The true cost of Recharge on a $5M subscription brand

At $5M in subscription GMV (gross merchandise value, the total dollars flowing through your subscriptions), the Recharge line on your P&L reads as a tidy SaaS fee. What it does not read as is a 1.49% tax on every subscription dollar, stacked on top of the Shopify Payments spread on the same order, plus a per-order charge that compounds with every renewal. And none of that captures the involuntary churn quietly draining a fifth to a third of your subscriber losses through payment failures you never bothered to recover.

Modeled correctly, the total cost of Recharge on a $5M subscription brand runs about 1.56% of GMV at the app layer, 3.8% to 4.5% all-in once you add payments, and then a separate five-figure leak on top that never shows up on the pricing page. This post builds that model, stress-tests it against Ordergroove and a direct Stripe setup, and finds the GMV thresholds where the math actually changes.

What Recharge actually charges: the three-layer fee model

Recharge has three fee layers, and the trap is that operators only budget the first one. Layer one is the SaaS subscription: $99/mo on Starter, $499/mo on Plus. Layer two is the GMV percentage: 1.49% of every dollar on Starter, 1.34% on Plus. Layer three is the per-order flat fee: $0.19 on both plans, charged on every recurring and one-time order.

Run the Starter numbers on a $5M brand doing roughly 12,500 orders a year (about a $400 average order value): $99 × 12 = $1,188 in SaaS, plus 1.49% × $5,000,000 = $74,500 in GMV fees, plus $0.19 × 12,500 = $2,375 in per-order fees. Total: $78,063 a year, or 1.56% of GMV. The subscription is 1.5% of that number. The percentage line is 95% of it.

When I talk to founders running a subscription brand this size, the layer they consistently miss is that middle one. They see "$99 a month" on the plan page and file it next to their email tool. The pattern we see again and again is that the GMV percentage doesn't appear anywhere until you pull the actual invoices, because the dashboard nets it out of payouts rather than billing it as a line.

That flat $0.19 is the quiet one to watch. At $400 AOV it is nothing, about 0.05% of the order. But subscription brands skew toward low-ticket replenishment: coffee, supplements, pet food, refills. At a $60 AOV that same $0.19 is 0.32% of the order, more than six times the drag. One operator on a food-and-supplement subscription flagged exactly this while building a unit-economics model: the platform "takes something" on every order that runs through the app, and it does not scale down with your ticket. If your AOV is under $80, model the flat fee as its own line, because it is doing real damage.

Stacking Recharge on Shopify Payments: the true processing rate

Here is where the 1.49% headline falls apart. Recharge is an app layer. It does not replace your payment processor, it sits on top of it. So on a single $400 subscription order you pay the Recharge fee and the Shopify Payments fee, on the same dollar.

Shopify Payments runs 2.9% + $0.30 on the Basic plan, dropping through the tiers to 2.15% + $0.30 on Plus. Add the Recharge Starter fee (1.49% + $0.19) and the combined rate at $400 AOV lands at 4.5% of GMV on Shopify Basic, 4.2% on Standard, 4.0% on Advanced, and 3.8% on Plus. That is the number your subscription revenue is actually paying to move, not the 1.49% on the app's pricing page, and how large a share of total revenue subscriptions represent varies by category (our average ecommerce subscription revenue share by vertical benchmark puts the range in context).

The CFO move here is to book the platform fee as part of COGS, inside your contribution margin, not marketing. When we've reviewed a subscription P&L with an operator, the fee that compounds like this belongs in cost of goods, the same bucket as shipping, because it scales one-for-one with every order shipped. Park it in "software" or "marketing" and your contribution margin per subscriber reads better than it is, which is exactly how brands end up scaling a channel that loses money at the unit level.

What Ordergroove and direct Stripe actually cost

Two alternatives change the fee math in opposite directions. Ordergroove is the enterprise option: it does not publish pricing and, critically, it does not process payments. It routes orders through your own gateway, so there is no processing markup layered on top. Third-party comparisons put its minimum around $2,917/mo, with a revenue share in the 0.5-1.5% range and a $5,000-$25,000 implementation. Modeled at a midpoint it lands near $62,000 a year, but every one of those inputs is an estimate. If you are seriously comparing it at $5M, get a real quote before you trust any number here, including mine. If you are at the stage of comparing platforms, Recharge vs Smartrr vs Skio breaks down how those alternatives stack up on fees and features.

Direct Stripe pulls the fee down hardest. Stripe Billing, the subscription management layer, is 0.7% of billing volume pay-as-you-go, on top of the 2.9% + $0.30 you already pay for Stripe Payments. Because you are already paying Payments, the incremental cost of adding Billing is just that 0.7% plus the per-order Payments fee, which pencils to about $38,750 a year at $5M, or 0.78% incremental. On the fee line alone, it is half the cost of Recharge.

The catch is that Stripe Billing is billing infrastructure, not a subscription product. It does not give a DTC brand a customer portal, a "manage my subscription" UX, swap-and-skip flows, or a native Shopify integration, and choosing the right subscription app for Shopify involves more than the fee line alone (see Recharge vs Bold vs Loop for a side-by-side). You build or buy all of that. Directional engineering-service estimates for an in-house build run an estimated $40,000-$120,000 up front and $50,000-$100,000 a year to maintain. That is the real crossover: Stripe Billing wins on fees above roughly $5M GMV only if you have the engineering capacity to close the product gap, otherwise you are buying back the saving in dev time.

The cost nobody models: involuntary churn as a revenue leak

Everything above is on the pricing page somewhere. The largest cost usually isn't. Involuntary churn, the subscribers you lose to failed payments rather than deliberate cancellation, runs 20-40% of total subscription churn depending on category (third-party category data puts digital goods near 29% and merchandise near 23%). First-attempt payment failure averages 7.9% across industries and closer to 9% on recurring charges, driven mostly by insufficient funds and expired or reissued cards.

On a $5M brand at 6% monthly churn with 25% of that churn involuntary, roughly $75,000 of GMV sits in the recoverable at-risk pool from payment failures at any given time, before you recover a cent. That is one month of steady-state exposure; push involuntary share to 40% and the monthly pool climbs toward $120,000. This is the number that never appears on a Recharge invoice, and the annualized leak behind it dwarfs the platform fee.

The reason operators miss it is reporting. A $30M+ CPG brand we know only caught a payment-fee spike by auditing line items, not from any dashboard, and involuntary churn hides the same way: the platform reports a failed-payment cancellation as ordinary churn, so it looks like a retention problem instead of a recoverable billing problem. When we've struggled to explain a soft LTV number, the answer is often sitting in the decline codes, not the cancel-reason survey.

The dunning gap: what you recover vs what you leave on the table

Failed payments are recoverable, and this is where the real money moves. Dunning, the automated retry-and-notify sequence after a card fails, recovers about 50% of failed charges on static rule-based retries. AI-optimized retry timing pushes that above 70%. Recurly's data shows 90% of all recoverable revenue is captured within the first 10 days of the failure, so front-loading retry budget matters far more than stretching the cadence out for weeks.

Do the arithmetic on that 20-point swing. On the $75,000 at risk in the base case, weak dunning recovers about $37,500 and leaves $37,500 on the floor; strong dunning recovers about $52,500, cutting the loss to $22,500. That is a $15,000 difference on one month's at-risk pool, roughly $180,000 a year on this single churn scenario, and it scales up fast at higher churn or higher involuntary share. For most brands this gap dwarfs the delta between Recharge plans, which is why obsessing over Starter-versus-Plus while running default dunning is optimizing the wrong line.

The Recharge line on your P&L is the cost you can see. The involuntary-churn leak is the one that is usually bigger, entirely recoverable, and completely invisible until someone reads the decline codes instead of the dashboard. Fix the dunning before you renegotiate the platform.

The decision framework: stay, switch, or build

Put it together and the thresholds are clean. Under about $2M subscription GMV, Recharge Starter is the right call, the flat SaaS fee is trivial and you are not big enough for the GMV percentage to hurt. From roughly $3.2M to $10M, Recharge Plus starts to pencil, because the 0.15-point rate cut finally beats the higher subscription fee. Evaluate Ordergroove at $5M-plus only if you need true omnichannel or retail-media subscription, and only after a direct quote. Consider direct Stripe Billing above $5M only if you have engineering capacity to build the subscriber experience Recharge gives you out of the box.

But whatever tier you land on, the sequence matters. When I talk to founders agonizing over a $2,700 delta between Recharge plans, the first question is always what their dunning recovery rate is, because a 20-point improvement there is worth five to ten times the plan saving, and a fractional CFO will typically run this cost model before recommending any platform change. Fix the leak you cannot see before you optimize the fee you can.

Sources and methodology

Recharge published pricing is the anchor for every platform-fee number here. Starter is $99/mo at 1.49% + $0.19 per transaction; Plus is $499/mo at 1.34% + $0.19; Custom is quote-based. Fees apply to one-time and recurring orders. See getrecharge.com/pricing. Recharge's support documentation indicates fees can apply even on refunded orders, which is worth verifying against current policy before modeling refunds out of your cost base.

Stripe Billing and Payments rates come from Stripe's own pricing pages. Billing is 0.7% of billing volume pay-as-you-go; Payments is 2.9% + $0.30 domestic, with cross-border and FX surcharges on international cards. The incremental-cost model assumes you already run Stripe Payments and are adding only the Billing layer. See stripe.com/billing/pricing.

Payment-failure and dunning benchmarks are triangulated across several published sources. First-attempt failure of 7.9% cross-industry is from Kaplan's 2025 subscription statistics; the 90%-within-10-days recovery window and the 50%-to-70% dunning recovery spread are from Recurly's failed-payment recovery analysis. Involuntary-share-of-churn figures draw on Churnkey and Stripe category data and Paysafe's hidden-cost-of-failed-payments research.

Ordergroove figures are estimates, not published pricing. Ordergroove does not disclose rates publicly and routes payments through the merchant's own gateway. Minimum, revenue-share, and implementation ranges are compiled from third-party comparison content, including Loopwork's platform comparison. Treat any Ordergroove number as directional until you obtain a direct quote.

All models use a $5M GMV, ~12,500-order, $400-AOV base case. Lower-AOV brands should re-run the per-order flat fee, which becomes proportionally much larger below $80 AOV. Involuntary-churn dollar figures are approximate and sensitive to churn rate and involuntary share, both of which vary by category. The "$75,000 at risk" figure models roughly one month of steady-state involuntary exposure (GMV × monthly churn rate × involuntary share), not cumulative annual churn; it approximates the recoverable pool sitting in the failed-payment queue at any given time. A reader who annualizes 6% monthly churn over 12 months will arrive at a larger number; the dunning ROI comparison in the table above holds either way.

Frequently asked questions