What Is Dunning?
Dunning is the process of communicating with customers to recover overdue or failed payments. In the SaaS and subscription business context, dunning refers specifically to the automated system that handles failed subscription payment recovery — the sequence of payment retries, email notifications, and card-update prompts that kick in when a customer's recurring charge doesn't go through.
When a customer's credit card expires, their bank flags a transaction as suspicious, or their account temporarily runs short, their subscription payment fails. Without dunning, that subscriber quietly falls off — a form of churn the customer never intended. Dunning software intercepts that failure, classifies why it happened, and triggers a recovery workflow designed to get the subscription back on track with minimal disruption to the customer relationship.
Dunning is sometimes called failed payment recovery, payment retry automation, or involuntary churn prevention. All three phrases describe the same core process.
Failed card payments are the silent revenue leak most SaaS founders don't notice until they look at the numbers.
Where the Term "Dunning" Comes From
The word "dunning" dates back to 17th-century England. A "dun" was slang for a debt collector, and "dunning" came to describe the act of persistently demanding payment from someone who owed money. The term entered formal financial lexicon to describe any series of increasingly urgent communications sent to collect a past-due account.
In the modern SaaS world, the tone is entirely different from its debt-collection origins. SaaS dunning isn't adversarial — the customer usually had no idea their payment failed. The best dunning sequences are helpful, empathetic, and informational. They explain what happened, reassure the customer their account is fine, and make updating payment details as frictionless as possible. The historical connotation of aggressive collection doesn't apply.
Why Dunning Matters for SaaS
Failed payments are the most underappreciated revenue leak in subscription businesses. According to research from Stripe, Baremetrics, and ChartMogul, the industry average for involuntary churn sits between 7–9% of MRR per month — meaning nearly 1 in 10 subscription dollars is being lost to payment failures at any given time.
Average MRR lost monthly to failed payments in SaaS businesses — most of it recoverable with the right dunning system in place.
What makes this loss particularly painful is that it's largely preventable. Unlike voluntary churn — where a customer consciously decides your product isn't worth the price — involuntary churn involves customers who intended to stay. Their credit card expired, their bank blocked an international charge, or their account temporarily had insufficient funds. These are recoverable situations if you act quickly and communicate the right message.
The math becomes compelling very quickly. At $30K MRR, if 9% fails monthly ($2,700) and you recover 60% of that ($1,620/mo), that's $19,440 recovered per year. A dunning tool at $49–$149/mo pays for itself in the first week of every month. And unlike growth tactics that require constant marketing spend, dunning recovery is pure margin improvement — it doesn't cost you anything to re-activate a customer who was already paying.
How Dunning Works
Modern dunning software operates in two parallel tracks that work together:
- Payment retry logic — automatically rechallenging failed payments at optimized intervals, using ML models that analyze decline codes, time of day, card type, and bank behavior to find the best retry window
- Customer communication sequences — sending personalized emails (and sometimes SMS) to notify the customer of the failure and guide them to update their payment details or take action with their bank
When a subscription charge fails, the dunning system receives a webhook from the payment processor (Stripe, Braintree, Chargebee, etc.) containing the failure details, including a decline code that describes why the payment failed. The dunning system uses this code — along with historical patterns for that card type, bank, and customer — to determine the best response.
Some failures resolve on their own through retries (insufficient funds that clear in a day or two). Others require customer action (expired cards need to be updated; fraud-blocked charges need the customer to call their bank). The best dunning software distinguishes between these scenarios and handles them differently — which is why failure reason classification is such a critical capability.
Understanding Payment Failure Reasons
Not all failed payments are the same, and treating them the same is one of the costliest mistakes in dunning. Here are the four primary failure categories:
1. Expired Card
The customer's card expired and their bank declined the charge. The customer may not even know their card expired — banks don't always send proactive notifications. The correct response: tell the customer exactly which card expired and give them a one-tap link to update their payment method. Retrying without informing the customer accomplishes nothing.
2. Insufficient Funds
The customer's account didn't have enough funds at the time of the charge. This is usually temporary — their paycheck clears in two days, or they paid a large bill that week. The correct response: reassure the customer their account is safe, let them know the charge will retry automatically, and give them a way to manually trigger a retry once they're ready. Telling this customer to "update their card" is confusing and unnecessary.
3. Bank Decline (Do Not Honor)
The customer's bank declined the charge for a vague reason — often a risk flag for an unfamiliar merchant or international transaction. The correct response: explain that their bank blocked the charge, tell them what to say when they call their bank ("please whitelist a recurring charge from [merchant name]"), and give them a callback link for when it's resolved.
4. Fraud Block / Suspected Fraud
The bank or card network flagged the transaction as potentially fraudulent — even though it's a legitimate recurring subscription. Similar to a bank decline but with more urgency. The correct response: explain clearly that this is a known subscription charge, provide specific instructions for contacting their bank, and offer a temporary alternative payment method.
Modern dunning platforms surface recovery analytics so you can see exactly what's being recovered and why.
Anatomy of a Dunning Sequence
A well-designed dunning sequence typically spans 7–21 days from the initial failure. Here's what a best-practice sequence looks like:
- Day 0 (immediate): Failure detected. Email sent within 1–2 hours, classified by failure type. Retry attempted if it's a temporary failure type (insufficient funds, soft decline).
- Day 2–3: Second retry attempt. Follow-up email if the first email went unopened. Different subject line, gentle escalation in urgency.
- Day 5–7: Third retry. Email with card update link for failure types that require customer action. Link should be deep — no login required, direct to payment update screen.
- Day 10–12: Escalation email. Mention the specific service at risk. Make it personal. "Your [plan name] subscription will pause in [X] days."
- Day 14–16: Final retry attempt with ML-optimized timing. Last-chance email with urgency framing. Offer live support chat if available.
- Day 19–21: Subscription paused or cancelled. Off-boarding email with easy reactivation link (you may still recover this customer later).
The exact cadence varies by business model, average contract value, and customer relationship. Enterprise SaaS with annual plans may run longer sequences; high-volume consumer subscriptions may compress to 7 days to reduce cart abandonment-style friction.
Smart Retries vs. AI-Personalized Emails
These two capabilities are often conflated but solve different problems:
Smart retries operate at the payment processor layer. They use machine learning to analyze when a retry is most likely to succeed — considering the time of day, day of week, historical behavior for that bank and card type, and the specific decline code. A smart retry might wait until Tuesday morning to rechallenge a declined card that statistically clears more often on weekday mornings. Smart retries are invisible to the customer and recover payments without any customer action.
AI-personalized emails operate at the customer communication layer. They're the mechanism for recovering failures that retries alone can't fix — expired cards, hard declines, and fraud blocks. AI personalization goes beyond inserting the customer's name: it tailors the entire email — the explanation, the call-to-action, the tone — based on why the payment failed and what the customer needs to do. AI-written emails that match the failure context achieve dramatically higher open rates and click-through rates than template emails.
The highest-performing dunning systems combine both: smart retries to handle recoverable failures silently, and AI-personalized emails to communicate contextually with customers who need to take action. See our full comparison of dunning tools that combine both approaches.
What Recovery Rates to Expect
Recovery rates vary significantly based on the quality of the dunning system:
- No dunning (Stripe defaults only): 10–15% of failed payments recovered
- Basic dunning (generic retries + template emails): 15–30%
- Smart retries + template emails: 25–40%
- Smart retries + failure-reason classification + AI emails: 55–70%+
The difference between 20% and 67% recovery doesn't sound dramatic until you run the numbers on real MRR. At $100K MRR with 9% monthly failure ($9,000/mo), moving from 20% recovery ($1,800) to 67% recovery ($6,030) is a difference of $4,230/mo — or $50,760/year. That's revenue that exists in your customer base right now, waiting to be recovered.
Industry benchmarks from Baremetrics and ChartMogul consistently support the 9% monthly failure rate across subscription businesses, and leading dunning tools like Churn Buster report 50.3% average recovery rates. Tools with failure reason classification and AI personalization push that to 67%+.
Does Stripe Have Built-In Dunning?
Yes — Stripe includes two built-in dunning mechanisms: Smart Retries and customer emails.
Stripe's Smart Retries use ML to optimize retry timing and are genuinely useful — they handle the payment retry layer reasonably well. Stripe's built-in email reminders are configurable in the Stripe Dashboard: you can enable pre-renewal reminders and failed payment emails with card update links.
The limitations are real, though. Stripe's emails are:
- Generic — the same email goes to every customer regardless of why their payment failed
- Unbranded or minimally branded — they look like system notifications, not brand communications
- Low-personalization — no failure reason context, no intelligent sequence escalation
- Non-analyzable — Stripe doesn't provide dunning-specific recovery analytics dashboards
For most businesses above $10K MRR, a dedicated dunning tool that plugs into Stripe via OAuth will recover significantly more revenue than Stripe's defaults — typically paying for itself by the first recovery of the month. Learn more about involuntary churn and how to stop it.
Stop losing revenue to failed payments
DunningBee classifies every failed payment by type and sends AI-written emails matched to each reason. 67% average recovery rate. Takes 2 minutes to set up on Stripe.
Try DunningBee Free for 14 Days →Frequently Asked Questions
Related reading: What is involuntary churn? · Best dunning software 2026 · DunningBee review