Acceptance Rate Without Cascading: Why First-Attempt Approval Is the Number to Fix

Most payment teams report one acceptance rate — and it usually includes transactions that only went through on the second or third processor. That blended number hides a routing problem: if the cascade is doing the heavy lifting, your first choice of acquirer is wrong too often. Acceptance rate without cascading — the share approved on the first attempt — shows how well your routing actually performs, and it's the number worth fixing first.

Two acceptance rates hiding in one number

A payment can be counted as approved in two different ways. Blended acceptance rate counts it if any attempt succeeded, including a retry on another processor. First-attempt acceptance rate counts it only if the very first authorization was approved. The two are often reported as one number, and they answer different questions (what the business collected versus how well the first route fits the payment).

• First-attempt AR = payments approved on the first attempt ÷ all first attempts × 100.
• Blended AR = payments approved on any attempt ÷ all payments × 100.

A simple illustration (round numbers, not a benchmark): out of 10,000 payments, 7,800 are approved on the first attempt, so first-attempt AR is 78%. Another 700 are declined by the first acquirer and approved on a second or third one, so blended AR is (7,800 + 700) ÷ 10,000 = 85%. The dashboard says 85%, but only 78 points come from your first routing choice. The other 7 percentage points were pulled back by the cascade.

Before comparing any approval rate with a benchmark or another provider's figure, check which attempts it counts (see approval rate in the glossary). A cascade is the automatic re-submission of a declined payment to the next processor (see how cascade logic recovers declined payments for the mechanics).

Why a cascade can mask a routing problem

A cascade turns a wrong first choice into a slower and more expensive approval. The blended figure shows none of the costs below.

• Extra authorization attempts cost money. Acquirers and processors typically charge for every authorization request, approved or not, so a payment that needs three attempts is authorized at up to three times the cost.
• Card scheme rules limit retries. Visa and Mastercard both penalize excessive re-attempts of declined transactions. Acquirer guidance published in 2026 puts Visa's threshold at 20 re-attempts per card and merchant in 30 days for Category 2 decline codes (those that may be approved later), not for every soft decline; since 25 April 2026 the fee for each excess attempt is $0.25, up from $0.15. There are no retries at all after a hard decline. For Mastercard, fees start after 10 attempts in 24 hours or 35 in 30 days on the same card and merchant in standard regions; Brazil and LATAM use a stricter limit of 7 attempts in 24 hours and 35 in 30 days. Thresholds and fees vary by region and change over time, so confirm the current figures with your acquirer. A cascade multiplies attempts and uses this allowance faster.
• Checkout takes longer. Every extra attempt adds a round trip to another processor, and sometimes another authentication step, while the shopper waits.
• Issuer trust suffers. Issuers see every attempt. A card that is declined and re-presented within seconds can look like a pattern, which may lower the odds of approval for later payments on that card.

What raises approval on the first attempt

The fix is to choose the first route using what is known about the payment. The signals that usually move first-attempt approval:

• Issuer BIN and issuer country: send each card range to the acquirer with the best approval history for it.
• Merchant category code (MCC): acquirers and issuers treat categories differently.
• Amount and currency: some connections approve large tickets or specific currencies better than others.
• Local acquiring: a domestic card processed by a local acquirer typically approves more often than one processed cross-border.
• ML balancing on history: shifting traffic between acquirers based on observed approval history instead of fixed percentages.

Set these as routing rules and let ML balancing adjust shares within them. Fraud screening belongs in the same decision: Payneteasy runs 150+ auto-learning fraud filters in the routing layer, so routing choices are made with risk signals in view.

Where cascading still belongs

Cascading is insurance, not a way to raise acceptance rate. It protects revenue against processor outages and soft declines that no first route can predict, and it should stay switched on. Keep its contribution visible as its own line so it does not hide a weak first route. Background: payment cascading explained in the glossary and cascade recovery in practice.

Measure the split

Report first-attempt AR and the share saved by the cascade as two separate lines. In Payneteasy, approval and decline statistics by card type, issuer country, decline reason and processor are available in the dashboard and through the read-only MCP server, and each order shows its routing path and processing steps, including whether the payment went to a second provider. That is enough to build the split from order data, including by asking an AI assistant over MCP. For how that connection stays read-only, see how Payneteasy MCP secures AI access.

Three acceptance metrics compared

How to start measuring first-attempt acceptance

1. Tag every authorization as a first attempt or a cascade attempt, and keep one payment ID across the attempts.
2. Take a baseline: first-attempt AR by BIN range and issuer country over a few weeks of volume.
3. Find the segments where the cascade saves more than a threshold you set. These are the candidates for a different first route.
4. Reconfigure the routing rules for those segments, then measure again: first-attempt AR should rise and the cascade-saved share should fall.