Chapter 26 — Key Takeaways

Technology: POS Systems, Online Ordering, Delivery Platforms, Reservation Tech, and the Digital Restaurant

The core claims

1. The POS is the system of record, not a cash register. Every number in Parts III, IV, and VII — menu mix, theoretical food usage, sales per labor hour, ticket time, turn time, RevPASH, the audit trail, the weekly flash report — is a query against what a server entered on a terminal. Choose the stack badly and you pay twice: in fees, and in every decision you then make blind.

2. Payment processing is the largest and least examined line in the budget. At Bellwether it is \$43,573 — 59% of the entire technology cost and more than every software subscription combined. It never arrives as a bill; it is netted out before the deposit lands. A cost you never see is a cost you never manage.

3. The fee base is bigger than your sales. The processor charges on the settled transaction, which carries the sales tax you are holding for the state and the tip you are holding for your staff. Bellwether's card volume is \$1,743,750** against \$1,395,000 of card sales — 1.25× — and one dollar in five** of the processing bill is charged on money that was never the restaurant's.

4. Your processor keeps 20 cents of every dollar you pay it. Interchange takes 73% and goes to the bank that issued the guest's card; network assessments take 7%. That is why "shopping processors" has a ceiling and why pricing model and card mix move more money than any sales call.

5. A KDS is justified by a timestamp, not by the death of paper. Paper tickets produce no record, which means no ticket time, which means Chapter 14's 22-minute not-to-exceed is a feeling rather than a standard. Watch for early bumping — the metric becomes the target the moment anyone is judged on it.

6. Every un-integrated boundary is a permanent labor cost. Bellwether's five boundaries, done by hand, are 227.5 hours and \$6,643 a year — booked as labor, never attributed to the software decision that caused it. But integration is not free either: certified connections carry fees, break on updates, and fail silently, which is worse than a manual process that fails loudly.

7. On a marketplace order you rent a transaction; on a first-party order you acquire a guest. The fee is not the main difference. Chapter 23 established that the second visit is where profitability lives, and the marketplace owns it. Chapter 28 owns the economics.

8. Lock-in is arithmetic, not a clause. Switching Bellwether's POS costs \$14,294 — three years of the subscription. When the renewal comes with a \$720 increase, you will pay it. All of your leverage exists before the signature; none of it exists after.

9. Technology is 4.73% of sales and has no line on the P&L. It hides inside other operating expenses, where it is the largest single occupant — a third of the line, bigger than utilities, bigger than insurance, bigger than marketing, and 28% of the plan's operating profit.

The formulas

$$\text{Effective rate} = \frac{\text{total fees for the period}}{\text{total card volume for the period}}$$

$$\text{Processing as \% of sales} = \frac{\text{total fees for the period}}{\text{total net sales for the period}}$$

Compute both, every month. The first tells you whether your agreement is competitive. The second is what lands on your P&L. At Bellwether they are 2.50% and 2.81% — and budgeting the first leaves you \$4,823 short for the year.

CARD VOLUME =  card sales  +  sales tax collected on them  +  tips added on card

The rules of thumb

The key terms

Point of sale (POS) · kitchen display system (KDS) · inventory management software · scheduling software · reservation platform · online ordering (first-party vs. marketplace) · integration and API · payment processing · interchange · effective rate · downgrade · surcharging (vs. cash discounting) · data ownership · technology cost as a percentage of sales

What you should be able to do Monday morning

Pull last month's merchant statement and compute your effective rate twice — total fees over card volume, and total fees over net sales. Then request the interchange category detail, ask in writing what every fee under \$60 actually buys, verify the batch closed every night, and send the statement to two competing processors asking them to quote interchange-plus against it.

Then open the POS and check three settings: is a cover count required to tender, is the open-item button behind a manager code, and can a server void their own closed check. Those three take ten minutes and they determine whether the next six months of reports are worth reading.