About Quebracho Empanadas MN
A Minneapolis empanada brand that spent five years turning a family recipe into a small, operational business — and the data journal that grew out of it.
The restaurant story
Quebracho Empanadas MN started as a home kitchen project in Minneapolis, built around an Argentine family recipe. Over several years it grew from pop-up orders and farmers-market weekends into a small operational business producing a few hundred hand-folded empanadas per week for retail customers and wholesale accounts across the Twin Cities.
Along the way the business picked up press attention from HuffPost, the Star Tribune, Minnesota Monthly, The Current (MPR), and the White on Rice podcast. In 2024 Quebracho Empanadas was recognized in the University of Minnesota Carlson School of Management’s MN Cup in the Food / Agriculture / Beverage division — the largest statewide startup competition in the country.
The operational side of the business — producing consistent product, predicting weekly demand, managing perishable inventory, scaling a wholesale channel without wrecking margins — turned into a surprising source of data lessons. Running a kitchen at this scale means making dozens of small probabilistic decisions every week: how much chicken filling do we prep? how many empanadas will the coffee-shop account buy this weekend? at what point do we raise wholesale prices?
Every one of those is a forecasting problem. And most small-business operators answer them with intuition shaped by whatever happened last week. That intuition is surprisingly accurate for simple cases and surprisingly wrong in exactly the places that hurt margins.
The journal
This site’s Journal is the write-up of those lessons. The posts are written for other small-business operators who don’t have a data team and don’t want one. The math is kept simple. The tools are a spreadsheet, a pencil, and occasionally a Python script someone posts from a coffee shop.
Topics include:
- Demand forecasting when you have 12 weeks of sales data, not 12 years
- Inventory as a prediction problem with an explicit service-level target
- Why point estimates lie to small-business owners and what to report instead
- Pricing experiments that actually generate information
- The difference between operational data and reporting data — and why confusing the two costs time
Contact & press
For press inquiries about the original Quebracho Empanadas story, or for permission to reference the archive, reach out via the contact channel on the publishing company’s site. For the data-journal content, permission is granted to quote or link freely with attribution.