its pricing page doesn't say which elements [9].
Who this fits: senders with at least a few hundred recipients per campaign who will act on results. Who should skip testing for now: lists under about 500, where the sample-size check below shows most differences are noise.
What you can test, by platform and plan
Read on each vendor's official pages on 2026-10-29. Not stated means the page we read did not say.
| Platform and plan | Subject | Content | Sender | Send time | Winner chosen by |
|---|---|---|---|---|---|
| MailerLite Free | Yes [1] [3] | Yes [1] [3] | Yes [1] [3] | Yes [1] [3] | Clicks, opens or manual [1] |
| MailerLite Comfort and Power | Yes, plus multivariate up to 8 versions [2] | Yes, multivariate [2] | Yes, multivariate [2] | Yes, multivariate sends to all recipients [2] | Clicks, opens or manual for split tests [1] |
| Kit Free and Creator | 2 subject lines [5] [7] [10] | No [7] | Not stated | Not stated | Open rate only [5] |
| Kit Pro | Up to 5 subject lines [5] [7] | Yes, 2 versions [6] | Not stated | Not stated | Opens for subject tests [5]; clicks for content tests [6] |
| Moosend trial and paid plans | Yes [8] [11] | Body copy, images and CTAs [8] | Sender name [8] | Not stated | A winning parameter you set; test runs up to 24 hours; can send winner automatically [8] |
| beehiiv Free | No [9] | No [9] | No [9] | No [9] | Not applicable |
| beehiiv Lite and Pro | A/B testing listed; elements not stated [9] | Not stated | Not stated | Not stated | Not stated |
How the test groups work
- MailerLite split tests: a slider sets the share of the list that gets A and B, with the winner going to the rest, or you can test across the entire list; the testing period is yours to set [1].
- MailerLite multivariate: up to 3 of 4 variables and 8 versions; lists under 1,000 recipients must send variations to 100% of recipients [2].
- MailerLite automations: A/B split steps in workflows are for Comfort and Power, with up to three paths [4].
- Kit subject tests: two 15% groups (30% total) for two versions, about 7.7% each for three to five versions; the test runs 30 to 240 minutes, default 240 [5].
- Kit content tests: two 15% groups; one type of A/B test at a time, so you can't combine a subject and content test [6].
- Moosend describes its A/B test as two versions that differ in one key variable, and lets a test run for up to 24 hours [8].
Quick sample-size check before you trust a winner
This is rough arithmetic, not a vendor formula. For a click rate p measured on n recipients, a 95% margin of error is about 1.96 times the square root of p times (1 minus p) divided by n. If the two versions' rates are closer together than their margins, don't call a winner.
Example: a Kit two-version test on 1,000 recipients sends each version to 150 people (15%) [5]. At a 5% click rate, the margin is 1.96 times the square root of 0.05 times 0.95 divided by 150, about 3.5 percentage points. Version A at 4% (6 clicks) and version B at 6% (9 clicks) are within that margin: three clicks is not a result. At 5,000 recipients, each group is 750 and the margin shrinks to about 1.6 points.
| Recipients per version | Margin at a 3% click rate | Margin at a 5% click rate | Margin at a 25% open rate |
|---|---|---|---|
| 150 | About 2.7 points | About 3.5 points | About 6.9 points |
| 500 | About 1.5 points | About 1.9 points | About 3.8 points |
| 1,000 | About 1.1 points | About 1.4 points | About 2.7 points |
| 5,000 | About 0.5 points | About 0.6 points | About 1.2 points |
Worked example: choosing a test for a 2,000-subscriber list
Hypothetical newsletter with 2,000 recipients. On Kit Creator, a two-subject-line test puts 300 people in each version, and the winner is picked on opens [5], which Apple Mail Privacy Protection can inflate. On Kit Pro, a content test also uses 300 per version but picks on clicks [6]. On MailerLite, a sender or content split test with the slider at 50% would put 500 in each version and can be judged on clicks [1].
With 500 per version and a 3% click rate, the margin is about 1.5 points, so only a large gap, such as 2% against 5%, is worth acting on. Run the same type of test on several sends and look for a consistent direction before changing your template.
Testing checklist
- Test one variable at a time; Moosend describes its tests as two versions differing in one key variable [8].
- Pick clicks as the winning metric when the platform allows it.
- Write down the hypothesis before sending, such as a question in the subject will get more clicks.
- Check the per-version sample against the margin table before declaring a winner.
- Let the test run long enough to cover your readers' usual reading time; Kit's default is 240 minutes [5].
- Record results in a simple sheet: date, variable, versions, recipients per version, clicks, decision.
Questions before you decide
Which metric should decide the winner?
Clicks, where the platform allows it. Apple Mail Privacy Protection can register opens without a reader, so open-based winners are less reliable. MailerLite lets you choose clicks; Kit's subject line tests use opens only, while its content tests use clicks.
How big does my list need to be?
Kit recommends at least 500 recipients for a two-subject-line test, 750 for three, 1,000 for four and 1,250 for five. MailerLite requires at least 4 subscribers to run a test, but says smaller multivariate groups may give results that are not significant or reliable.
Can I test more than two versions on a free plan?
Not on Kit, where more than two subject lines requires Pro. MailerLite's features page says up to 5 versions of a single element are available to all users, while its split test help article describes A and B versions; multivariate testing needs Comfort or Power.
Sources & verification
Product details and prices can change. Check the linked provider before buying.
- MailerLite Help: How to create an A/B split test campaign Accessed 29 Oct 2026
- MailerLite Help: How to use multivariate testing Accessed 29 Oct 2026
- MailerLite: A/B testing features Accessed 29 Oct 2026
- MailerLite Help: How to use automation steps (Rules and Actions) Accessed 29 Oct 2026
- Kit Help: How to A/B test subject lines Accessed 29 Oct 2026
- Kit Help: How to A/B test email content Accessed 29 Oct 2026
- Kit Help: Creator vs Pro, choose the right plan Accessed 29 Oct 2026
- Moosend: A/B testing Accessed 29 Oct 2026
- beehiiv pricing Accessed 29 Oct 2026
- Kit Help: The Kit Free plan Accessed 29 Oct 2026
- Moosend pricing Accessed 29 Oct 2026
Sources link directly to providers. Product buttons may use separately labeled affiliate links. Read our disclosure.