Resources

Braintree Dunning: The Complete Guide

If you bill on Braintree, your recovery depends first on a question many merchants never ask: are your subscriptions actually in Braintree Recurring Billing, or is Braintree only the gateway and Vault underneath a billing system you run yourself? If it is the first, you have a native retry setting, a small set of email notifications, and an optional card-updating service. If it is the second, you have almost none of that until you build it. Either way, Braintree ships payment infrastructure with retry logic attached — not a decline-code recovery system, not outreach beyond email, not cancel deflection, and not one view across every platform you bill on. This guide explains where native Braintree dunning stops and what a dedicated layer adds.

What "Braintree dunning" actually is, out of the box

Dunning is everything after a subscription payment fails: retrying the charge, telling the customer, getting a working card on file, and deciding when to stop. In Braintree, those jobs only exist as a system if you use Recurring Billing — plans in the Control Panel or by API, payment methods in the Vault, subscriptions charged on their billing dates. Its native pieces are:

• Automatic retry logic — configured under Settings, then Account Settings, then Recurring Billing Options, by checking "Automatically retry failed transactions." When a transaction ends in a failed status, Braintree makes three built-in retries before marking the subscription Past Due, which can take up to two days. A processor-declined or gateway-rejected transaction goes Past Due immediately. Once Past Due, Braintree tries billing at least two more times; you set each interval, any value between 1 and 10 days.
• Email notifications for Past Due events — Braintree can email customers for First Decline, First Retry, Second Retry, and Past Due, the last sent a number of days after the first decline that you define. Notifications only fire for events that leave a subscription Past Due. The feature is not simply switched on: the account's authorized signer first contacts Support to request activation, then you create notifications per plan under Subscriptions' Email Notifications tab — sender, subject, plain-text body, and triggering events.
• Account Updater — a separate service that requests updated details for vaulted cards when a card expires or is replaced, so the Vault holds the new number or expiration date. It is not enabled by default; Braintree sets it up with you directly. It covers Braintree Direct merchants based in the US or transacting primarily with US customers, and Visa, Mastercard, and Discover — and whether a card updates depends on the issuing bank participating. Requests go out on a rolling basis when, for example, a card is expired or a recurring payment is due within two weeks.

One more behavior decides what "the end" looks like. When the two Past Due retries have also failed, you choose: cancel the subscription, leave it Past Due, or continue retrying — the default is once per billing cycle, on the usual billing date. Braintree also will not retry certain processor declines it lists as unlikely to succeed, including expired card and possible lost or stolen card codes. It is a fixed sequence with merchant-set gaps — not a different plan per failure.

Where native Braintree dunning stops

None of this is a criticism. Braintree is a gateway first, and the edges show up where recovery stops being a processing problem:

• The intervals are the system. Native logic asks for two numbers between 1 and 10 days. It does not reorganize around why the payment failed, beyond excluding codes it never retries. Insufficient funds on the wrong side of payday, an expired card that needs the customer, a soft decline that just needs time — all run through the same shape. No native playbook treats the decline code as the organizing system for retries and messages together.
• Email is the only native channel, and a narrow one. Four events tied to the Past Due state, a plain-text body, and activation through Support before anything can be configured. There is no SMS step for the customer who never opens email, no voice step for the account worth a call — and no notification for the moment a subscription is cancelled at the end of the cycle.
• Account Updater helps only where it reaches. Opt-in rather than default, limited by merchant type and geography, limited to the card types it supports, and limited again by issuer participation. When the answer is that the account is closed or the customer must be contacted, prevention has done all it can — and no coordinated outreach sequence waits behind it to get a new card.
• There is no cancel deflection in the stack. A subscriber who moves to cancel while a payment problem is unresolved meets whatever cancel flow you built yourself. Braintree's end-of-cycle options are about status — cancel, leave Past Due, keep retrying — not presenting a save offer and applying it.
• It only sees Braintree, and only Braintree's version of it. If some subscribers bill through Recurring Billing, others through your own code on the gateway, and others somewhere else, each world keeps its own failure list. The second group gets a surprise: because their subscriptions do not live in Recurring Billing, the retry logic and notifications above were never running for them at all.
• Attribution is status-centric. Native reporting tells you a subscription's status and transactions — Active again, still Past Due, balance owed. It is not built to separate "recovered by anyone while a system was active" from "recovered because of a specific attempt or message" — the split you need to judge any tool honestly, including this one. See our guide to failed payment recovery software.

What a dedicated recovery layer adds on Braintree

Swipe Again connects to Braintree and runs recovery as one sequence, coordinated with native retries so the same customer is never worked by two competing schedules:

• A decline-code playbook, not a retry setting. Each failure is routed by its code: up to 6 attempts over roughly 16 days for soft declines, payday-aware timing for insufficient funds, a dedicated day 1 / 3 / 7 cadence for expired cards, and no blind retries at all for hard declines like lost or stolen cards — those go straight to asking for a new card.
• Outreach beyond email, in your brand. A branded email you control for each day of the sequence, SMS on day 1 and day 7, and voice calls reserved for high-value accounts — every message carrying a secure, hosted card-update page under your brand.
• Cancel deflection in the same product. Reason-matched save offers — discount, pause, downgrade — presented at the cancel moment and applied to the subscription automatically, included with dunning rather than sold separately.
• Backfill on connect. Connecting pulls in your last 90 days of failed payments automatically, so recovery starts on the backlog — including subscriptions sitting Past Due under the once-per-cycle default — not just tomorrow's failures.
• One dashboard across platforms. If Braintree is one of several places you bill, every failure and recovered payment lands in the same view, with attribution split between Total Recovered and Directly Recovered.

One decision has to be made explicitly before any of this runs: which system owns retries. Braintree's Control Panel logic will keep firing its built-in attempts and its two Past Due intervals underneath any tool you add, and layering a second schedule over it is how customers get duplicate attempts and contradictory emails. Our engine coordinates with native retries and never double-retries — the same rule to apply to any tool you evaluate here, including ours. The discipline is in our dunning management guide; the category is compared in our guide to subscription dunning software.

What it costs, and what the pricing models do at scale

Braintree's retry logic and notifications are features of its recurring billing setup, while Account Updater pricing varies with your pricing model — Braintree directs merchants to contact them for fees. Swipe Again is one flat price — $179/month, or $1,799/year — with dunning and cancel deflection in the same plan and no percentage of recovered revenue, ever. The full breakdown is on the pricing page. Percentage pricing scales with what you recover; a flat fee stays where it started. Run any model at 3× your current failure volume before you sign — the math is in our best dunning software comparison.

And the flat fee carries the guarantee: recover $179 in your first 30 days, guaranteed — or Swipe Again is free the next 30 days. Recovered revenue means every failed payment collected while Swipe Again is active, from any recovery source, as defined in Section 3 of our Terms. For the counting rules behind that promise, see how Swipe Again counts recovered revenue.

Frequently asked questions

Should I turn off Braintree's automatic retry logic?

Not blindly, and not before the coordination decision is made. Swipe Again coordinates with Braintree's native retries so schedules never collide and the same failure is never double-retried. Never leave the Control Panel intervals running underneath a second, uncoordinated engine to compete for the same card.

Does Swipe Again replace Braintree?

No. Braintree remains your gateway, your Vault, and — if you use it — your recurring billing system of record. Swipe Again is the recovery layer that works the failures Braintree surfaces: the decline-code sequence, the outreach, the card-update flow, and cancel deflection.

What if I don't use Braintree Recurring Billing?

Then most of what merchants call "Braintree dunning" was never running for you: automatic retries and Past Due notifications are features of Recurring Billing subscriptions, not of transactions you create yourself through the gateway. A dedicated layer matters more there, not less — the playbook and outreach have to come from somewhere.

How is this different from Account Updater?

Account Updater is prevention: when it works, the vaulted card is refreshed before the customer has to act. It does not reach every card, issuer, or setup, and it sends no outreach when no update is available. Swipe Again runs recovery after a payment has failed — and reports Total Recovered versus Directly Recovered, separately.

The short version

Braintree's native dunning is a configurable retry setting with a small set of Past Due emails and an optional card-updating service — solid gateway machinery, in a gateway's shape. The sequence does not reorganize around the decline code, outreach stops at plain-text email, cancel deflection is not in the stack, and the system only exists for subscriptions in Recurring Billing. If your unrecovered gap is bigger than $179 a month, a dedicated layer closes it — and a 14-day free trial on your own backlog will prove it. For the full playbook, see our complete guide to recovering failed payments.