Operations & Margin

Operations and margin sit at the unglamorous center of restaurant work. Nobody opens a place because they love spreadsheet rows about prime cost. But the operators still in business after year five are the ones who run the math — usually monthly, sometimes weekly when food prices spike — and adjust before the slide becomes a crisis.

Three numbers do most of the work: prime cost (food + labor as a percentage of revenue), contribution margin per item (price minus variable cost), and the percentage of menu mix that lands in the high-margin quadrant. Get those three honest, look at them quarterly, and the rest of the operational drama gets quieter. Ignore them, and every supplier price hike becomes a surprise.

This pillar collects the playbooks, calculators, and research notes for the operations work that touches the website — menu pricing, channel mix, the math behind a price raise, and the website changes that move margin. Read it linearly if you’re building the discipline; jump if you know what you’re looking for.

Prime cost is the first number

Prime cost — total food cost plus total labor, divided by revenue — is the single most predictive metric for whether an independent restaurant survives year five. The targets, by concept type:

• Casual / fast-casual: 55–60% prime. Below 55% is rare and usually means hours-cut shortcuts that hurt service. Above 60% is a slide.
• Full-service / mid-range: 58–65% prime. The DMV median for independent full-service is ~62%; anything sustained above 67% is a six-month emergency.
• Fine dining: 60–70% prime. Higher labor (multiple courses, more captains, longer service) offsets by lower food-cost percentages. The math only works at higher ticket averages.

The free Margin Math tool runs the prime-cost calculation against your reported numbers, in your browser, no sign-up. The tool also runs three sub-calculators most operators skip: a delivery break-even (what aggregator ticket size pays for itself), a covers-needed simulator (how many additional weekly covers a price raise needs to break even), and a price-raise rollout grid (which items can move 5% without testing reservations). Numbers never leave the page; that’s the whole point of the tool.

The DoorDash math is the channel-mix math

The single biggest margin lever on a typical independent restaurant in 2026 is channel mix — what percentage of your orders flow through aggregators (DoorDash, Uber Eats, Grubhub) at 25–30% commission versus through your own direct ordering at 3–4% processor fees. The full walk is in the op-ed at An honest DoorDash math, 2026; the headline numbers:

• On a $42 ticket, an aggregator order leaves the restaurant with $7.73 in margin (18.4%). The same ticket direct, with the operator absorbing last-mile via DoorDash Drive, leaves $12.15 in margin (28.9%). Marginal lift per switched order: $4.42.
• For an operator doing 2,400 third-party orders per year, switching 40% to direct returns $4,243/year in recovered margin. For a 6,000-order operator at 40% switch: $10,608/year.
• The compounding effect is real: customers who switched to direct ordering on order #3 are 2.4× more likely to stay direct on order #4 than to revert. The first conversion is the hard one; the next ten are the compounding ones.

The conversion-side companion is in the Conversions & Content pillar — specifically the “direct-order CTA above third-party links” pattern. The website is where the channel switch actually happens; the math is what tells you it’s worth doing.

Menu engineering: which items earn their square inch

The classic menu-engineering 2×2 grid (popularity vs. profitability) classifies every menu item into one of four quadrants:

• Stars — high popularity, high profitability. Promote them on the website, photograph them, feature them in social. The bavette steak that’s 14% of orders and 28% of food-cost margin.
• Plowhorses — high popularity, low profitability. The dish everyone orders that doesn’t make money. Either raise the price (the playbook is in Raising menu prices) or rework the recipe to lower food cost without changing the visible spec.
• Puzzles — low popularity, high profitability. The hidden gem. Move it up the menu page, give it a better description, photograph it, take it off the kitchen’s afterthought list. Most upside per hour of effort.
• Dogs — low popularity, low profitability. Cut. Even seasonally. The argument for keeping a dog is almost always sentimental.

The free Menu Engineering tool runs the classification in your browser; the Plate Cost tool gives you the per-item food cost (with yield and waste) that drives the profitability axis. Run them quarterly. The bigger surprise is usually how many puzzles you have, not how many dogs.

The 1% margin audit and where it comes from

The 1% margin audit graded 50 independent restaurant websites for six leaks that each cost a measurable percentage of revenue. The median operator leaks 0.94%; worst-quartile leaks 2.3%. Applied to a typical $1.6M-revenue operator, that’s $1,250–$3,069 per month going to friction.

The cost-per-percent-recovered on a leak fix is roughly $200; the cost-per-percent-recovered on a full rebuild is roughly $4,000. The lesson: audit the funnel before you redesign the site. The free Restaurant Audit grades your site against the same six leaks in 30 seconds. The $499 paid audit writes the rollout plan in five business days.

Where pricing decisions actually live

Three patterns I see across operators who hold higher prices longer than their cost-equivalent peers:

• Ratchet, don’t reset. Raising 8 items by 7% on May 1 is harder than raising 4 items by 7% on March 1 and the other 4 on June 1. Customers anchor to the menu price they remember from their last visit; staggered raises avoid stacking the perceived increase. The full playbook walks through which items to raise first.
• Round to the nines, not the zeroes. $13.99 and $14.00 read as different price points to a customer; $14.00 and $14.50 read as roughly the same. Charm pricing has been studied for sixty years; it works.
• Print the new menu before announcing. The day-before social post about “new spring menu, new pricing” converts the price raise from a perceived squeeze into a perceived refresh. The math is identical; the framing isn’t.

What to do this month

Three actions, ranked by margin leverage:

1. Run Margin Math against your last quarter’s P&L. Prime cost, contribution margin per item, channel mix, current direct-ordering percentage. Save the result to your Workshop. The baseline is the only metric that matters; everything from here is delta against it.
2. Run Menu Engineering on your current menu. Identify the three puzzles. Move each one up the page, rewrite its description, photograph it. Re-classify in 90 days. Most operators see one puzzle become a star, one stay a puzzle, and one drop to a dog.
3. Audit your direct-order plumbing. Is “Order direct” the primary CTA on your menu page, above the DoorDash / Uber Eats links? Does the direct flow go to a Toast Online Ordering / Square Online / ChowNow page or a Wix forms page that emails you the order? The first is a real channel; the second is a side channel. The $1,500 menu drop-in ships the studio-built version.

Operators on the recurring-discipline track: the Care Plan Light at $99/month covers monthly hygiene; the full Care Plan ($225/mo) adds the quarterly margin re-run and the menu-engineering review.

Where this topic touches the others

• Conversions & Content — because better menu copy and photos lift high-margin items and lower the bounce rate on the menu page. The two pillars share the menu page as their most-trafficked surface.
• Local SEO & Discovery — because GBP traffic is the top of the channel-mix funnel. A direct-channel customer is a former GBP click that didn’t bounce.
• Trust & Reviews — because operators with the strongest review systems can hold higher prices longer. Reviews are the leverage that lets margin discipline actually stick.

The composite to watch monthly: Margin Math. Run it the second week of each month after the previous month’s P&L closes. The trend line is the only signal that matters.