Card networks publish the same fee schedule to everyone. The cost per dollar of volume isn't. That's a scale penalty smaller issuers and acquirers pay without anyone calling it a penalty.
Four structural mechanics create it
Fixed annual, licensing, and registration fees
Acquirers pay fixed license fees, FANF, and specialty registrations. Issuers pay base licensing fees, BIN maintenance, and access minimums. For the same licensing and registration footprint, a $500M institution pays roughly the same baseline as a $5B institution. Per dollar of volume, the smaller book bears it disproportionately.
Tier-based pricing on optional programs
Many publish volume-tier discounts. Smaller participants pay the standard rate while larger ones get the tier.
Behavior-based fee compounding
Fallback, retry, force post, and integrity fees are flat per event. The smaller portfolio has fewer good transactions to dilute the cost, so the blended impact runs higher.
Per-transaction minimums and flat event fees
Many charges are fixed per transaction or event, so low-ticket portfolios bear a higher basis-point cost. A $0.025 fee on a $50 transaction is 5 basis points; on a $500 transaction it's 0.5.
Beyond the fees, larger institutions negotiate accommodations on edge cases, waivers, and prioritized bulletin responses. Smaller institutions pay the published schedule as written.
The combined effect is meaningful. On smaller books, the per-dollar network cost can run materially higher than at scale, just from the disadvantages above. This compounds the broader trend of network fees outpacing payments volume. Before anybody mismanages anything.
What smaller players can actually do
Measure first
You can't negotiate or prioritize what you can't see. A quarterly decomposition of network spend by driver shows where the recoverable margin lives.
Prioritize the recoverable variance
Registration fees and structural minimums may look non-negotiable, but some can be restructured, reduced, or reallocated over time. Behavior-based leakage and optional program creep are usually the fastest place to start.
Build the operating cadence at fractional headcount
One owner connecting bulletin changes to invoice impact covers most of the ground.
Target the disputes you actually win
Network rules let you dispute specific fee categories where you have a defensible case. Most smaller institutions skip it because they don't think the dollars justify it. They usually do.
Use the right tooling, not spreadsheets and inboxes
Scheme-cost intelligence breaks down when bulletin changes live in email, invoice reviews live in Excel, and nothing connects the two. Smaller teams don't need a huge platform stack, just a repeatable way to track fee changes, map them to invoice impact, and separate structural from recoverable cost.
What's your biggest constraint: data access, internal ownership, or negotiating leverage?
Happy to walk through where smaller issuers and acquirers find the most margin in 2026.