Blockchain networks charge fees that shift by the minute. For payment service providers, this creates a structural problem: how do you price crypto transactions for merchants when the cost of settling those transactions can double or disappear between the time an order is placed and the time it clears? The answer is not a single number. It is a system. PSPs that get this right build durable merchant relationships. Those that do not absorb unpredictable losses or pass on surprise fees that erode trust quickly.
TL;DR
- Network fees on blockchains fluctuate constantly based on congestion and demand for transaction capacity [ideas.repec.org].
- PSPs face a structural mismatch: merchants want predictable pricing, but settlement costs are dynamic.
- Fee management is not a technical detail. It is a trust and margin problem.
- The best PSPs solve this through fee buffering, real-time monitoring, and policy-driven automation rather than manual adjustments.
- Infrastructure choices, including the underlying crypto payment gateway and custody layer, directly determine how well a PSP can absorb and manage fee volatility.
About the Author: Cregis is an enterprise-grade crypto financial infrastructure provider with nine years of operation and zero security incidents. Serving 3,500+ businesses across 50+ countries and securing over $300 billion in yearly transactions, Cregis builds the payment and custody infrastructure that PSPs rely on to operate at institutional scale.
Why do blockchain network fees fluctuate so unpredictably?
Blockchain fees are not set by a central authority. They emerge from competition for limited transaction capacity. When network demand increases, fees rise. When demand falls, fees drop. The underlying mechanisms vary by blockchain, but the pattern is consistent across networks.
On Ethereum, the fee structure introduces a base component that adjusts automatically based on network congestion, plus an optional priority component [ideas.repec.org]. This was designed to improve fee predictability, but it did not eliminate volatility. It shifted the volatility from chaotic bidding to a more structured but still reactive price signal [ui.adsabs.harvard.edu]. On Bitcoin and other chains, simpler fee-market models remain, where congestion can cause fees to spike sharply within minutes.
For a PSP, this means:
- A transaction priced at a certain fee estimate at checkout may cost significantly more by the time it is broadcast.
- During network congestion events, the gap between estimated and actual settlement cost can become a margin problem at scale.
- Chains with slower block times compound the risk because pending transactions may wait while fees continue to shift [consensys.io].
The dynamic fee design problem is fundamentally about timing: the moment of merchant pricing and the moment of network settlement are rarely the same moment.
What is the merchant trust problem that fee volatility creates?
Stepping back from the technical detail, the harder business problem is trust. Merchants do not care about settlement mechanics or fee structures. They care about three things: what they are charged, when they are paid, and whether the numbers are consistent.
Fee volatility breaks all three in different ways [optimus.tech]:
- Inconsistent charges: If a PSP passes through raw network fees, merchants see different costs for similar transaction sizes at different times. This is confusing and erodes confidence in the PSP's pricing model.
- Settlement delays: When PSPs underprice fees to stay competitive, transactions may confirm slowly or not at all during peak congestion. Delays damage merchant cash flow expectations.
- Surprise deductions: If a PSP absorbs fees up front and then adjusts settlement amounts retroactively to recover costs, merchants experience unexpected shortfalls. Even if this is disclosed in contracts, it creates friction.
The merchants most likely to leave a PSP are not the ones who experience high fees. They are the ones who experience unpredictable fees. Predictability, even at a slightly higher price point, builds more durable relationships than volatile cost-passing [optimus.tech].
How do PSPs actually manage fee volatility in practice?
Most PSPs combine several approaches. None of them alone is sufficient.
Fee buffering A PSP sets merchant-facing fees slightly above the historical average network cost for a given chain and transaction type. The buffer absorbs short-term spikes without passing them to the merchant. During low-fee periods, the buffer generates a small surplus that offsets high-fee periods. This requires careful calibration because an overly large buffer makes the PSP uncompetitive, while an overly thin buffer creates margin risk during congestion events.
Real-time fee monitoring and dynamic routing Rather than broadcasting all transactions on the same chain, sophisticated PSPs monitor fee conditions across multiple networks and route settlement through the lowest-cost path that meets the merchant's speed requirements. A stablecoin payment, for example, might settle on a lower-fee chain without the merchant noticing any difference in outcome.
Policy-driven automation Manual fee management does not scale [moderntreasury.com]. PSPs operating at volume need rules that fire automatically: if network fees on Chain A exceed a defined threshold, delay non-urgent batches, or switch to Chain B. These policy systems replace reactive human decisions with pre-approved institutional logic that operates consistently at any hour.
Batch settlement Grouping multiple merchant payouts into a single on-chain transaction reduces per-merchant fee exposure. This is standard practice but requires the PSP to have sufficient liquidity to hold funds during batching windows without creating settlement delay complaints.
What infrastructure choices determine how well a PSP handles this?
A related but distinct question is why some PSPs handle fee volatility better than others at the same transaction volume. The answer almost always traces back to infrastructure.
The underlying platform determines:
- How quickly the PSP receives fee signal updates from the network.
- Whether multi-chain routing is available without rebuilding the integration.
- How granularly fee policies can be configured per merchant, per currency, or per transaction type.
- Whether compliance checks run in parallel or add latency to the fee-setting window.
PSPs that rely on basic tools often find that fee management logic has to be built and maintained by their own engineering teams [moderntreasury.com]. Every network upgrade, every new chain, every shift in fee market design requires custom work. This creates both operational cost and reliability risk.
Infrastructure providers built for institutional use embed this logic into the platform itself. Fee monitoring, multi-chain settlement, and policy automation are built into the foundation, not engineering projects that sit on top of it.
Cregis serves as the Trust Layer for institutions managing cross-border payment flows, operating at scale without absorbing unpredictable fee volatility. The platform delivers real-time settlement visibility across multiple chains and assets, with automated policy controls that respond to fee and risk conditions before they become merchant-facing problems. For PSPs managing institutional portfolios, this infrastructure foundation makes consistent, trustworthy merchant pricing operationally viable [elliptic.co].
Frequently Asked Questions
What causes sudden spikes in crypto network fees? Spikes are typically caused by high demand for transaction capacity, often from large trading volumes, market events, or other surge conditions. Fee structures adjust upward automatically when networks experience consistent high congestion [ideas.repec.org].
Can PSPs offer flat-fee pricing for crypto payments? Yes, but it requires fee buffering. The PSP absorbs the difference between the flat merchant rate and actual network costs, which means margin management is critical.
What is the role of stablecoins in reducing fee unpredictability? Stablecoins settled on lower-fee chains reduce the absolute fee exposure significantly. The fee is still variable, but the base level is much lower, making buffer management easier.
How does multi-chain routing help with fee management? By routing the same economic transaction through different networks based on real-time fee conditions, PSPs can systematically choose cheaper settlement paths without changing the merchant's experience.
What is a Policy Engine in the context of crypto payments? A Policy Engine is a rules-based system that converts defined conditions, such as fee thresholds or risk signals, into automated actions like delays, routing changes, or transaction approvals.
Does fee management affect AML compliance? Routing and batching decisions can affect which chain a transaction settles on, which in turn affects which compliance tools apply. PSPs need infrastructure where compliance monitoring travels with the transaction regardless of routing decisions.
How quickly can fee conditions change on major networks? Fee conditions shift materially within minutes during periods of high network activity [ui.adsabs.harvard.edu].
About Cregis
Cregis provides enterprise-grade crypto financial infrastructure for institutions, PSPs, banks, and financial service companies operating at scale. With nine years of operations and zero security incidents, Cregis secures over $300 billion in yearly transactions across 3,500+ businesses in 50+ countries. Its platform covers wallet infrastructure, stablecoin payment processing, and policy automation that manages compliance and operational controls, all built to the first tier of security standards in the industry, including SOC 2 Type II, ISO 27001, and PCI DSS certification. For PSPs managing fee volatility at institutional scale, Cregis provides the Trust Layer foundation that makes consistent, trustworthy merchant pricing operationally viable.
Ready to see how institutional-grade infrastructure handles fee volatility in practice? Visit Cregis to learn more.
