Not everything worth optimizing is code.
I've been helping a small family food business in Scarborough think through their catering operations, and the delivery-versus-pickup decision turned out to be a cleaner tradeoff problem than half the feature flags I've shipped.
Pickup minimizes variables. No delivery window to miss, no driver, no traffic. The cost is pushed onto the customer, and you lose the people who won't drive.
Delivery removes friction for the customer but adds a stack of failure modes: timing, temperature, a driver, an address, someone on site to receive it. Every one is a thing that can go wrong on the day.
The interesting part is that the right answer isn't a blanket policy — it depends on order size and occasion. Small weekday orders lean pickup; big catered events lean delivery. They wrote up the full decision framework here if you want the non-engineer version: halal catering delivery vs pickup in Toronto.
The takeaway that stuck with me: "reduce the number of things that can fail on the day" is just as true for a food order as it is for a deploy.
Top comments (0)