Dunning emails are the customer communications a subscription business sends after a payment fails, with one job: get the customer to update their billing details in fewer than two clicks. Done well, they're the single highest-leverage retention tool in subscription billing. Done badly, they read like collections notices and chase customers out the door.
Across the 148 Baremetrics customers using Recover for dunning automation, the median customer earned 410% ROI on their Baremetrics subscription in a single month. 82% of those customers saw Recover pay for itself within month one. The difference between the customers in the 82% and the rest is rarely the underlying retry logic — it's the emails.
This guide covers everything you need to write dunning emails that work: subject line patterns, body structure, CTA design, the optimal cadence (7 emails over 30 days), how SMS fits into the flow, and how to personalise messaging by failure reason. Plus real templates you can copy.
Subscription businesses lose an average of 9% of MRR to failed payments every month. For a business at $100K MRR, that's $9,000 walking out the door — every month — without any customer deciding to leave.
The customers who churn this way didn't pick a competitor. They didn't get a better deal. Their card expired or was reissued by their bank, the next charge failed, and nobody told them in a way that registered. The bigger framing: research from Paddle suggests involuntary churn accounts for 20–40% of total churn in subscription businesses. That's the share of churn that has nothing to do with product satisfaction or pricing — and almost all of it is preventable with the right dunning email sequence.
Dunning emails are how you reach into that 20–40% and pull customers back.
Most dunning emails underperform for one reason: they sound like collections notices. "FAILED PAYMENT." "FINAL NOTICE." "ACTION REQUIRED."
These patterns kill recovery rate. They feel adversarial. They lower trust. They get marked as spam. And — most importantly — they treat the customer like a delinquent debtor instead of a person whose card expired.
The most effective dunning emails read like a helpful CX team flagging an admin issue. Polite, neutral, low-friction, and specific about what to do next.
"Nobody wants to hear that their credit card has failed and they're delinquent on something. Because Recover is so easy to use and customizable, it allows our CX team to handle the payment recovery process instead of marketing, which customers seem to appreciate." — Matt Gartland, CEO, SPI Media (Smart Passive Income)
That framing — dunning as customer experience, not as billing — is the single most predictive factor for recovery rate. Every choice we make below about subject lines, body copy, CTAs, and cadence flows from it.
A dunning email has more moving parts than it looks. To convert, all five need to work.
The subject has one job: get opened. On the typical phone screen, your customer sees ~35 characters. You need to signal the problem and the solution simultaneously.
Patterns that work:
Patterns to avoid:
If your subject line could be sent by a debt collector, rewrite it.
Three sentences is often enough.
The 3-sentence template:
That's it. Long emails are worse emails. Customers skim. The whole message should be readable in the preview pane without scrolling.
A variation, tone-matched to a B2B SaaS:
Hi [First name], Quick heads-up — the card we have on file for [Workspace name] couldn't be charged for this month's subscription. It's usually an expired card or a recent reissue from your bank. Updating it takes about 30 seconds: [Update billing] Once that's done, we'll retry the charge automatically. Nothing else needed on your end. Thanks, The [Product] team
A variation, tone-matched to a subscription community/DTC:
Hi [First name], Just a quick note — your monthly [Product] payment didn't go through. Almost always it's an expired card or a small banking issue, easy to fix. Here's the one-click update: [Update billing] If you've got any questions about your subscription, just reply to this email and we'll help. — [Founder / CX team name]
Notice what's NOT in either: marketing language, urgency manipulation, threats about service interruption, multiple CTAs, footer noise, promotional links.
The CTA is the entire reason the email exists. Make it impossible to miss and impossible to misuse.
Rules:
If you use Recover, the hosted payment-update URL handles this — customers land directly on a one-click form, no portal navigation. If you're rolling your own, audit the path from inbox to updated card and count the clicks. More than two means you're losing customers at the friction.
When you send matters almost as much as what you send.
Branded dunning emails outperform plain-text ones for trust and engagement, even though plain-text feels more "personal." Counter-intuitive but real: customers want to know the email is genuinely from the company they're paying.
Recover lets you control all of this without writing code — branding, sender identity, colors, and template. If your current dunning system is rendering plain "your payment failed" emails branded as Stripe/Braintree/etc., you're missing recovery on every send.
A single dunning email recovers a fraction of what a properly sequenced flow does. Most failed payments aren't ignored on purpose — they're missed. People are busy, emails get buried, and customers genuinely don't notice their card expired.
Recover ships a 7-email sequence by default, running from the day of failure through 13 days delinquent. Here's the full cadence with the purpose of each step:
| # | Day | Tone | Purpose | Key element |
|---|---|---|---|---|
| 1 | 0 | Friendly | "Hey, your payment didn't go through" | Sent within 24h of failure |
| 2 | 3 | Friendly | Gentle reminder; surface common reasons | "Probably just an expired card" framing |
| 3 | 7 | Direct | Surface the consequence (service may pause) | Light urgency, still helpful tone |
| 4 | 10 | Direct | Account-level urgency | Mention any features they'll lose access to |
| 5 | 13 | Direct | Last chance before lockout | Lands before the typical day-15 cutoff |
| 6 | 20 | Re-engagement | Account paused; offer help | Survey-style; ask if anything's wrong |
| 7 | 27 | Final | Final notice before hard cancellation | Calm, factual, with one final update link |
The 13-day email is the most important email in the entire sequence. Most subscription businesses suspend access at day 15 — locking customers out of the product they were paying for. The day-13 email lands right before that lockout, giving customers one final chance to recover their account before involuntary churn becomes irreversible. Skipping this step is the most common mistake we see in dunning sequences. Add it and recovery rate moves immediately.
Email 1 (Day 0):
Subject: Quick fix needed for your [Product] subscription Hi [First name], looks like the card we have on file for [Workspace] didn't go through this month. Usually it's an expired card or a recent bank reissue. Update takes 30 seconds: [Update billing]
Email 2 (Day 3):
Subject: Still trying to charge your card Just a heads-up — we tried again, and the payment for [Product] still isn't going through. Most often this is a card expiry or insufficient funds issue. Easy fix here: [Update billing]
Email 3 (Day 7):
Subject: Your [Product] subscription needs attention [First name], we haven't been able to charge your card for the last week. Your subscription is still active right now, but it'll be paused soon if we can't get the payment through. Here's the one-click update: [Update billing]
Email 5 (Day 13) — the critical one:
Subject: Your subscription pauses tomorrow Hi [First name], unless we can charge your card in the next 24 hours, your [Product] subscription will pause and you'll lose access to [specific feature]. The fix takes 30 seconds: [Update billing]. If you'd like to cancel or downgrade instead, just reply to this email.
Email 7 (Day 27):
Subject: Last chance — closing your [Product] account Hi [First name], we've tried to charge your card several times over the past month without success. We'll close your account at the end of this week unless we hear from you. If this was an admin issue, you can still update your card here: [Update billing]. If you'd prefer to cancel cleanly, reply and we'll take care of it.
The pattern across all seven: factual, neutral tone, one CTA, low-friction update path, optional support link for customers who need help.
Don't squeeze the cadence too tight. Emails 1 and 2 within 3 days starts to feel like badgering; emails 3 and 4 spaced 7+ days apart feel disconnected. The cadence above (0, 3, 7, 10, 13, 20, 27) balances persistence with respect for the customer's inbox.
If you're using Recover, the spacing and timing are fully customisable per step — adjust to your audience's specific behaviour patterns. If you're using a payment-processor's default dunning, you typically don't have this granularity, which is one of the bigger limitations of native processor dunning vs a dedicated layer.
Email is the primary channel for dunning. But email open rates drop sharply as a sequence progresses — the customer who didn't open email 3 probably isn't going to open emails 4, 5, 6, or 7 either. That's where SMS earns its place.
Baremetrics Recover ships SMS as a first-class dunning channel alongside email — not a Twilio bolt-on, not a separate flow. SMS is layered into the same sequence with the same customisation controls.
When SMS works best:
SMS template patterns:
[Product]: Your subscription pauses in 24h. Update your card here: [short URL]. Reply STOP to unsubscribe.
Hi [First name], [Product] couldn't process your payment. Update billing: [short URL] or reply HELP.
Two rules for dunning SMS:
SMS isn't a replacement for the email sequence — it's an amplifier on the steps where email engagement is lowest.
A card declined for insufficient funds is a completely different problem from an expired card, which is a completely different problem from a geographic restriction. Sending the same email to all three wastes the recovery opportunity.
The three most common failure reasons and how to write for each:
This is the most common cause and the easiest to fix. The customer needs a 30-second update.
Subject: Your [Product] card update Looks like the card we have on file for [Product] has expired. Update takes 30 seconds: [Update billing]. We'll retry the charge automatically once your new details are saved.
Lead with the cause, lead with the fix. No drama needed.
This is often a temporary cash-flow issue — payday hasn't hit, or the bank is reviewing the charge. Often resolves itself within 24–48 hours via smart retries before any customer action is needed.
For the email that does fire:
Subject: We'll try again soon Hi [First name], we couldn't charge your card today — looks like a temporary issue. We'll automatically try again in a couple of days. If you'd like to update the card on file or use a different one, you can do that here: [Update billing].
Empathy + retry transparency + optional self-service. Don't make this feel like a problem the customer urgently caused.
Card was declined by the issuer (bank), not by your processor. The customer needs to contact their bank.
Subject: Your bank declined this charge Hi [First name], your bank's authorising the charge to [Product] was declined — this typically happens when the bank flags an international or unusual transaction. The easiest fix is a quick call to the number on the back of your card to clear the charge, then we'll retry automatically. Alternatively, you can update to a different payment method: [Update billing].
Specific about the cause + practical about the fix + alternative for customers who can't or won't call their bank.
If you're using Recover, you can configure separate email content per failure reason — the system identifies the failure type from the payment processor response and triggers the matching template automatically.
Some customers won't recover, no matter how good your emails are. Sometimes the card issue is real — they genuinely can't afford the subscription anymore, or they're going through something difficult. The right move at the end of the sequence is to give them options other than "pay or be cancelled."
Effective alternatives to include in the last 2–3 emails:
Nifty gives customers a few ways to get more help aside from updating their billing details — including direct links to their billing FAQ and a chat option. Juicer does similar, pointing customers to their declined-payments FAQ first. Both are good examples of the "alternative options" close done well.
Three published examples worth studying:
Signable — short and respectful, with the update link prominent. Good example of the 3-sentence body.
CanIRank — strong subject line pattern; treats the customer like a person, not a debtor.
SoapBox — clean structure, single CTA, brand voice intact.
For more template examples (over 30 real-world SaaS dunning emails), the full breakdown is in our dunning best practices guide.
Once you have a dunning email sequence running, the metrics that matter most are:
Recover provides all of these out of the box. Whatever tool you use, slice recovery rate by step and by failure reason — averages hide the actionable signal.
Two options for any subscription business:
Stripe, Braintree, Recurly, and Chargebee all ship basic dunning capabilities — typically 3–5 emails, limited customisation, no SMS, no segment-based exclusion. Free if you're already using the processor. The right choice if you're early-stage and need something running today.
Limitations: limited customisation, billing-system-toned defaults, no segment exclusion, no FAQ-by-failure-reason templates.
A dedicated tool that layers on top of your payment processor. Customisable email sequence, SMS channel, in-app reminders, paywalls, hosted payment-update URL, segment-based exclusion, and attempted-recovery-rate metrics.
Across the 148 businesses using Recover in December 2024, the median customer earned 410% ROI on their Baremetrics subscription in a single month. The aggregate recovery across that month was over $1.35 million. Best for businesses where customer-experience tone matters more than the few hours of setup time.
Start a free trial of Baremetrics Recover →
A standard sequence is 6–7 emails over 27–30 days, weighted toward the first two weeks after a failed payment. Recover's default sends 7 emails, with the most important one landing at 13 days delinquent — right before most subscription businesses suspend access at day 15. Single-email dunning recovers a fraction of what a properly sequenced flow does.
Within 24 hours of the failed payment. The first email is the highest-converting one in the sequence — the issue is fresh in the customer's mind, the card is probably still nearby, and the cause (often an expired card) is easy to fix. Delaying the first email by even a day measurably lowers recovery rate.
Branded. Counter-intuitive but true: customers want to see your logo, your brand colours, and a named sender — it confirms the email is genuinely from a company they're paying. Plain-text emails feel more "personal" to senders but actually lower trust on the receiving end. Branded HTML emails recover better.
The day-13 email — the one that lands just before most subscription businesses lock customers out at day 15. It's the customer's final practical chance to update their billing before involuntary churn becomes irreversible. Many dunning sequences skip this step entirely; adding it back is one of the most reliable ways to lift recovery rate.
Signal both the problem and the solution in 35 characters or fewer (mobile preview length). Patterns that work: "Quick fix needed for your [Product] subscription", "Your card expired — here's a one-click update", "Action needed: payment didn't go through". Avoid all-caps urgency ("FAILED PAYMENT", "OVERDUE") which lowers open rates and triggers spam filters.
No for the early cadence (days 0–7). Recovery rate dips materially for weekend sends — customers are checking inboxes less, and the email goes stale by Monday. Yes for later cadence (days 13+) where urgency is the point. Tuesday–Thursday, 9–11am local time, is the highest-recovery window for the first few emails.
Yes — especially for later-cadence emails (day 7+) where email engagement starts to drop. SMS open rates are typically 90%+ vs 20-30% for email, but the channel is best used as an amplifier on the same cadence, not a replacement for the email sequence. Recover ships SMS as a first-class channel alongside email.
No. Pause the subscription instead. Pausing preserves the customer's data, subscription history, and re-activation path; cancellation throws all of that out and makes recovery harder if the customer comes back later. Hard cancellation should typically only happen after 30 days of failed recovery attempts.
Most retention work focuses on the customers who actively cancel — voluntary churn. Important, but not where the easiest revenue is. The easiest revenue is the 20–40% of churn that's involuntary: customers who never decided to leave, whose subscriptions ended because of a payment failure that nobody surfaced to them in a way that registered.
Dunning emails are the single most direct way to reach into that revenue and pull it back. The patterns above — short body, clear CTA, branded identity, 7-email sequence, day-13 last-chance email, SMS amplifier on the later steps, personalisation by failure reason — aren't theoretical. They're what Baremetrics customers using Recover have used to recover $1.35M in a single month at a median 410% ROI on their subscription.
If your current dunning setup is your payment processor's defaults, you're almost certainly leaving recoverable revenue on the table. Even small upgrades — a customised template, an extra email at day 13, a single SMS at day 7 — move the metric measurably.
Start a free trial of Baremetrics Recover →