The Optimization Playbook
Eight structured actions that compound: how to reduce your cost basis, qualify for better interchange programs, and price your sell rate to the market — without switching processors.
22 min read
What payment optimization actually means
Payment optimization is the systematic process of improving what you pay to process transactions and what you earn from them. It is not a one-time project. It is a discipline that compounds as your volume grows, your card mix data matures, and your negotiating leverage expands.
The discipline has three structural levers. Every improvement traces back to one of them.
Lever 1 — Reduce your cost basis (buy-rate optimization)
Target: the processor/acquirer markup layer of the MDR stack — the negotiable portion of your costs (see Chapter 5 for the full MDR waterfall). At $100M+ GMV, a 10 bps reduction in processor markup is $100,000 per year. The lever also includes structural improvements like switching from flat-rate to interchange-plus to eliminate hidden cost variance. Above $50M GMV, flat-rate pricing is a tax on success — and the migration math almost always works.
Lever 2 — Optimize interchange qualification
Interchange rates are not fixed per merchant. They are determined per transaction by whether you meet the network’s qualification criteria. Submitting Level 2/Level 3 data on commercial cards, maintaining correct MCC classification, hitting settlement timing windows, and using authentication tools (3D Secure) all directly affect the interchange rate the transaction qualifies for. Failure to meet criteria is a “downgrade” — the transaction falls to a worse, more expensive tier. Most platforms have never measured their downgrade rate.
Lever 3 — Raise your sell rate (revenue optimization)
The sell rate you charge merchants is your primary revenue lever. Most platforms under-price relative to the value they deliver, particularly as the platform becomes more embedded in merchant workflows. The five-minute Stripe Connect implementation that launched three years ago is now a system the merchant cannot leave without rebuilding their entire revenue operation. The price should reflect that. Regular sell-rate benchmarking and strategic repricing on new cohorts are underused tools.
The power of the three levers is that they compound. A platform that reduces buy rate by 15 bps, captures 25 bps of L2/L3 savings on a 20% commercial-card mix, and raises sell rate 15 bps has improved economics by 40+ effective basis points on the portfolio — without changing its payment stack. The Charge Forward Payments Revenue Calculator quantifies each lever against your specific card mix and ticket size.
Charge Forward Insight
The cleanest diagnostic question to ask any internal payments team is: “what is your downgrade rate, and what is it costing us?” Across Charge Forward advisory engagements, the modal answer is “we don’t track that.” That answer alone is worth 10–25 bps of recoverable economics in most portfolios. If you cannot answer the question, you have not yet started optimizing.
Level 2 and Level 3 data optimization
Level 2/Level 3 data is the highest-value underutilized lever in embedded payments. For platforms serving B2B merchants — construction, field services, legal, fleet management, professional services — L2/L3 optimization can be the single largest annual improvement available without any commercial renegotiation. And it is overwhelmingly the case that platforms running on standard PFaaS configurations are not capturing it.
What L2 and L3 data are
Card networks define three levels of data richness for commercial card transactions. Each level unlocks progressively lower interchange rates by demonstrating that the transaction carries lower fraud risk and better reconciliation data:
| Level | Required Fields | Typical Savings vs. Standard CNP Commercial | Who Benefits Most |
|---|---|---|---|
| Level 1 (Basic) | Card number, expiry, CVV, amount | Baseline — no savings | All merchants |
| Level 2 | Customer reference / PO number, tax amount, merchant postal code | ~20–30 bps vs. standard commercial CNP | B2B, government contractors, professional services |
| Level 3 | L2 fields + line-item detail: item description, quantity, unit price, commodity code, shipping | ~40–60 bps vs. standard commercial CNP | B2B with itemized invoices: construction, field services, fleet, legal, procurement |
Source: Visa U.S. Interchange Reimbursement Fee schedule, effective October 18, 2025; Mastercard U.S. Region Interchange Programs and Rates, effective April 11, 2025; Charge Forward Interchange Fees research (April 2026).
The savings are visible directly in the published interchange tables. Visa Commercial Card Not Present (standard): 2.70% + $0.10. Visa Commercial Level II qualified: 2.50% + $0.10 — a 20 bps saving. Level 3 qualification on the same transaction can save an additional 20–40 bps versus the Level II rate. These are not modeled estimates — they are line items on a published schedule.
IMPORTANT — Visa Level 3 update, April 2026: Starting April 1, 2026, Visa supports Level 3 data submission as the qualified pathway for commercial CNP commercial rates (Level 2-only is no longer a distinct qualified tier on Visa). Mastercard continues to support L2 and L3 independently. Platforms building L2/L3 capability today should design for Level 3 compliance from the outset to maximize qualification success on Visa transactions.
Which verticals benefit most
The L2/L3 opportunity is proportional to your commercial-card mix — the share of card volume where cardholders are using corporate, purchasing, or fleet cards rather than consumer credit. Verticals with structural B2B orientation benefit disproportionately:
• Construction (subcontractor and supplier payments): High-value B2B invoices, typically CNP. Commercial card mix can exceed 50% of card volume.
• Field services (HVAC, plumbing, electrical): Service invoices to commercial property owners and managers. Mix of consumer and commercial, particularly for larger commercial service accounts.
• Fleet and transportation management: Corporate fleet cards are inherently commercial purchasing cards — near-100% commercial mix.
• Legal services: B2B billing through platforms like Clio and LawPay; corporate clients frequently use purchasing cards.
• Professional services (accounting, consulting): CNP billing platforms with meaningful commercial card exposure.
The B2C verticals — consumer restaurants, fitness memberships, healthcare copays — have minimal L2/L3 opportunity because cardholders are paying with personal consumer credit or debit, not corporate purchasing cards.
Processor support for L2/L3
Not all processors make L2/L3 submission straightforward. Current documented support — verify the specifics with your provider:
• Adyen: Documented L2/L3 data submission and validation. Enterprise-grade implementation with April 2026 Visa Level 3 compliance built in.
• Finix: Native L2/L3 support built into the payment API; structured data fields for commercial card optimization.
• Stripe: Documents passing payment line items to participate in Level 2/Level 3 programs for eligible commercial cards.
• Braintree / PayPal: Documents Level 2/3 processing including the Level 3 line-item payload structures.
• PFaaS providers (Worldpay Payrix, Finix managed, Rainforest, NMI, others): Variable. Confirm with your provider whether L2/L3 data flows through to the network or is discarded in processing. This is a contract question, not a technical one — the answer is in your processor’s certification documentation.
The Charge Forward Vendor Database tracks L2/L3 support across the major PFaaS providers and is the fastest way to confirm a specific vendor’s current capabilities.
L2/L3 impact at scale
The table below shows the annual dollar value of L2/L3 optimization across GMV tiers and commercial-card mix levels. Savings are illustrative based on 25 bps (L2) and 50 bps (L3) improvement on the qualifying commercial card volume.
| Annual GMV | Commercial Card Mix | L2 Savings (25 bps) | L3 Savings (50 bps) |
|---|---|---|---|
| $50M | 10% (~$5M) | $12,500 | $25,000 |
| $50M | 20% (~$10M) | $25,000 | $50,000 |
| $50M | 30% (~$15M) | $37,500 | $75,000 |
| $100M | 10% (~$10M) | $25,000 | $50,000 |
| $100M | 20% (~$20M) | $50,000 | $100,000 |
| $100M | 30% (~$30M) | $75,000 | $150,000 |
| $250M | 10% (~$25M) | $62,500 | $125,000 |
| $250M | 20% (~$50M) | $125,000 | $250,000 |
| $250M | 30% (~$75M) | $187,500 | $375,000 |
| $500M | 10% (~$50M) | $125,000 | $250,000 |
| $500M | 20% (~$100M) | $250,000 | $500,000 |
| $500M | 30% (~$150M) | $375,000 | $750,000 |
How Visa and Mastercard actually shape your economics
Visa and Mastercard are rule-setters, not banks. They do not earn interchange — that goes to the issuing bank. They earn revenue through assessment and network fees charged on transaction volume. Understanding exactly what they do (and what they do not do) is essential for navigating the economics of your program.
Four things the networks actually do
1. Set interchange rates. The networks publish interchange rate tables by card type, channel, MCC, and data quality. These tables update semi-annually (Mastercard) and periodically (Visa). They are the single most important input to your cost model.
2. Charge network/assessment fees. Visa and Mastercard earn revenue from assessments — not interchange. Common fees include the Visa US Assessment (~0.14% of credit volume, ~0.13% of debit volume), Mastercard NABU ($0.0195 per authorization), Visa APF ($0.0195 per credit auth, $0.0155 per debit/prepaid auth), and the Visa FANF (Fixed Acquirer Network Fee, charged at the merchant tax ID level with complex tiering by channel, volume, and MCC).
3. Operate dispute and chargeback processes. Both networks run dispute resolution programs with escalating consequences. Visa has consolidated its programs into the Visa Acquirer Monitoring Program (VAMP), with “Above Standard” dispute ratio thresholds around 0.5%–0.7%. Mastercard’s Excessive Chargeback Program flags merchants at 100+ chargebacks and 1.5%–2.99% ratio. Monthly fines of $50–$100 per chargeback once in monitoring, with risk of card acceptance termination at extreme levels.
4. Define qualification criteria for interchange programs. The networks specify exactly what data, timing, authentication, and processing requirements must be met for a transaction to qualify for a specific interchange rate. Failure to meet criteria triggers a downgrade. Managing downgrades is core operational discipline.
Network assessment fees reference
| Fee Name | Network | Rate | Trigger |
|---|---|---|---|
| Assessment (brand usage) — Credit | Visa US | ~0.14% of volume | Per Visa credit transaction |
| Assessment (brand usage) — Debit | Visa US | ~0.13% of volume | Per Visa debit transaction |
| APF — Credit Auth | Visa US | $0.0195 per auth | Per Visa credit authorization |
| APF — Debit/Prepaid Auth | Visa US | $0.0155 per auth | Per Visa debit/prepaid authorization |
| FANF (Fixed Acquirer Network Fee) | Visa US | Variable by merchant; tiered by channel & MCC | Per merchant tax ID |
| NABU (Network Access & Brand Usage) | Mastercard US | $0.0195 per auth | Per Mastercard credit/sig-debit auth |
| Cross-Border Assessment | Visa / MC US | 0.60% (USD settled); 1.00% (non-USD) | Non-US-issued cards used at US merchants |
| Acquirer Service Fee | Visa Canada | 0.09% of volume | All Visa cards acquired in Canada |
| Acquirer Volume Assessment | Mastercard Canada | 0.090% (9 bps) | Assessable Mastercard volume in Canada |
Network rules that move your economics
For embedded payments platforms, the most impactful network rules govern:
• MCC assignment. Your MCC determines which interchange programs your transactions can qualify for. An incorrect MCC can permanently exclude you from lower-rate programs. Review your MCC against the network category definitions before launch and after any material expansion into new merchant segments.
• Cross-border fees. Transactions where the cardholder’s issuing country differs from the merchant’s country attract cross-border assessments (0.60%–1.00% depending on currency settlement). For US platforms serving Canadian merchants — or vice versa — this is a significant cost driver.
• Settlement timing. Transactions settled outside the network’s required timeframe (typically within 24 hours of authorization for most programs) may downgrade to more expensive interchange categories.
The structural factors that determine your blended cost
Beyond interchange optimization, several structural and operational factors shape your blended payment economics. Two platforms at identical GMV can have very different cost structures because of how their merchants accept payments.
Card mix — the most underestimated variable
Your card mix — the distribution of transaction volume across card types — is the single largest determinant of your blended interchange cost.
| Card Type | Typical Interchange Range (CNP) | Impact on Blended Cost |
|---|---|---|
| Regulated Debit (Durbin) | 0.05% + $0.21 | Dramatically lowers blended cost; fixed fee dominates on small tickets |
| Unregulated Consumer Debit | 1.05%–1.65% + $0.15 | Middle tier — volume-sensitive |
| Consumer Credit (standard) | 1.65%–2.50% + $0.10 | Core cost driver for B2C platforms |
| Consumer Prepaid (unregulated) | 1.75%–1.90% + $0.20–$0.25 | Often neglected; similar to consumer credit |
| Commercial / Purchasing | 2.50%–2.70% + $0.10 (without L2/L3) | Highest cost category without optimization |
| Commercial (with L2/L3) | ~2.10%–2.40% + $0.10 | Significant savings vs. standard commercial |
CP vs. CNP mix
Card-Present (CP) transactions carry materially lower interchange than Card-Not-Present (CNP) — typically 40–100 basis points lower on consumer credit. Your CP/CNP ratio is largely set by your vertical’s operational model:
• Restaurants, auto repair, in-clinic healthcare: 60–75% CP — structurally advantaged on interchange.
• Property management, legal, fitness memberships, construction: 85–95% CNP — structurally disadvantaged, which creates strong incentive to route large transactions through ACH/EFT.
• Field services (HVAC, plumbing): ~50/50 — a mix opportunity to optimize by routing large invoices to ACH.
Where CNP is unavoidable, 3D Secure authentication can qualify CNP transactions for lower interchange by reducing fraud risk signals. Implementation requires integration with an authentication service and increases transaction latency slightly. The tradeoff is almost always worth it on portfolios with material CNP volume.
Transaction size and fixed per-item fees
Fixed per-item fees ($0.10–$0.30 per transaction) affect small-ticket transactions disproportionately. A $0.30 per-transaction fee is:
• 3.0% of a $10 QSR tab
• 0.06% of a $500 HVAC service call
• 0.002% of a $15,000 construction invoice
For low-ATV verticals (restaurants, fitness), minimizing the fixed per-transaction component of your buy rate is worth more than equivalent percentage-rate improvements. For high-ATV verticals, the percentage component dominates.
Chargeback rates by vertical
Chargeback rates are both a direct COGS component (typically $15–$25 per dispute in direct fees) and a risk-program exposure (exceeding network thresholds triggers fines and processing restrictions). The variance across verticals is an order of magnitude.
| Vertical | Avg. Chargeback Rate | Risk Level | Key Drivers |
|---|---|---|---|
| Restaurants (QSR/In-person) | 0.01%–0.12% | Very Low | Low ticket, CP, immediate consumption |
| Full-Service Restaurants | ~0.15%–0.25% | Low | Delivery dispute friction growing |
| Property Management | ~0.05%–0.15% | Very Low | ACH-dominant reduces card disputes |
| Auto Repair | ~0.12%–0.25% | Low-Moderate | Service quality disputes; fleet card “not authorized” |
| Healthcare / Dental | ~0.15%–0.30% | Low-Moderate | Billing confusion, insurance mismatch |
| Field Services (HVAC/Plumbing) | ~0.10%–0.20% | Low-Moderate | Service disputes; unauthorized charges on filed cards |
| Fitness / Health & Wellness | 0.50%–0.86% | Moderate-High | Recurring billing, cancellation friction, friendly fraud |
| Legal Services | ~0.15%–0.35% | Low-Moderate | Post-engagement disputes, fee disagreements |
| Software / SaaS | 0.50%–0.66% | Moderate | Recurring billing, forgotten subscriptions |
| Education & Training | 0.80%–1.02% | High | Student refund disputes, event cancellations |
Operational implication: platforms serving fitness/wellness or education/training face fundamentally different risk economics than restaurant or field service peers. Chargeback management infrastructure — dispute response automation, cancellation flow optimization, pre-dispute tools (Ethoca Alerts, Verifi Rapid Dispute Resolution) — is not optional at meaningful scale in these verticals. Pricing for risk has to follow.
ACH vs. cards
ACH (Automated Clearing House) transfers process at a fraction of card interchange — typically $0.25–$0.75 flat per transaction, regardless of amount. For high-ATV verticals, the economics are not close:
• A $1,302 average rent payment: CNP card cost ~$26 (2.0% interchange); ACH cost ~$0.50 — a $25.50 per-transaction difference.
• A $50,000 construction invoice: CNP card cost ~$1,000–$1,350; ACH cost ~$0.50–$1.00 — ACH is essentially the only economically rational instrument.
This is why property management and construction platforms route the majority of high-value transactions through ACH, with cards reserved for cases where speed or convenience justifies the premium.
Surcharging — passing fees to customers
Surcharging — adding a line item to the customer’s bill to recover card processing costs — is legal in most US states (a few exceptions remain) and has been legal across all Canadian provinces since 2022. The rules:
• Visa and Mastercard both permit surcharging but require pre-notification, cap surcharges at the actual processing cost (not to exceed 3% for Visa, 4% for Mastercard), and require clear disclosure at checkout.
• Not permitted on debit cards or prepaid cards — credit cards only.
• Where the platform passes the surcharge through to the merchant’s end customer, the merchant’s net payment cost approaches zero — card acceptance becomes effectively neutral from a cost perspective.
• Risk: surcharging can reduce card conversion rates and add friction in consumer-facing transactions. B2B billing and professional services platforms typically see minimal conversion impact; consumer checkouts can see meaningful conversion erosion.
Charge Forward Insight
Surcharging is the most common “quick win” advisory clients ask about — and the most common one that misfires. In B2B verticals, it can meaningfully restructure your payment economics. In B2C, the conversion hit often exceeds the savings. Run an A/B test on a meaningful cohort before rolling it out portfolio-wide. The Charge Forward Embedded Payments Maturity Framework flags surcharging as a Stage 3 (Margin Expansion) move — premature deployment at Stage 1 or 2 is a frequent source of merchant churn.
The optimization impact model
The table below shows the annual dollar impact of three optimization actions at each GMV tier: a 10 bps buy-rate improvement, a 25 bps composite improvement (a mix of interchange qualification and processor renegotiation), and a 15 bps sell-rate increase.
| Annual GMV | 10 bps Improvement | 25 bps Improvement | Sell Rate +15 bps | Combined (50 bps) |
|---|---|---|---|---|
| $50M | $50,000 | $125,000 | $75,000 | $250,000 |
| $100M | $100,000 | $250,000 | $150,000 | $500,000 |
| $250M | $250,000 | $625,000 | $375,000 | $1,250,000 |
| $500M | $500,000 | $1,250,000 | $750,000 | $2,500,000 |
| $1B | $1,000,000 | $2,500,000 | $1,500,000 | $5,000,000 |
How to read this: at $250M GMV, a comprehensive optimization review that delivers 50 total effective basis points of improvement is $1.25M per year. Over three years, that is $3.75M — sufficient to fund a complete payments technology rebuild or a dedicated payments product team. The case for investing seriously in optimization is straightforward at any meaningful GMV tier. The question is not whether the economics justify it — it is whether your organization has the prioritization and process discipline to execute.
How to run a payment optimization review
The eight-step process below is designed for platforms between $25M and $1B+ GMV. It can be completed in four to eight weeks with internal resources, or accelerated with external advisory support. Each step produces a specific output that feeds the next.
Steps 1–5 require access to your processor’s interchange detail reports (typically in your reconciliation or reporting portal). Steps 6–8 are commercial and can be led by your CFO or VP Finance with support from your payments or engineering lead. Total time investment: 40–80 hours for an internal team running the review for the first time.
| Step | Action | What You Are Looking For | Typical Finding |
|---|---|---|---|
| 1 | Audit your buy rate — get interchange detail, not just blended rate | Transaction-level interchange qualification by program and card type | Many platforms discover 10–20% of volume is downgrading to expensive non-qualified tiers |
| 2 | Analyze card mix (credit / debit / commercial / prepaid split) | Exact percentage of commercial card volume — this drives the L2/L3 opportunity | B2B-facing platforms often underestimate commercial card share by 5–15 points |
| 3 | Identify interchange downgrades | Settlement timing failures, missing CVV/AVS data, absent L2/L3 data on commercial cards | Recoverable 10–40 bps depending on volume and vertical |
| 4 | Assess L2/L3 opportunity | Commercial card transactions without L2/L3 data attached | 30–80 bps savings available on qualifying commercial volume |
| 5 | Benchmark sell rate against market | Effective MDR vs. published benchmarks for your vertical and size | Most platforms are 15–50 bps below market sell-rate ceiling |
| 6 | Review processor contract terms | Fee schedule, volume tiers, term commitments, downgrade rules | Stale contracts often contain above-market markups never renegotiated |
| 7 | Model the impact of switching from flat-rate or tiered to IC+ | Project economics under IC+ using your actual card mix data | IC+ switch typically saves 20–60 bps at meaningful commercial card volumes |
| 8 | Negotiate or issue RFP if savings are material | Define “material” as >$100K annual run-rate savings, or >25 bps improvement available | Formal RFP processes regularly achieve 15–30 bps improvement at $50M+ GMV |
Charge Forward Insight
Across advisory engagements, three patterns recur consistently with platforms that have never run a formal optimization review. One: the flat-rate trap — platforms processing $25M–$100M on Stripe or Square standard pricing are typically paying 60–100 bps more than they need to, and the IC+ migration is the single highest-ROI move at this scale. Two: the commercial card blind spot — platforms with B2B merchants are often processing commercial cards at standard CNP rates (2.50%–2.70%) because nobody asked whether L2/L3 data was being submitted; in almost every case, it is not, and the technical fix is weeks, not months. Three: the stale contract — processors rarely proactively renegotiate as a platform’s volume grows. A contract signed at $10M GMV is often still in effect at $150M GMV, with markups 3–5x higher than the platform could achieve with a competitive RFP today. These patterns are not edge cases. They are the modal state of embedded payments platforms at growth stage. Optimization is not about sophistication — it is about attention.
Where optimization fits on the maturity curve
Optimization is not a single playbook applied uniformly. The right next move depends on where your platform sits across the maturity spectrum. The Charge Forward Embedded Payments Maturity Framework maps platforms across five stages — Capability, PFaaS Transition, Margin Expansion, Orchestration, and Fintech — each anchored to a GMV band and a set of stage-specific moves.
Lever 1 (buy-rate reduction) and Lever 3 (sell-rate optimization) apply at every stage, but the playbook differs. At Stage 1 (Capability, <$10M GMV), the optimization conversation is mostly about getting onto a sensible vendor stack and avoiding flat-rate cost creep. At Stage 2 (PFaaS Transition, $10M–$50M GMV), the L2/L3 implementation and the IC+ migration are the headline moves. At Stage 3 (Margin Expansion, $50M–$250M GMV), surcharging, ACH routing, commercial-card optimization, and sell-rate repricing dominate. At Stage 4 (Orchestration, $250M–$1B GMV), multi-processor routing, network tokenization, and downgrade-rate management become viable. At Stage 5 (Fintech, $1B+ GMV), the platform is running its own optimization stack — the question shifts to whether to internalize processing entirely.
Running the eight-step review without first identifying your stage produces a defensible but generic action list. Pairing the review with the Maturity Framework produces a prioritized, stage-specific playbook — which is the difference between an optimization theory deck and an optimization plan a CFO can execute.
What’s next
Chapters 5 and 6 together have built the complete foundation for understanding and capturing your payment economics. You now have the interchange tables, pricing-model frameworks, financial modeling structure, and optimization process to run a professional-grade payments program at any stage of scale.
Chapter 7 — “From Integration to Adoption” — turns to the harder operational problem that determines whether your optimized economics actually generate revenue: attach rate. The most optimized payment stack in the world is worth nothing on the 40% of your merchants who are still processing somewhere else. Chapter 7 covers the five levers (product integration, onboarding, sales incentives, UX, pricing) that move attach rate from median to leader.
For the full benchmark dataset referenced throughout this chapter, see Chapter 11. To quantify the optimization opportunity for your specific platform, run the Charge Forward Payments Revenue Calculator. To identify which optimization moves apply at your stage, download the Embedded Payments Maturity Framework.
SOURCES & REFERENCES
Visa US Interchange Reimbursement Fee schedule, effective October 18, 2025; Mastercard US Region Interchange Programs and Rates, effective April 11, 2025; Visa and Mastercard US/Canada assessment schedules; Visa Acquirer Monitoring Program (VAMP) and Mastercard Excessive Chargeback Program documentation.
Worldpay/Payrix Vertical SaaS Benchmarking Study (January 2025); Ethoca 2025 State of Chargebacks Report; Clearly Payments 2024 Chargeback Benchmarks.
Charge Forward Embedded Payments Benchmark Report (April 2026); Charge Forward Interchange Fees research (April 2026); see Chapter 11 for the full benchmark dataset and source table.
Public Charge Forward tools referenced in this chapter: Payments Revenue Calculator, Vendor Database, Embedded Payments Maturity Framework, Embedded Payments Fit Assessment, Payment Model Fit Navigator. All available at chargeforward.io/tools.
By Jane Podbelskaya · Updated