Credit Card Interchange Revenue per Active Account: Banking Fee Income in 2027

A regional issuer watches its fee line shrink while accounts grow

Picture a $14 billion-asset regional bank with 620,000 open credit card accounts. Over eighteen months the card team added 90,000 net new accounts through a branch-referral push and a digital pre-approval funnel. Total interchange revenue rose about 6 percent, and the quarterly deck showed a green arrow. The CFO signed off. Nine months later the card P&L was under water on a contribution basis and nobody could explain why, because the headline number had never gone red.

The problem is that total interchange is a product of two things — how many accounts transact and how much interchange each transacting account throws off — and account growth can mask a collapse in the second term for a long time. When the team finally divided the fee line by average monthly active accounts, the picture inverted. Gross interchange per active account had fallen from roughly $268 to $221, an 18 percent decline, while the account base grew 17 percent. The two effects nearly cancelled in the total. Revenue per account is the metric that does not let you hide that.

Three things drove the decline, and each is common enough to be worth naming. First, the new accounts came in through a channel that skewed near-prime, with average credit lines around $2,400 versus $9,800 for the legacy book. Lower lines mean lower spend headroom, and lower spend means less interchange regardless of rate. Second, the promotional offer used to acquire them was a 0 percent intro APR on purchases for fifteen months, which drives balance-building behavior rather than transaction frequency; cardholders parking a single large balance generate one interchange event, not forty. Third, the acquisition creative featured groceries and gas as hero categories, and cardholders obliged — supermarket and fuel volume climbed from about 19 percent of the book to nearly 31 percent, and those merchant category codes carry materially lower interchange rates than dining, travel, or card-not-present retail.

None of those three failures were visible in total revenue. All three were visible in interchange revenue per active account within one reporting cycle. That is the entire case for the metric: it isolates the unit economics of a card program from the growth story sitting on top of it, and in banking fee income the two move independently far more often than executives expect. Heading into 2027, with organic card growth slowing across most U.S. issuers and acquisition costs elevated, the per-account view stops being a nice-to-have diagnostic and starts being the number the card P&L actually turns on.

The corrective the regional bank ran took two quarters. They segmented the book by acquisition cohort and credit tier, found that the near-prime digital cohort was producing roughly $140 gross per active account against a fully-loaded acquisition cost north of $180, and paused that channel. They re-weighted the rewards structure toward dining and travel, which pulled the merchant mix back roughly four points. And they instrumented an activation campaign against the 31 percent of accounts that had not transacted in ninety days. Per-account interchange recovered to about $252 within three quarters — still below the legacy peak, but the trajectory had reversed and, more importantly, leadership could see it happening in something other than a lagging total.

How interchange actually reaches the issuer's fee line

Interchange is not a fee the bank charges the cardholder. It is a fee the merchant's acquiring bank pays the card-issuing bank on every purchase, set by the network's published rate schedule, and it moves through settlement rather than through any customer-facing statement. Understanding the plumbing matters because every lever available to a card team acts on one specific link in that chain.

A purchase runs like this. The cardholder taps or enters a card at a merchant. The merchant's acquirer routes an authorization request through Visa, Mastercard, or another network to the issuing bank, which checks the line, applies fraud scoring, and approves or declines. At settlement — usually the next business day — the acquirer pays the merchant the transaction amount minus the merchant discount rate. Out of that discount rate, interchange flows to the issuer, network assessments flow to the network, and the remainder is the acquirer's and processor's margin. Interchange is typically the largest of the three components, which is why the issuer's fee line is so sensitive to the rate that applies.

The rate that applies is not one number. Networks publish extensive interchange schedules with rates that vary by card product, merchant category, transaction environment, and data quality. A rewards consumer credit card at a restaurant carries a materially different rate than a standard consumer card at a supermarket, which in turn differs from a commercial card at a card-not-present retailer. Most rates take the form of a percentage plus a fixed per-item amount — the fixed component is why average transaction value matters so much, and the percentage component is why merchant mix matters so much. Regulated debit interchange, capped under the Federal Reserve's Regulation II for issuers above the asset threshold, sits in an entirely separate regime and should never be blended into a credit interchange metric. Doing so is one of the most common ways a banking KPI ends up meaningless.

Two definitional choices decide whether the resulting number is usable. The first is the denominator. An active account should mean an account with at least one posted purchase transaction in the measurement month, averaged across the months in the period — not total open accounts, and not accounts with a balance. Using open accounts inflates the base and understates unit economics; using accounts-with-a-balance mixes lending behavior into a transaction metric. If a portfolio has 620,000 open accounts and 415,000 that transacted in the month, the activation rate is about 67 percent, and both the numerator-per-active and the activation rate itself deserve to be tracked, because they respond to completely different interventions.

The second choice is which numerator. Gross interchange is the raw fee inflow and is the right number for diagnosing merchant mix and spend behavior, because it moves only when transaction volume, transaction size, or rate mix moves. Net interchange subtracts the accrued rewards liability, network assessments, and typically fraud losses, and is the right number for pricing decisions, because a rewards program can be generous enough to consume the entire gross fee. A card earning a blended 1.75 percent in interchange while paying 2 percent flat cash back is running a structurally negative contribution before a single operating cost is allocated. Track both; report both; never let a deck show one and label it the other.

One more mechanical point that trips teams up: rewards liability is an accrual, not a cash outflow, and it accrues when points are earned rather than when they are redeemed. A portfolio that just launched a rich earn structure will book the full liability immediately while redemption behavior — and any breakage assumption — plays out over years. If the accrual assumption is wrong, net interchange per active account is wrong in the same direction, and the error compounds quietly. Have the finance team restate the last eight quarters whenever the breakage assumption changes, or the trend line becomes uninterpretable.

The numbers: what good and bad look like in a real portfolio

Concrete ranges make the metric operational. The figures below reflect the general shape of U.S. consumer and commercial card economics; every issuer should recompute against its own settlement data rather than importing benchmarks wholesale, because portfolio composition swamps industry averages.

Gross interchange per active account, annual. A standard consumer rewards portfolio typically lands somewhere in the $220–$380 range. Below roughly $180 signals either very low spend per account or a merchant mix badly weighted toward low-rate categories. Above $450 generally indicates either a premium travel product with genuinely high-spend cardholders or a commercial book. Premium travel cards with affluent, high-frequency spenders can clear $500–$600. Retail store cards — narrow acceptance, frequent promotional financing, lower activation — commonly sit near $180 or below.

Net interchange per active account, annual. After rewards accrual, network assessments, and fraud, a healthy consumer rewards portfolio generally holds $120–$200. Under $100 is a warning that the earn rate is too rich relative to the blended interchange the merchant mix actually produces. Above $200 net on a rewards product usually means either a well-capped bonus structure or a mix genuinely skewed toward high-rate categories.

Average transaction value. Consumer credit typically runs in the $45–$65 band; commercial card ATV runs materially higher, often $80–$120 or above depending on whether the program covers T&E, purchasing, or fleet. ATV matters disproportionately because of the fixed per-item component in most interchange rates. Take a rate of roughly 2 percent plus ten cents: a $50 ticket yields about $1.10, an effective 2.2 percent; a $100 ticket yields about $2.10, an effective 2.1 percent. The fixed component is worth 20 basis points on a $50 ticket and only 10 on a $100 one, which means small-ticket-heavy portfolios earn a higher effective rate per dollar but need far more transactions to reach the same per-account revenue.

Transactions per active account per month. This is the most under-tracked input and often the most actionable. A primary-wallet card typically sees 18–30 posted purchases a month; a secondary card sees 4–8. The gap between those two states is worth more per account than almost any rate optimization, because it multiplies rather than adds. Moving a cohort from secondary to primary status roughly triples its interchange contribution without touching the rewards structure at all.

Activation rate. Consumer credit portfolios commonly run 65–75 percent monthly active against open accounts. Store cards often sit in the 50–60 percent range. The arithmetic here is worth internalizing: a book of one million open accounts at 65 percent activation has 650,000 revenue-generating accounts, while a competitor's 800,000-account book at 80 percent has 640,000 — nearly identical productive scale from a base 20 percent smaller. Activation is a real lever on total fee income and costs far less to move than acquisition.

Merchant category mix. Rough working targets for a healthy consumer rewards book: supermarkets and fuel combined under 25 percent of purchase volume, dining and travel above 15 percent, and card-not-present retail above 25 percent. When low-rate categories climb past 35 percent of volume, gross interchange per active account will decline even with flat or growing total spend, and it will keep declining until the mix reverses. Government payments, utilities, and certain education categories carry among the lowest rates in the schedule; a book heavily exposed to them should expect structurally suppressed yield and plan accordingly rather than chasing it.

Seasonality. Fourth-quarter purchase volume commonly runs 20–30 percent above the trailing average for consumer portfolios. Any target set off a Q4 run rate will miss badly in the first quarter. Use trailing-twelve-month figures for benchmarking and year-over-year same-quarter comparisons for trend, never sequential quarters against an annualized Q4.

Putting the arithmetic together for a single account: 22 transactions a month at a $55 average ticket is about $1,210 in monthly purchase volume, or roughly $14,500 a year. At a blended effective rate of 1.75 percent that is approximately $254 in gross interchange. Subtract a 1.2 percent effective rewards cost — allowing for tiered earn and breakage — and roughly 14 basis points of assessments and fraud, and net contribution lands near $50–$55 per active account per year. That is the honest unit economics of a mid-tier consumer rewards card, and it explains why annual fees, interest income, and interchange have to be evaluated together rather than in isolation.

Trade-offs: every lever on this metric costs something elsewhere

There is no move that raises interchange revenue per active account for free. Each lever trades against acquisition volume, credit risk, rewards expense, or customer satisfaction, and card teams that ignore the second-order effect tend to win the metric and lose the P&L.

Richer rewards to drive spend. Raising the earn rate reliably increases transaction frequency and share of wallet, which raises gross interchange per active account. It also raises rewards expense on every dollar, including dollars that would have been spent anyway. The break-even is straightforward but frequently skipped: incremental gross interchange must exceed incremental rewards cost across the whole spend base, not just the incremental spend. A blanket 3 percent earn against a blended 1.75 percent interchange is negative on contact. Category-capped structures — an elevated rate on a specific category up to a quarterly ceiling — are the standard containment mechanism precisely because they bound the liability while still shifting behavior.

Shifting merchant mix through targeted offers. Bonus earn on dining or travel pulls volume toward higher-rate categories and can move blended yield by a real margin. The trade-off is that the promotional cost is immediate and certain while the mix shift is gradual and partly temporary; measure sustained mix ninety days after the promotion ends, not during it, or you will capitalize a rented behavior change.

Chasing higher credit tiers. Prime and super-prime cardholders spend more, carry higher lines, and skew toward higher-rate merchant categories. They are also expensive to acquire, heavily courted by competitors, and more likely to be transactors who generate little interest income. A book that optimizes purely for interchange per active account will drift upmarket and may sacrifice net interest margin in the process. This is a portfolio strategy decision, not a KPI decision, and it should be made explicitly.

Activation campaigns on dormant accounts. Cheapest lever available in most portfolios. A dormant account costs money to maintain and produces zero interchange; converting even a fraction of the dormant base is high-return. The caution is definitional — a campaign that produces one qualifying transaction and then silence has technically raised the activation rate while lowering interchange per active account, because it added a near-zero-revenue account to the denominator. Judge activation work on sustained ninety-day transaction counts, not on campaign-window conversions.

Annual fees. An annual fee funds a richer earn structure and improves total revenue per account, but it suppresses acquisition volume and raises attrition among light spenders. It also does not appear in interchange at all, so a fee-funded premium card will show strong gross interchange, weak net interchange, and healthy total contribution. Report the fee line alongside the interchange metric or the card will look worse than it is.

The discipline that makes this tractable is refusing to evaluate any lever on gross interchange alone. Every proposal should carry a projected effect on gross interchange per active account, net interchange per active account, activation rate, and acquisition cost per account, with the last two included specifically because they are where the hidden cost usually lands.

Pitfalls that quietly corrupt the number

Blending debit into credit. Regulated debit interchange operates under a statutory cap for issuers above the Federal Reserve's asset threshold and behaves nothing like credit interchange. A blended figure will drift with debit-versus-credit mix rather than with anything a card team controls. Report them as separate metrics with separate benchmarks, always.

Using open accounts as the denominator. The most common error and the most distorting. It understates per-account economics by whatever the dormancy rate is — 30 percent dormancy makes the number 30 percent too low — and it moves when dormancy moves rather than when spend behavior moves, which is the opposite of what you want from a diagnostic.

Reporting gross while making net decisions. A product manager looking at gross interchange will approve a richer earn structure that gross interchange rewards and net interchange punishes. If only one number can go in the executive deck, make it net, and put gross in the appendix for diagnosis.

Annualizing off a fourth quarter. Holiday volume inflates the run rate by 20–30 percent. Annualizing it produces a target the first quarter cannot reach, followed by a mid-year reforecast that reads as a miss but was a measurement error.

Ignoring cohort composition. A single portfolio-level number hides everything. Compute the metric by acquisition cohort, credit tier, product, and channel. In the regional-bank example above, the portfolio number moved 18 percent while one cohort had moved more than 40 percent — the aggregate was averaging a healthy legacy book against a badly underperforming new one, and only the cut revealed it.

Treating rewards accrual as fixed. Breakage assumptions change, redemption behavior shifts when a program adds transfer partners or richer redemption options, and each change restates the effective rewards cost. When it changes, restate history before comparing trend. An unrestated series will show a step change that looks like a business event and is actually an accounting one.

Missing rate-schedule changes. Networks update interchange schedules periodically, and category definitions shift. A blended-rate decline that is really a schedule change gets misdiagnosed as a behavior problem, and the team spends a quarter running promotions against something no promotion can fix. Reconcile blended effective rate against the published schedule at least semiannually.

No anomaly alerting. A single large merchant reclassifying its category code, a processor data feed dropping a segment, or a promotional program ending can move the number several percent overnight. Set an alert on a week-over-week change beyond about 5 percent against the trailing four-week average, and require someone to close each alert with a written cause. Most will be data issues; the ones that are not are the ones worth knowing about within a week rather than at quarter close.

Attributing the whole card P&L to this metric. Interchange is one of three or four revenue streams on a card program alongside net interest income, annual fees, and penalty fees. It is the cleanest signal of transaction health, which is why it deserves its own tracked line, but it should never be optimized in a way that quietly destroys the others.

Related questions

How often should this metric be recalculated?

Gross interchange per active account weekly, to catch data anomalies and promotional effects while they are still actionable. Net interchange monthly, after rewards accrual and fraud losses close. Cohort, tier, and merchant-mix cuts quarterly, since those move too slowly to warrant more frequent review.

Does the Durbin Amendment cap credit card interchange?

No. Regulation II caps debit interchange for issuers above the statutory asset threshold. Credit card interchange is not subject to that cap, though network-published rate schedules effectively set what issuers earn. Keep debit and credit interchange as separate metrics with separate benchmarks.

Can net interchange per active account go negative?

Yes. If the rewards earn rate plus network assessments and fraud losses exceeds the blended interchange rate the merchant mix actually produces, net contribution is negative. Flat high-earn structures on low-rate merchant mixes are the usual cause; category caps are the usual fix.

Why does average transaction value matter so much?

Most interchange rates combine a percentage with a fixed per-item amount. The fixed component is a larger share of a small ticket, so small-ticket portfolios earn a higher effective rate per dollar but need far more transactions to reach the same per-account revenue.

Should commercial and consumer cards share a benchmark?

No. Commercial programs run higher average transaction values, different merchant mixes, and different rate schedules, so per-account interchange typically runs well above consumer levels. Benchmarking them together makes both numbers uninterpretable. Track separately and compare each against its own product-type peer set.

FAQ

What exactly counts as an active account in this metric?

An account with at least one posted purchase transaction during the measurement month, averaged across the months in the reporting period. Not total open accounts, and not accounts carrying a balance — a revolving account with no purchases generates interest income but no interchange, and including it distorts a transaction metric with lending behavior.

Which is the better headline number, gross or net interchange per active account?

Net, for any decision involving pricing or product structure, because it reflects rewards liability and network costs. Gross is the better diagnostic for spend behavior and merchant mix, since it moves only when volume, ticket size, or rate mix moves. Mature card teams report both and label them clearly.

How much does merchant category mix actually move the number?

Substantially. Interchange rates vary widely across merchant categories, and a shift of ten or more percentage points of volume from higher-rate categories like dining and travel into lower-rate ones like supermarkets and fuel can cut gross interchange per active account by a double-digit percentage with no change in total cardholder spend at all.

What is the fastest lever to raise this metric?

Activation, in most portfolios. Dormant accounts cost money to maintain and generate zero interchange, and reactivation costs far less than acquisition. The condition is that reactivation must be sustained — measure transaction counts ninety days after the campaign, because a single qualifying transaction adds an account to the denominator without adding meaningful revenue.

How should seasonality be handled in targets?

Use trailing-twelve-month figures for benchmarking and year-over-year same-quarter comparisons for trend. Fourth-quarter volume commonly runs 20–30 percent above the trailing average, so annualizing a Q4 run rate produces a first-quarter target the portfolio cannot reach and a reforecast that reads as a miss but was a measurement error.

Why not just track total interchange revenue?

Because account growth and per-account productivity move independently and can offset each other completely. A portfolio can grow accounts 17 percent while per-account interchange falls 18 percent and show a flat-to-positive total for many quarters. The per-account view separates the growth story from the unit economics underneath it.

Sources

• Federal Reserve — Regulation II (Debit Card Interchange Fees and Routing)
• Visa USA — Interchange Reimbursement Fees
• Mastercard — Interchange Rates and Criteria
• Consumer Financial Protection Bureau — Consumer Credit Card Market Report
• Federal Reserve — Payments Study
• FDIC — Quarterly Banking Profile
• Federal Reserve Bank of Kansas City — Payments Research
• Financial Accounting Standards Board — Revenue from Contracts with Customers (ASC 606)

Related on PULSE

• [Fee per Transaction in Wealth Management: Advisory Revenue Yield](/knowledge/ik0540)
• [Loan-to-Deposit Ratio as a Liquidity and Revenue Metric for Credit Unions](/knowledge/ik0508)
• [Transaction Revenue Per Active User in Digital Wallets like PayPal](/knowledge/ik0521)
• [Top 10 Banking Net Interest Margin Revenue KPIs](/knowledge/ik0593)
• [Membership Dues Revenue per Active Member: Association Renewal KPI](/knowledge/ik0569)