DishCost: a $39/month recipe costing tool for independent restaurants
Highlights:
Built to fill a clear gap: nothing meaningful exists between spreadsheets ($0) and MarketMan ($199/mo)
Targets 660,000+ independent US restaurants running on 3-9% margins
Recipe builder, ingredient price tracking, automatic food cost calculations
SvelteKit 2 on Cloudflare Workers with D1
In-house product we built, use, and keep developing
The problem
Most independent restaurant owners don’t know what their dishes actually cost them.
That sounds unlikely for people running a business, but it makes sense when you look at how restaurants work. A chef-owner running a single location works 12-hour days. They manage the kitchen, deal with suppliers, handle staff. Sitting down to calculate that their mushroom risotto costs $4.12 per plate, that they’re selling it for $14, and that puts them at 29.4% food cost (which is fine)… that calculation either happens on the back of a napkin or it doesn’t happen at all.
The standard advice is “target 28-32% food cost.” But most restaurants can’t tell you their actual number. They’re eyeballing it. One restaurant owner described their process as “a two-hour waking nightmare” of hand-counting inventory and guesstimating.
And that’s not a small problem. US restaurants lose an estimated $25 billion a year to food waste and cost mismanagement. An individual restaurant doing $500K in revenue at 35% food cost instead of 30% is losing $25,000 a year. That’s the difference between making money and not.
Existing tools don’t help
There are tools that solve this. MarketMan, xtraCHEF, MarginEdge. They work. But they share the same issues.
MarketMan charges $127-339 per month per location, plus a setup fee. MarginEdge runs $300-500+/month. These are prices for restaurant chains, not for someone running a single taqueria.
One user described MarketMan’s onboarding as “very tedious, taking 2-4 weeks for a small restaurant.” MarginEdge “takes several months to really get the data in.” A chef working 12-hour days isn’t spending their one free evening importing inventory data.
And once you’re in, good luck getting out. MarketMan requires 12-month contracts with a 60-day cancellation notice. xtraCHEF requires a Toast POS subscription with a 2-year commitment. Trustpilot is full of restaurant owners paying thousands for software they stopped using months ago, unable to cancel. One described trying to leave as “jumping through flaming hoops.”
There’s a clear gap between “do it yourself in Google Sheets” and “pay $200/month for a system you’ll spend weeks setting up.” That’s where DishCost sits.
What DishCost does
The core question DishCost answers: “what does each dish on my menu actually cost me?”
You add your recipes and ingredients. You enter what you’re paying your suppliers for each ingredient. DishCost calculates the cost per dish and your food cost percentage. When a supplier raises prices (chicken goes from $2.80/lb to $3.40/lb), you update it once and every recipe using chicken recalculates automatically.
That’s the thing spreadsheets can’t do. In a spreadsheet, when an ingredient price changes, you have to find and update every formula that references it. In DishCost, you change it once. The dashboard shows you which recipes are now above your target margin. You decide what to adjust.
No inventory counting, no POS integration, no invoice scanning. Just the thing restaurant owners actually need: know what your food costs and when something changes.
It’s $29-39/month with no contracts. Cancel whenever you want.
How we built it
DishCost runs on SvelteKit 2 deployed to Cloudflare Workers, with Cloudflare D1 as the database. Pages load fast regardless of where the user is. We used Tailwind CSS 4 for the interface and Drizzle ORM for database access.
Most of DishCost was built using AI coding agents. The project doubled as a testing ground for how far agent-assisted development can go on a real product, not a demo. We shipped a full SaaS in weeks.
Where it’s going
DishCost is live and actively developed. The roadmap follows what restaurant owners ask for:
Price change alerts (“chicken went up 15% this week”)
Recipe scaling (double a recipe, costs recalculate)
Menu engineering suggestions (“your burger is underpriced based on your costs”)
Supplier price comparisons
We’re not trying to build the next MarketMan. DishCost does one job: help a restaurant owner know their food costs without spending weeks or hundreds of dollars to get there.