> Could economies of scale drive the cost below $5/meal? $2?
Even if you have minimum wage employees, you're still paying people $15/hour to make the food, and $15/hour to deliver it. How many servings can a cook make in an hour? How many deliveries can a driver do in an hour?
Ten deliveries an hour gets you down to $1.50 per delivery. In some areas and times of day where deliveries line up just right, they can maybe do more, but if it's hot food, there's a limit as to how much the driver can carry at once. If it's all point-to-point back and forth between the kitchen and destinations, forget it.
Let's say the cook can make 30 servings in an hour. So that's $2 just to pay for labor already. That doesn't factor in costs for ingredients, kitchen space, cooking equipment, and maintenance for that. It doesn't account for marketing. It doesn't account for the R&D required to build and maintain the ordering app/website, or for server costs.
Speaking of space requirements, this is actually really problematic from a cost perspective. You need the kitchen to be within a reasonable delivery time from your customers. For a dense city, you're probably good for volume of orders, but you'll probably need to locate your kitchen inside the city limits, where real estate can be expensive. For a suburb, no matter where you put your kitchen you run the risk of not having enough delivery orders within a reasonable distance.
A company with other revenue could certainly sell this as a sort of loss leader, but I can't imagine a way to do this for $2 per meal, and probably not even $5 per meal. Hell, if I walk to my local grocery and buy ingredients for a basic meal, I'd be hard pressed to walk out of there spending less than $5. Economies of scale could reduce the cost of materials, but the labor and infra costs would surpass the savings.