Why does nobody else build this calculator?

Because everyone who could build it is on the other side of the trade.

Jobber, Housecall Pro, Square, QuickBooks and Wave all publish free tools. None of them publishes one that puts their processing rate next to everyone else's and adds up the year. That is not a conspiracy, it is just an obvious commercial fact: processing is one of the best revenue lines any of them has, and a page that helps you spend less on it is a page that costs them money. The research behind this tool found no free comparison from any incumbent, and a fair amount of resentment about it in the reviews.

The result is that most contractors know their headline rate and nothing else. They can tell you "2.9%" and not what it came to last year, not what the same money would have cost somewhere else, and not what the identical card would have cost if it had been tapped instead of typed. Those three gaps are worth real money.

The number that should be on the wall is the annual one. At $25,000 a month and a 3.3% plus 30 cents online rate, processing costs about $10,000 a year. That is not a line item. That is a used truck, or a part-time helper for a season, and it is being spent 30 cents at a time where nobody looks at it.

What do the processors actually charge in 2026?

Published rates read off each company's own pricing page on 8 August 2026. Where a company publishes a range or a "starting at" number, that is said plainly rather than smoothed over.

Rates read 8 August 2026 from: Square's US pricing page · Stripe's pricing page · PayPal's merchant fee schedule (rates effective 15 July 2026) · Intuit's QuickBooks payment rates page (marked accurate as of 30 April 2026) · Jobber's pricing page · Housecall Pro's payments feature page ("card processing fees start at 2.59%") · Wave's pricing page · Helcim's pricing page. Square Invoices Plus at $20/mo and Housecall Pro's reported 2.49%–3.49% range are from an August 2026 price survey rather than a live vendor page — treat those two as softer. Rates in this category move; check before you sign anything.

Three things worth pulling out of that table.

The same card costs three different amounts. Square charges 2.6% plus 15 cents if you tap it on a reader in the driveway, 3.3% plus 30 cents if the customer pays an emailed invoice, and 3.5% plus 15 cents if you type the number in from a card you kept on file. On a $850 ticket that is $22.25, $28.35 and $29.90. Nothing about the work changed. If you are keying cards on file for a recurring maintenance customer, you are on the worst rate in the building and there is usually a cheaper way to do the same thing.

The fixed cents are a small-ticket tax. Wave's free tier charges 2.9% plus 60 cents. On a $1,000 invoice that 60 cents is 0.06% and irrelevant. On a $45 dog-walk or a $60 filter change it is 1.3%, and the all-in rate is 4.2%. Stripe's in-person 5 cents versus PayPal's 49 cents is the same story pointing the other way.

ACH is priced completely differently. Card fees scale with the invoice forever. ACH is either a small percentage with a cap or a flat charge, because moving money between two bank accounts costs roughly the same whether it is $50 or $50,000. That is why the gap between card and ACH explodes on big invoices, and why every big-ticket trade should be asking for a transfer as a matter of routine.

When is ACH worth pushing, and when is it not?

Here is the same invoice at six sizes, on Square's free online rate of 3.3% plus 30 cents against Square's ACH at 1% with a $1 minimum, and against Stripe's ACH at 0.8% capped at $5.

The cap is the whole story. An uncapped 1% ACH still costs $180 on an $18,000 draw, which is better than $594 but not free. A capped ACH costs $5 and stops. If your work runs to five-figure invoices — roofing, remodels, commercial service contracts — the cap is worth more than any rate negotiation you will ever have.

The honest other side. ACH is slower, typically three to five business days to settle, and it can fail days later for insufficient funds, which a card declines instantly instead. Some residential customers do not want to type a routing number into anything. And for a $180 drain clear the whole conversation costs more in time than the $4 you saved. The rule that survives contact with a real week: card for anything under about $1,000, ask about a transfer above it, insist on one above about $5,000.

Can you pass the fee on to the customer?

Sometimes, and it is not the free money it looks like.

Start with the arithmetic, because it defeats most people before the law does. Add a 3% surcharge to a $1,000 job and the customer is charged $1,030. Your processor charges its rate on the whole $1,030, not on the original $1,000. At 2.9% plus 30 cents that is $29.87 plus $0.30 = $30.17, so you keep $999.83. You are 17 cents short of the price you quoted, and you have added a line to the invoice that some customers will argue about. On a $150 service call with the same 3% surcharge, the customer pays $154.50, the fee is $4.78, and you keep $149.72 — the fixed 30 cents eats the surcharge and then some.

The break-even surcharge is always a bit more than the processor's percentage, and it is bigger the smaller the ticket. The calculator above works it out exactly. Roughly: break-even surcharge = (invoice + fixed fee) ÷ (invoice × (1 − rate)) − 1.

Then the rules

• Three states ban credit card surcharges outright — Connecticut, Maine and Massachusetts. Several more restrict them: Colorado caps at 2%, New York, New Jersey and Georgia limit you to your actual cost of acceptance, Minnesota has its own cap and disclosure rules. Texas has a prohibition on the books that is currently unenforced after a court ruling. Sources disagree about California, which some list as a ban and others list as allowed with disclosure, so that is a question for a lawyer in your state rather than a web page.
• Debit and prepaid cards cannot be surcharged anywhere in the US under card network rules, even in states where credit surcharging is legal. A flat "3% card fee" applied to every payment is the most common way small businesses break this.
• The networks cap it. A Visa surcharge is limited to 3%, or your actual cost of acceptance if that is lower. You also owe the networks 30 days' written notice before you start.
• It has to be visible. Signage at the point of sale and on any website that takes payment, and the surcharge itemised as its own line on the receipt.
• A cash discount is a different animal. Price the job at the card price and give a discount for cash or check, and you are not surcharging at all. It sidesteps most of the state restrictions. It has its own disclosure rules and it is not a loophole to be casual about.

Surcharge rules summarised 8 August 2026 from PaymentCloud's and Nickel's 2026 state-by-state surcharge guides, which agree on the network caps, on the debit prohibition and on Connecticut, Maine and Massachusetts, and disagree on California. This is a summary of secondary sources for planning, not legal advice. Confirm with your processor and your own state before you add a fee to an invoice.

The practical read for a small shop: surcharging is legal in most of the country, it is fiddly, it recovers slightly less than you think, and it occasionally annoys a customer at exactly the wrong moment. The version that works with almost no friction is the one in Step 2 — do not add a fee, offer to remove one. "Card's fine, or send a bank transfer and I'll take the processing fee off." Same money, and you are the one doing a favour.

What does waiting for your money actually cost?

Nothing in fees, and a real amount in working capital.

Whatever the funding delay, the same fee comes out. What changes is the size of the float that permanently sits inside the processor. An average month has about 21.7 business days, so at $25,000 a month a business day is about $1,152. Next-day funding parks $1,152 of your money on the other side of the wall at all times, and five-business-day funding parks about $5,760. That money never comes back to you; it just stays one payment cycle behind you forever. It is not lost, it is unavailable, and the difference matters at eight o'clock on a Monday when a supplier wants cash for a delivery.

Which is what makes instant payout so tempting and so expensive. Square charges 1.95% for an instant transfer, on top of the processing fee. Run $25,000 a month through it and you have spent $5,850 a year to move money you had already earned about three days earlier. Jobber charges 1% for payouts on weekends and holidays, and a Jobber user's complaint in the research put the other half of it plainly: "Ever Since Jobber started the 'Instant Payout', i now have to wait the full 5 business days to get paid." Faster is a product now, which means standard got slower.

Instant payout is worth it exactly once: when the alternative is not making payroll or not buying materials for tomorrow's job. As a habit it is one of the most expensive forms of credit a small business can get into, and it does not show up as a loan on anything.

Held funds, which is the risk that actually hurts

Fees are annoying and predictable. A freeze is neither.

The single most quoted line in the research: "Square froze my account with $4,200 in it. No warning, no phone call." Review and hold periods commonly run around 90 days, and the documented triggers include a sudden jump in processing volume, an unusual pattern against your normal, too many disputes, a high-risk category, or bank details that do not verify cleanly. Read that list again as a contractor. A sudden jump in volume is what April through July is.

This is not a Square problem. It is an aggregator problem, and it is the direct cost of the thing that made aggregators popular: you got approved in four minutes without underwriting, so the underwriting happens later, on your money. A traditional merchant account with a real underwriting process up front is far less likely to do this and takes days to open.

Three defences that cost nothing. Keep a second way to be paid that runs through a different company, so a freeze is a headache rather than an emergency. Tell the processor before a big month if you know one is coming — a $40,000 week after a run of $8,000 weeks looks like fraud to a model and like a good spring to a human. And do not let your operating balance live inside the processor; sweep it to your bank.

The other half of the same problem is disputes. A Housecall Pro user in the research put it as: "if a customer files a dispute, you are almost guaranteed to lose that money." Card disputes are weighted toward the cardholder, and a signed scope of work with a signature on it is the only thing that reliably helps. That is what the contract generator and the signature capture in the estimate generator are for.

Where is this calculator enough, and where is it not?

It is enough to answer the three questions people actually ask: what am I paying a year, would somebody else be cheaper at my volume and ticket, and is it worth asking for a bank transfer on this invoice. All three are arithmetic on published rates, and all three are things you can act on this afternoon.

It is enough to walk into a rate conversation with a number. It is enough to notice that you are keying cards on file at 3.5% when tapping them would cost 2.6%.

Here is what it cannot do, stated plainly.

• It cannot process a payment. It is a page of arithmetic in your browser. It is not connected to any processor, it cannot move a dollar, and it never will. To take a card you need a merchant account or an aggregator.
• The cheapest headline rate is not always the cheapest all-in. A 2.59% rate inside a $79 a month subscription costs more than a 3.3% rate with no subscription until you are running about $10,600 a month through it at an $850 ticket. The table above adds the software cost for exactly this reason, and the crossover moves with your volume.
• Interchange-plus needs an assumption. Helcim's row depends on what your actual interchange is, which depends on your customer mix. Debit cards are cheap, business rewards cards are expensive, and nobody can tell you your blend without your statement. The number in the box is an average, and averages lie about individual months.
• It does not model the extras. Chargeback fees, monthly minimums, PCI compliance fees, statement fees, terminal rental, early termination penalties. Aggregators like Square and Stripe genuinely do not charge most of these. Traditional merchant accounts frequently do, and that is where a quoted 2.1% turns into an effective 3%.
• It cannot price the risk of a freeze. There is no percentage you can put on 90 days without your cash in the middle of your season. It is real, it is rare, and it belongs in the decision even though it cannot go in a spreadsheet.
• Rates move. Every figure here was read on 8 August 2026. PayPal's current schedule took effect on 15 July 2026. Intuit's page is marked accurate as of 30 April 2026. Check before you commit to anything.

How do you use it?

1. Get one real month off your processor's statement. Total processed and number of payments. Divide for the average ticket. Do not estimate this — the ticket is the input that moves the answer most and everyone's guess is high.
2. Set the channel honestly. Most contractors think they are on the in-person rate and are actually on the online or keyed one, because the invoice gets emailed and paid from a phone. Look at a statement line and see what rate you were actually charged.
3. Run it with ACH at 0% first so you can see what card-only costs you a year. That is the number worth being annoyed about.
4. Then push the ACH share up to what you could realistically get. Ten percent of volume on bank transfer is easy for most shops. Half is achievable if your work is big-ticket and you ask every time.
5. Take your biggest normal invoice to Step 2 and see what one phone call is worth. That is the number that changes behaviour.
6. Check the surcharge math before you charge one, and check your state. Then consider offering the discount instead of adding the fee.
7. Print the summary and put it next to the phone when your processor's account manager calls to offer you a "better rate."

Common questions