How Recurring Payments Work
Recurring payments follow a structured lifecycle that begins with a customer's consent and ends with automated charges every billing period — no manual action required from either party after the initial setup.
Initial Authorization
The customer enters payment details at checkout and explicitly agrees to recurring charges. This first transaction is a cardholder-initiated transaction (CIT) and must include consent language — "You authorize [Merchant] to charge your card $49/month until you cancel." The card network rules require this agreement to be explicit and documented.
Credential Storage
The payment processor or vault tokenizes the card, replacing the raw card number (PAN) with a network or processor token. This token is stored against the customer record and used for all future charges. Card-on-file management is critical here — a stale or invalid token is the first cause of recurring payment failures.
Scheduled Billing
On each billing date, the merchant's platform submits a merchant-initiated transaction (MIT) flagged with the stored-credential indicator and the network transaction ID from the original CIT. This flag tells the issuing bank the transaction was authorized by prior agreement — it reduces fraud friction and keeps authorization rates higher than ad-hoc charges.
Decline Handling
When a charge fails, the dunning process begins. Soft declines (insufficient funds, temporary hold) trigger an automated retry sequence — typically 3-5 attempts over 7-14 days. Hard declines (stolen card, closed account) skip retries and trigger cancellation or update flows. Dunning logic is the difference between recovering revenue and losing a subscriber permanently.
Renewal or Cancellation
If all retries succeed, the subscription renews and the cycle restarts. If retries exhaust without success, the subscription enters a grace period or cancels — depending on business rules. Merchants with smart dunning recover 20-40% of initially failed charges within 30 days.
Why Recurring Payments Matter
Predictable, automated revenue is the financial foundation of every subscription business — and recurring payments are the mechanism that makes it possible.
For ecommerce merchants, the shift from one-time sales to recurring revenue dramatically changes unit economics. A customer who buys once generates a single transaction. A subscriber generates revenue every cycle, compounding lifetime value (LTV) over months or years. According to Zuora's Subscription Economy Index, subscription businesses grew revenue 3.7x faster than S&P 500 companies over the past decade.
The risks are real, too. Visa and Mastercard data shows that approximately 10-15% of recurring card charges fail on the first attempt in any given billing cycle — primarily from expired cards, insufficient funds, or bank-side declines. Left unaddressed, those failures become involuntary churn. ProfitWell research found that involuntary churn accounts for 20-40% of total churn across SaaS and subscription ecommerce businesses. That means nearly half of subscriber losses have nothing to do with product dissatisfaction — they're payment infrastructure problems.
Businesses with robust recurring payment infrastructure — smart retries, card updaters, proactive dunning — recover 15-30% of revenue that would otherwise be permanently lost.
Recurring Payments vs. One-Time Payments
Both use the same card rails, but the rules, risk profile, and operational requirements differ significantly.
One-time payments are straightforward: the customer initiates a charge at checkout, the transaction is authorized, and the relationship with that stored credential ends. Recurring payments create an ongoing obligation that must be managed across card reissues, account changes, and customer lifecycle events.
| Recurring Payments | One-Time Payments | |
|---|---|---|
| Initiated by | Merchant (after initial consent) | Cardholder at checkout |
| Credential storage | Required across billing cycles | Optional (guest checkout) |
| Network flags | MIT + stored-credential indicator | CIT, no special flag |
| Decline handling | Dunning + retry sequences | Customer re-enters at point of sale |
| PCI scope | Higher — credential stored long-term | Lower — credential used once |
| Churn risk | Involuntary churn from payment failures | N/A — single transaction |
| Revenue predictability | High — scheduled cash flows | Low — depends on repeat purchase behavior |
Types of Recurring Payments
Not all recurring billing models work the same way, and the right model depends on your product, pricing strategy, and customer behavior.
Fixed recurring billing charges the same amount every cycle — $29/month, $299/year. This is the simplest model, easiest to predict, and lowest-friction for customers. SaaS tools and content subscriptions typically use fixed billing.
Variable recurring billing charges a different amount each cycle based on usage or consumption during the prior period. Utility companies, cloud providers, and usage-based SaaS platforms use this model. Variable billing requires capturing usage data before each billing run and communicating the upcoming charge to reduce disputes.
Metered billing charges based on specific units consumed — API calls, seats, messages sent. The amount varies per cycle but is calculated from measurable, auditable events. Metered billing supports usage-based pricing (UBP), the fastest-growing pricing model in B2B SaaS.
Installment payments split a one-time purchase into a fixed number of equal charges over time (e.g., 3 payments of $100). Unlike true subscriptions, installments have a defined end date. They're technically recurring transactions but governed by different business logic — the merchant must handle pay-in-full scenarios, defaults, and early payoff.
Hybrid billing combines a fixed base fee with variable usage charges — for example, a $49/month platform fee plus $0.01 per transaction processed. Common in payment platforms, telecoms, and B2B SaaS with consumption components.
Best Practices
Getting recurring payments right requires discipline at both the merchant operations level and the technical implementation level.
For merchants:
Send pre-billing notifications 3-7 days before each renewal for charges above $10, or any annual renewal regardless of amount. This is not just a best practice — Visa and Mastercard mandate advance notice for free trial conversions and annual renewals. Customers who know a charge is coming dispute it less.
Make cancellation easy. Friction in the cancellation flow increases involuntary disputes and chargebacks from customers who couldn't figure out how to stop being charged. A simple cancel button reduces disputes more than it increases voluntary churn.
Send failed payment notifications immediately. When a charge fails, email the customer within hours — not days. Include a direct link to update their payment method. The faster the outreach, the higher the recovery rate.
For developers:
Tag every merchant-initiated transaction correctly with the MIT indicator and the network transaction ID from the original authorization. Missing or incorrect flags lead to higher decline rates and potential network rule violations.
Implement exponential back-off in retry logic — not fixed-interval retries. Retrying on day 3, 7, and 14 outperforms daily retries because it avoids repeatedly hitting accounts at the same point in the customer's payment cycle.
Use card account updater (CAU) services from Visa and Mastercard to automatically refresh stored credentials when a card is reissued. CAU enrollment alone can reduce card-related declines by 30-40%.
Never store raw PANs. Use tokenization — either processor tokens or network tokens. Network tokens (Visa Token Service, Mastercard MDES) carry higher authorization rates because they persist across card reissues and carry transaction-level cryptograms that reduce fraud friction.
Common Mistakes
Even experienced merchants make recurring billing errors that quietly drain revenue or create compliance risk.
Not flagging transactions as MITs. Submitting a recurring charge without the stored-credential indicator is a network rule violation. It also means the transaction is processed with the risk profile of a card-not-present one-time charge, which leads to unnecessary declines.
Retrying hard declines. Hard declines (stolen card, closed account, "do not honor — fraud") should not be retried. Repeated attempts on a hard-declined card trigger fraud signals at the issuing bank and can result in the merchant being flagged for suspicious behavior.
Ignoring card expiry proactively. Waiting until a charge fails to discover an expired card is avoidable. Account updater services notify merchants of expiry before the billing date. Not using them means avoidable payment failures at scale.
Annual renewals without notice. Charging a customer $299 without a pre-renewal notification is both a compliance issue (Visa/MC mandate advance notice) and a chargeback magnet. Annual renewals without notice generate dispute rates 3-5x higher than monthly renewals with proper communication.
Flat retry schedules. Retrying every 24 hours on a failed charge wastes retry attempts and increases the risk of the customer canceling or disputing before recovery succeeds. Intelligent timing — targeting post-payday dates, varying the day of week — consistently outperforms naive retry logic.
Recurring Payments and Tagada
Tagada is built for merchants who rely on recurring revenue — subscriptions, memberships, and installment plans across ecommerce and SaaS.
Rather than managing subscription billing in a single processor, Tagada orchestrates recurring charges across multiple payment processors simultaneously. If a card fails at one processor, Tagada can route the retry to a different acquirer, improving recovery rates without additional development work on the merchant side.
How Tagada handles payment recovery
Tagada's dunning engine applies intelligent retry logic — timing retries based on decline codes, card type, and historical recovery patterns — rather than a fixed schedule. Merchants using Tagada's payment recovery typically recover 25-35% of initially failed recurring charges within 30 days.
For stored credentials, Tagada uses network tokenization through Visa Token Service and Mastercard MDES. Tokens stay valid when cards are reissued, reducing the card-expiry declines that silently erode subscription revenue over time. All merchant-initiated transactions submitted through Tagada are automatically flagged with the correct stored-credential indicators and network transaction IDs — removing a common source of decline rate degradation.
Tagada also surfaces real-time subscription health metrics: failed payment rate by processor, dunning recovery rate by cohort, and involuntary churn attribution — giving operations teams the visibility to act before small billing issues become revenue problems.
For merchants building subscription products, Tagada's payment orchestration layer means recurring billing reliability is a solved problem, not a permanent engineering investment.