DEV Community

Swipe Again for Swipe Again

Posted on

Your Dunning Email Has to Survive the Inbox, Not the Editor

Every dunning email is written in an editor and judged in an inbox, and those are different countries. In the editor, your message stands alone: full width, brand colours rendered, the subject line given your complete attention. In the inbox it arrives as one grey line among forty, sent by a billing address the customer half-recognises, previewed in whatever fragment the mail client chooses to show, on a phone, between a delivery notification and a newsletter. Templates that win awards in the editor die quietly in that list. The only test that matters is the second one.

Start with the sender and subject, because for most recipients that is the entire email. If the from-name is a platform default or a no-reply address, the message begins with a stranger asking for money — and strangers asking for money are what spam filters and delete gestures are for. The subject lines that survive are boring in a specific way: they name the product, name the problem, and promise a short path to fixing it. Cleverness is wasted here. A customer scanning a crowded inbox is not looking to be charmed; they are deciding, in under a second, whether this line concerns them. Make the answer obvious.

Then the preview fragment, the accidental first sentence. Many templates open with a logo, a navigation bar, or a line of legal small print, so the inbox preview shows the least useful words in the message while the actual news — your card could not be charged; here is the two-minute fix — sits below the fold of a message that may never be opened. Writing dunning email templates tested in a real inbox means drafting backwards from that preview: the first line of the body is the summary, written to read well as a fragment, because for a large share of recipients the fragment is all that will ever be read.

The body itself has one job per message, and the inbox test enforces it brutally. On a phone, a dunning email that asks the customer to update a card, explains the retry schedule, reassures them about their data, and cross-promotes a feature is four messages wearing one coat, and the thumb chooses none of them. One message, one action, one button that works without a login wall if the platform allows it. Every additional request taxes the single request that pays.

There is a timing dimension to the inbox test as well. An email that lands at 2 a.m. in the customer's timezone is read — if at all — in the morning triage, the most ruthless sorting of the day, where anything non-urgent is swept away in bulk. The same message at a civilised hour joins the ordinary flow of things a person handles. Templates are usually reviewed without a clock on the wall; inboxes have clocks on every wall.

So run the test literally. Send the sequence to a real mailbox, on a real phone, among a day of real mail, and try to complete the card update one-handed with the attention you would actually give it. Time it. Notice where your eye goes and where your thumb hesitates. Every hesitation you feel is a percentage point of recovery your customers will not supply. The editor will tell you the email is clear, complete, and on brand. The inbox tells you whether it works. Believe the inbox, revise until it passes there, and only then let the sequence anywhere near a failing renewal — because that renewal gets one crowded, distracted, half-second hearing, and the template either survives it or the subscription does not.

Published by Swipe Again — failed payment recovery at a flat $179/mo.

Top comments (0)