Subject-Line A/B Testing for Lifecycle Emails

A lifecycle email's subject line is the one variable you get almost no signal on from a single send. Add up to 4 more subject variants to the main one, and each recipient is deterministically assigned a variant based on their email address — stable across retries and resends — so open rate and response rate show up side by side on the Responses dashboard once the send goes out.

Subject

Your trial ends soon

A lifecycle email's subject line is the one variable you usually get no real signal on from a single send. Add up to 4 more subject variants to the main one, and each recipient is deterministically assigned a variant based on their email address — stable across retries — so open rate and response rate show up side by side once the send goes out.

A trial-expiry or renewal email only gets one real shot per recipient — testing the subject line on that one shot, rather than guessing, is worth the extra thirty seconds of setup.

Test up to 5 subject lines at once

Add up to 4 more variants under the main subject field on a one-time send — 2 to 5 total. Each recipient's assignment is a deterministic hash of their email address, so a retried or resent message never reshuffles who saw which line.

Read the results side by side

The Responses page adds a variant comparison once a multi-subject send goes out — open rate and response rate per variant, with the current leader flagged, cumulative across every A/B send the project has ever run rather than scoped to just the latest one.

Carry the winner into your ongoing cadence

Recurring sends stay single-subject — there's no variant picker on the schedule form. Once a variant wins on a one-time test, set it as the single subject line on the recurring schedule that reaches this lifecycle stage going forward.

It rides along for free downstream

The assigned variant index lives in the recipient's own live-view token, so every interaction event and project webhook payload for that recipient already carries it — no extra joins needed to see whether a winning subject line also produced more clicks or purchases, not just opens.

Getting started

On a one-time lifecycle send (a trial-expiry or renewal reminder), add 2-5 subject-line variants and send it. Check the variant comparison on the Responses dashboard, then set the winning subject as the single line on your recurring schedule for that same lifecycle stage.

A typical build-and-send sequence

  1. 1

    Add up to 4 more subject-line variants

    Under the main subject field on a one-time send, add 2 to 5 total subject lines to test against each other.

  2. 2

    Send it

    Each recipient is deterministically assigned one variant based on their email address — stable across retries so nobody sees a different subject on a resend.

  3. 3

    Read the results

    The Responses page adds a variant comparison — open rate and response rate side by side, with the current leader flagged, cumulative across every A/B send the project has run.

  4. 4

    Carry the winner into your recurring send

    Recurring sends stay single-subject — once a variant wins, set it as the one subject line on the schedule that reaches this lifecycle stage going forward.

Frequently asked questions

Can a recurring/scheduled send also run subject-line A/B variants?

No — recurring sends stay single-subject; there's no variant picker on the schedule form. A/B testing runs on a one-time send, and the winning subject line then gets set as the single subject on the recurring schedule.

How is each recipient assigned a subject-line variant?

Deterministically, from a hash of their email address — the same recipient always gets the same variant, even across retries or a resend, so results aren't skewed by re-shuffling.

How many subject lines can I test at once?

Up to 5 total (the main subject plus up to 4 more variants) — 2 to 5 subject lines split-test against each other on a single send.

Do the results only cover the most recent A/B send, or all of them?

Results are cumulative for the project — the variant comparison on the Responses page covers every A/B send the project has ever run, not just the most recent one, with labels supplied by the most recent multi-subject send.

Does the subject-line variant affect anything else, like which webhook payload a recipient's events carry?

Yes — the assigned variant index rides in the recipient's own live-view token, so it appears on every downstream InteractionEvent and project webhook payload for that recipient with zero extra lookups.

Build this in the studio

Start on the free plan — every interactive block and the full fallback engine are included on every tier.