Payment Definition, Types & Core Concepts

Learn what a payment is, how payment types differ, and why these distinctions matter for system design and compliance.

What Is a Payment?

At its most fundamental level, a payment is the transfer of value from one party to another in satisfaction of an obligation. While this definition sounds straightforward, the mechanics underpinning modern payments involve dozens of institutions, multiple technology layers, and complex legal frameworks operating across milliseconds or multi-day settlement windows.

In the context of financial services, payments are classified along several axes — direction of flow, timing of settlement, reversibility, and the rail (network) over which they travel. Understanding these axes is essential whether you're designing a payment product, writing API documentation, or building compliance controls.

Push vs. Pull Payments

The most fundamental distinction in payments is between push and pull transactions:

• Push payments are initiated by the sender. The payer instructs their bank or payment service provider to move funds to a recipient. Wire transfers, ACH credits, and RTP transfers are push payments. The payer retains control — they decide when and how much to send.
• Pull payments are initiated by the recipient. The payee uses a prior authorization to debit the payer's account. Direct debits, card payments, and ACH debits are pull payments. The payee pulls funds from the payer's account on agreed terms.

This distinction has profound implications for disputes and reversibility. Pull payments can generally be reversed (chargebacks, return items). Push payments, particularly wire transfers, are typically final once the receiving bank credits the beneficiary account.

Payment Types by Rail

Payment "rails" refer to the infrastructure and rules networks over which value travels. Each rail has different characteristics for speed, reversibility, cost, and coverage:

Key Participants in a Payment Transaction

A single card payment at a coffee shop involves at least five distinct entities, each with a defined role and contractual relationship:

• Cardholder (Payer) — The customer whose card is being charged. Holds a relationship with the issuing bank.
• Merchant (Payee) — Accepts the payment. Has a merchant account and processing agreement with an acquirer.
• Issuing Bank (Issuer) — The bank that issued the customer's card. Authorizes or declines transactions and holds the liability for credit issuance.
• Acquiring Bank (Acquirer) — The merchant's bank. Receives the transaction request from the merchant and routes it to the card network for authorization.
• Card Network — Visa, Mastercard, American Express. Operates the rails, sets interchange rules, and routes authorization messages between issuer and acquirer.

When a payment service provider (PSP) like Stripe or Adyen is involved, it typically performs the acquirer function or aggregates processing on behalf of multiple acquirers, adding a technology layer between the merchant and the underlying financial infrastructure.

Payment Timing: When Do Funds Actually Move?

There is often confusion between authorization, clearing, and settlement — three distinct events that collectively complete a payment:

• Authorization (T+0): The issuer confirms the cardholder has sufficient funds/credit and places a hold. Funds do not move yet. This happens in sub-2 seconds for card transactions.
• Clearing (T+0 to T+1): Transaction data is exchanged between acquirer and issuer for settlement calculation. Batch clearing typically occurs overnight via the card network.
• Settlement (T+1 to T+2): Actual fund movement. The network debits the issuer and credits the acquirer. The acquirer credits the merchant (minus interchange and processing fees).

This time lag is economically significant. A merchant's bank account is typically credited 1–2 business days after the transaction, even though the customer's account was debited at the point of authorization. Instant settlement products (RTP, Visa Direct, Mastercard Send) are closing this gap for specific use cases.