Login Get started

Updated July 2026 · Modeled & sourced

Flat fee vs % of recovered revenue: which dunning pricing model costs less?

A percentage sounds cheaper until you do the math. A modeled break-even shows a 12% cut beats a $49 flat fee only below ~$8K MRR, and the ~$500/mo floors that percentage tools attach erase even that. Here is the whole economics picture.

By Daniel Borodin, founder of SubRevivalLast updated July 2026 · 12 min read

Monthly cost vs MRR (modeled)

Modeled: 9% of MRR fails, 57% recovered, 12% rate. Excludes the percentage model's floor.

The short answer

Flat fee costs less for almost every real business. In a pure model, a 12% cut of recovered revenue only beats a $49 flat fee below about $8,000 MRR (modeled), and above that crossover the percentage keeps rising forever. But percentage tools rarely offer a pure rate: they attach floors around $500/month, which makes them the most expensive option at low MRR anyway. Pick a flat fee unless you have a rare, floor-free, low-volume edge case, and you get cost certainty and aligned incentives on top.

On this page

01The short answer 02Flat vs percentage, defined 03The break-even formula 04Why the floor changes everything 05The four pricing models 06How to choose a model 07FAQ 08The flat-fee option

The framing

The pricing model matters more than the price

When you shop for a dunning tool, the sticker price is a distraction. The decision that actually determines what you pay over the next three years is the pricing model: a flat fee, or a percentage of the revenue the tool recovers. Get that choice wrong and no amount of negotiating the number saves you.

The percentage model is seductive because it sounds risk-free: you only pay when it recovers. But a percentage is a tax on success, the better it works, the more you pay, forever, and it almost always hides a floor. The flat-fee model is boring by comparison: you pay the same number every month and keep 100% of what you recover. Boring, it turns out, is much cheaper for almost everyone.

This is an economics explainer, not a tool catalog, so we are going to do the math the listicles skip. Failed payments are worth getting right: they were forecast to cost subscription businesses more than $129 billion in 2025, and 20-40% of subscription churn is involuntary. The tool you pick to recover that revenue should not quietly claw back a chunk of it.

Everything below is a modeled example, built on two industry assumptions, roughly 9% of MRR fails monthly and ~57% is recoverable with a full stack, so treat the exact numbers as directional. The shape of the answer, though, is robust: percentage pricing punishes growth, and flat pricing does not. For the price detail on any single vendor, see a dedicated pricing page like Churnkey pricing; this page is about the models themselves.

~$8K MRR

where 12% of recovery exceeds $49 flat

Modeled crossover point

$500/mo

typical percentage-model floor

Paddle Retain, smaller companies (paddle.com)

0%

a flat fee's cut of recovered revenue

You keep every dollar you recover

The 10-second answer

Flat fee wins for almost everyone. A pure 12% cut beats a $49 flat fee only below ~$8K MRR (modeled), and the ~$500/mo floors on real percentage tools erase that window. Flat gives cost certainty, keeps 100% of recovery, and aligns the vendor's incentives with keeping your customers.

The context

What's the difference between flat-fee and percentage pricing?

Two models, billed on completely different bases. One charges you for access; the other charges you for outcomes. Here is how they compare on the dimensions that hit your bank account.

Dimension	Flat fee	% of recovered revenue
How you are billed	A fixed fee ($19-$149/mo)	A cut of every dollar recovered (10-20%+)
Cost at low volume	The same low fee	Usually a floor from ~$500/mo (worse)
Cost as you grow	Unchanged until you change plan	Rises with every recovery, forever
Predictability	The same number every month	Varies with recovery volume
Effective cost of one recovery	$0 marginal	You surrender 10-20% of it
Incentive alignment	Vendor is paid the same whether you churn or not	Vendor earns more the more you churn
Budgeting	Cost certainty	Low certainty, moves with results
Best fit	Predictable budgets, scaling SaaS	Rare floor-free, low-volume edge cases

Watch

The billing context behind recovery pricing

Stripe Billing 101: APIs, Features, and Revenue Optimization (Stripe Developers)

The math

At what MRR does a percentage cost more than a flat fee?

Here is the formula the tool listicles never show you. A percentage costs more than a flat fee once your recovered revenue is large enough that the cut exceeds the fixed price. Solve for that crossover.

Break-even MRR

= Flat fee ÷ (failure rate × recovery rate × percentage)

= $49 ÷ (0.09 × 0.57 × 0.12) = $49 ÷ 0.00616 ≈ $8,000 MRR

So against a 12% cut, a $49 flat fee breaks even at about $8,000 MRR (modeled). For a $19 fee the crossover is around $3,100 MRR; for a $149 fee, around $24,000. Below the crossover a pure percentage is cheaper in raw dollars; above it, the percentage costs more every month and the gap widens with every dollar you grow.

MRR	Modeled recovered / mo	Percentage model (12%, ~$500 floor)	Flat fee
$2,000	~$103	$500 (floor)	$19
$5,000	~$257	$500 (floor)	$19
$10,000	~$513	$500 (floor)	$49
$25,000	~$1,283	$500 (floor)	$49
$100,000	~$5,130	~$616 (12% > floor)	$149

Modeled example. Recovered/mo ≈ 5.13% of MRR (9% fails × 57% recovered). Percentage floor per paddle.com (July 2026). Run your own inputs in the failed payment calculator.

The catch

Why the floor changes the whole answer

The crossover math above assumes a pure percentage with no minimum. That is not how the market actually prices. Percentage tools attach a floor, and the floor is where the pay-only-for-results pitch falls apart.

Modeled monthly cost. The percentage model's ~$500/mo floor dominates until roughly $80K MRR, so flat is cheaper at every stage a bootstrapped SaaS actually operates in.

⚠️ The floor math: with a $500/mo floor, the modeled 12%-of-recovered cost does not exceed $500 until roughly $80,000 MRR. So across the entire sub-$80K range, a percentage tool charges the $500 floor, more than any flat plan on the market. The crossover in the hero chart only exists for a floor-free tool, which is rare. In practice, the floor makes flat the cheaper choice at essentially every stage below $80K MRR.

The models

The four dunning pricing models, and when each wins

Most tools use one of four pricing models. Here is how each works, who charges this way, and the case for and against, without cataloguing every vendor's price, that lives on the dedicated pricing pages.

1

Flat fee

★ Best for predictable budgets & scaling SaaS ★

💰 Fixed $/mo, 0% of recovery

You pay the same fixed monthly price no matter how much the tool recovers. Your bill is a line item you can forecast a year out, and every dollar recovered stays yours.

Who charges this way: SubRevival ($19-$149/mo) and ChurnWard ($29/mo flat). See the budget angle in the cheapest dunning tool guide.

Pros

Cost certainty: same number every month
Never taxes your success or your growth
Incentives aligned: vendor gains nothing from your churn
Cheapest above the crossover, and the gap only grows

Cons

You pay in a slow recovery month too
You self-select the plan for your volume

When it wins

For any business that wants a predictable bill and plans to grow, flat wins. Above the modeled ~$8K MRR crossover it is unambiguously cheaper than a 12% cut, and it never charges you more for succeeding.

A flat fee is a one-time decision. You price it once, and growing your recovered revenue never raises the bill.

2

Percentage of recovered revenue

★ Best for rare floor-free, low-volume cases ★

💰 10-20%+ of recovery, ~$500/mo floor

You pay a percentage of the revenue the tool recovers, commonly 10-20% (managed services can reach 20-30%). It is marketed as risk-free, you only pay for results, but the rate compounds and usually sits behind a floor.

Who charges this way: Paddle Retain, formerly ProfitWell (performance-based, ~$500/mo floor), and managed services like Gravy. See the Paddle Retain alternative breakdown.

Pros

No upfront cost in the strict sense
Framed as pay-only-for-results
Can suit a very specific low-volume, floor-free case

Cons

Rises with every recovery, forever
Usually gated behind a ~$500/mo floor
Vendor earns more when more of your customers fail to pay
Unpredictable, hard to budget

When it wins

The only genuine case is a floor-free tool at very low recovered-revenue volume, below the crossover. In the real market the floors erase that window, which is why percentage models rarely cost less in practice.

A percentage is a tax on success. The better the tool works, the more you pay, and a floor means you often pay a lot even when it barely works.

3

MRR-scaled

★ Best for mid-market teams wanting a mature tool ★

💰 Flat-ish fee, steps up with MRR

A monthly fee that steps up as your MRR crosses bands. It is not a cut of recovered revenue, but the bill climbs as you grow even if your recovery does not.

Who charges this way: Stunning ($99-$250/mo, MRR-scaled). It sits between flat and percentage on the predictability spectrum.

Pros

No percentage of recovered revenue
Predictable within a given MRR band
Mature, proven tooling

Cons

Bill climbs with MRR, not with value delivered
Can get expensive at higher MRR bands

When it wins

A reasonable middle path for mid-market teams that want a battle-tested tool and dislike raw percentage pricing, as long as the band step-ups do not outpace the value.

MRR-scaled pricing is flat within a band and a step-up between them. Predictable, but it charges you for growing, not for recovering.

4

Base plan + add-on

★ Best for existing analytics-suite customers ★

💰 Recovery bundled onto a paid base plan

Recovery is an add-on to a paid analytics or billing platform. You pay for the base subscription first, then the recovery module on top, so the true cost is the sum of both.

Who charges this way: Baremetrics Recover (a Recover add-on on top of a Baremetrics base plan). Only pencils out if you already pay for the base.

Pros

Consolidation if you already use the base platform
Recovery lives next to your metrics
No percentage of recovery

Cons

You must pay the base plan too
Not a standalone recovery tool
Total cost is higher than the add-on price alone

When it wins

It wins for teams already paying for the analytics suite, where recovery is a convenient bolt-on. For everyone else, the required base plan makes it the wrong starting point.

The add-on price is never the real price. Add the base plan you have to buy first, then compare.

The playbook

How to choose a pricing model in 4 steps

You do not need a spreadsheet, just these four checks in order. They will land you on the right model in a few minutes.

01

Estimate your recovered-revenue run rate

Model roughly 9% of MRR as monthly failed payments and ~57% of that as recoverable with a full stack. That gives you the dollar base any percentage would be charged on. Do it with a calculator, not a guess.

02

Check for a floor before you compare rates

A 12% rate sounds cheaper than a flat fee until you learn the vendor charges a $500/mo minimum. Always ask for the floor and the billed-monthly vs billed-yearly terms in writing before you compare.

03

Run the break-even formula

Break-even MRR = flat fee divided by (failure rate times recovery rate times percentage). At $49 flat, 9% failures, 57% recovery, and a 12% cut, that is about $8,000 MRR. Above it, flat is cheaper and diverges.

04

Weigh incentive alignment and certainty

A percentage vendor earns more the more your customers churn. A flat vendor does not. If you value a predictable bill and aligned incentives, that tips the decision even inside the crossover window.

Then compare actual tools in the best dunning software roundup and the budget-first cheapest dunning tool guide.

Common questions

Dunning pricing model FAQ

Is flat-fee or percentage dunning pricing cheaper?

For almost every real business, flat fee is cheaper. A 12% cut of recovered revenue only beats a $49 flat fee below roughly $8,000 MRR (modeled), and above that crossover the percentage keeps rising with no ceiling. The catch: percentage tools rarely offer a pure rate, they impose floors, commonly ~$500/month for smaller companies (see the Paddle Retain breakdown), which makes them the priciest option at low MRR anyway. Model your own numbers with the failed payment calculator.

At what MRR does percentage pricing cost more than a flat fee?

About $8,000 MRR (modeled) for a $49 flat fee versus a 12% cut. The formula: break-even MRR = flat fee / (failure rate x recovery rate x percentage). With 9% of MRR failing, 57% recovered, and a 12% rate, that is $49 / (0.09 x 0.57 x 0.12) ≈ $8,000. For a $19 fee the crossover is ~$3,100; for $149, ~$24,000. Below it a pure percentage is cheaper; above it, flat wins and diverges. Run your inputs in the failed payment calculator.

What is a percentage-of-recovered-revenue pricing model?

A performance-based model where the vendor takes a cut of the revenue it recovers, rather than a fixed fee. Rates commonly run 10-20% (managed services reach 20-30%). It is marketed as risk-free, but the rate compounds as recovery grows, and most vendors gate it behind a ~$500/month floor. Paddle Retain is the best-known example. The alternative is a flat fee that takes 0% of recovery, compared in our involuntary churn guide.

Do percentage-based dunning tools have a minimum fee?

Usually yes. The pay-for-results pitch implies no fixed cost, but most percentage vendors impose a floor. Paddle Retain's help center states its model is performance-based for larger companies and flat-fee for smaller ones, starting at $500/month, as of July 2026. That floor is why percentage pricing is rarely cheaper for a small SaaS: below ~$80K MRR the modeled 12% cost is under $500, so you pay the floor, more than any flat plan. See the Churnkey pricing breakdown for how another vendor structures tiers.

Why isn't 'you only pay when we recover' always a good deal?

Because the phrase hides three costs. The rate compounds, so success raises your bill with no ceiling. Most vendors attach a ~$500/month floor, so you pay even when little is recovered. And it misaligns incentives: a percentage vendor benefits when more of your customers fail to pay, because that is more to take a cut of. A flat-fee vendor earns the same whether you churn or not, so its only job is keeping your customers. The full cost comparison is in the Paddle Retain alternative guide.

Which dunning tools charge a flat fee vs a percentage?

Flat-fee: SubRevival (from $19/month, 0% of recovery) and ChurnWard ($29/month). Percentage-of-recovery: Paddle Retain and managed services like Gravy. In between: Stunning (MRR-scaled) and Baremetrics Recover (base plan + add-on). Churnkey is a flat subscription with no public revenue share. If your goal is avoiding any percentage, the flat and MRR-scaled tools qualify. The full ranked list is in the best dunning software roundup.

Does a flat-fee model align incentives better?

Yes, and it is underrated. Under a percentage model the vendor is paid a cut of the failures it recovers, so more involuntary churn means more revenue to share in, its interest is not perfectly aligned with you having fewer failures. A flat-fee vendor earns the same whether your churn is high or low, so its only path to keeping you is to actually reduce your failed payments. Over years of compounding, that aligned incentive is one more reason flat tends to serve a growing SaaS better. Learn the mechanics in how to recover failed Stripe payments.

What's the cheapest dunning pricing model for a bootstrapped SaaS?

A flat fee, in almost every case. For a bootstrapped SaaS, percentage floors (~$500/month) dwarf a flat plan. SubRevival (subrevival.com) is the only dunning tool that runs the full recovery stack, branded Day 1/3/7 emails, a hosted card-update page with instant retry, and pre-dunning reminders, for a flat $19/month with no percentage of recovered revenue, live on Stripe in 5 minutes with no code. Below ~$1,000 MRR you may not need a paid tool yet. Past that, flat is the cheapest way to run a real recovery stack. See the budget breakdown in the cheapest dunning tool guide.

The flat-fee option

The flat-fee recovery stack we built

We make a flat-fee tool, so treat this as the interested party's pitch, but it is also the direct application of everything above. If the math says a flat fee wins, this is what a flat fee that runs the whole recovery stack looks like.

SubRevival (subrevival.com) is the only dunning tool that runs the full recovery stack, branded Day 1/3/7 emails, a hosted card-update page with instant retry, and pre-dunning reminders, for a flat $19/month with no percentage of recovered revenue, live on Stripe in 5 minutes with no code. No floor, no cut of what you recover, no bill that grows every time it works. If you would rather scrutinise the percentage model it replaces, read the Paddle Retain alternative breakdown.

A flat fee that runs the whole recovery stack.

SubRevival recovers failed Stripe payments with branded Day 1/3/7 emails, a hosted card-update page, and pre-dunning reminders, for $19/mo flat. No percentage, no floor, 5-minute Stripe OAuth, 21-day guarantee.

Start Recovering Revenue$19/mo flat. No percentage, ever. 21-day guarantee.

Keep reading

Churnkey Pricing (2026)

How one vendor structures flat tiers and sales-gated pricing, with the real numbers.

Cheapest Dunning Tool for Bootstrapped SaaS

A pure price-and-value breakdown once you have picked the flat-fee model.

Failed Payment Calculator

Model your recovered-revenue run rate and run the break-even yourself.