You've got two lines. Act like it.
The best preview text I shipped this quarter out-earned the subject line it sat under. Same subject, same creative, same send. All I changed was the line many forget to write.
Every email gets two lines in the inbox. The subject, and the preview underneath it. You write the subject, rewrite it, test it, sweat over it. The preview gets whatever's left, or nothing at all. That gap costs you revenue on every single send, and I ran five tests that put a number on it.
What's actually happening out there
The subject gets the full treatment, the preview gets the scraps, and the reader pays for it before they've even opened.
The numbers don't make this subtle
Five tests, five different clients, five funnel positions. Here's what moved.
Outdoor eyewear, browse abandon flow.
I tested a designed two-part inbox unit against a blank preview. Same subject, same send time, same creative below the fold.
Variant | Result |
|---|---|
Blank preview (control) | — |
Designed two-part inbox unit (winner) | +399% RPR |
Before you read that as my copywriting, don't. The control was blank. Beating nothing is the easiest win on the board. Read 399% as the cost of leaving the second line empty, not as clever words.
Oral care, site abandon flow.
The cleanest test I've run in a while. I changed one thing: added an emoji to the preview that shifted the tone and finished the subject line's implication.
Metric | Before | After |
|---|---|---|
Click-to-order rate | 18.2% | 37.9% |
RPR | — | +191% |
One character. When the variable is that narrow, the signal's about as clean as it gets. I didn't expect the gap to be that wide either.
Scrubs, checkout email #4.
Urgency-close preview against brand-voice preview, same subject line.
Variant | Result |
|---|---|
Urgency-close preview | — |
Brand-voice preview (winner) | +21.3% RPR |
Golf training, welcome flow.
Two tests here.
4A, authority-coherent preview vs curiosity-gap: opens identical, RPR +30.9%, click rate +100.8%
4B, offer-explicit preview vs control: RPR +25.4%, and the winner pulled fewer opens than the control
That last line matters. Hold onto it.
The preview isn't an open lever. It's a bouncer.
When the subject and preview work as a unit, they don't just pull opens. They pull the right opens.
Someone who reads a complete two-part hook opens ready to buy. Someone who reads the subject, sees nothing underneath, and opens anyway is just doing inbox triage.
That's why the golf resend is my favourite number in the set. The winning preview got fewer opens and made more money. It was explicit about the offer, so buyers self-selected in and everyone else bounced at the inbox. Fewer opens isn't a bug there. It's the preview doing its job.
How to actually use this
Treat the two lines like one creative brief. Subject is the hook. Preview is the qualifier that tells the right person "yes, this is for you" and tells the wrong person to skip it.
In practice:
✅ Subject is curiosity-driven? Preview resolves one layer of it. Don't stack a second question.
✅ Subject names a product or offer? Preview says why it's worth opening now. Specificity, benefit, scarcity.
✅ Subject is an announcement? Preview gives a reason to open before they're inside, not an echo.
🚫 Never let the preview default. Whatever your ESP auto-fills is the worst option available.
🎯 Aim for 80-100 characters and fill them on purpose.
The test to run: pull your five most-sent flows and read each inbox preview on your phone. If any say "View this email in your browser," "Check out our latest," or nothing at all, that's your first test.
This pattern shows up everywhere
Every account I get into, the subject lines are reasonable and the preview is an afterthought. That's not a brand problem, it's an industry default.
Your data stays yours. The patterns travel. This one showed up across browse abandon, site abandon, checkout, and welcome. Four funnel positions, same structural gap, same two lines going to waste.
If you want to know what that looks like on your specific flows, that's the work I do. Reply to this one, or hit the button below.
