Preview emails as specific subscribers

For most authors, previewing your email is casual—you might be checking some design details or tracing formatting issues first-hand. After a while, you've seen it all. But then there are the more nuanced cases: testing how your newsletter looks for different subscriber types, verifying paywall behavior, or making sure CTAs render correctly—which is why we added earlier this year the ability to view the paywall version of paid emails. Then there are the more esoteric cases where you want to test templating functionality or tags. Historically, the only way to do this was to actually send test emails to fake subscribers you've created, which match whatever state you're trying to test. This is time-consuming and cumbersome. Finally, though, we've got a better option.

You can now select any subscriber from your list directly in the preview panel. When you do, the email renders exactly as it would for that subscriber—using their metadata, subscription status, tags, and all the other personalization that makes your emails feel personal. No more creating fake test subscribers or sending yourself drafts just to see how {{ subscriber.metadata.first_name }} resolves for different people.

This works for both the email and web previews, and it's particularly useful for:

• Testing personalization tags with real subscriber data
• Verifying how premium content renders for different subscription types
• Checking that conditional logic works correctly (like showing different content based on tags or metadata)
• Seeing how your email looks for subscribers with different locales or custom metadata

The subscriber selector appears in the preview panel when you're editing an email. Just click the dropdown, search for a subscriber by email address, and select them. The preview updates instantly to show how that subscriber would see your email.

This replaces the old "free vs. premium" preview split—now you can preview as any subscriber, not just generic free or premium ones. Much more flexible, and way less work than the old approach.