2 ms·
I don't think anybody has figured out grocery delivery, but I think a lot of the obvious hazards are at least mapped by now. The first and most obvious thing,
by philwelch 8y ago
I don't think anybody has figured out grocery delivery, but I think a lot of the obvious hazards are at least mapped by now.
The first and most obvious thing, which Amazon is at least superficially best-equipped to handle, is the fact that existing grocery stores make for really shitty fulfillment centers. Grocery delivery services that literally send someone to the grocery store for you end up with probably even worse results than you would get just by hiring a part-time personal assistant and telling them to get you groceries. If you did that, at least you'd cut out the middleman, and the person would have better incentives to not fuck up your order in particular.
You can lay out a fulfillment center so that products can be picked and orders assembled efficiently and unambiguously. This doesn't match how a grocery store is laid out. A grocery store wants the shopper to be able to predict where to find a product, or at least a product category. Every single product is stocked directly next to every other product it possibly be confused for, usually all smushed up together or even hidden behind each other. Also, random members of the public walk through the shelves and throw them into chaos. Even the flow of going through the grocery store is designed for marketing return rather than efficiency.
The bigger part is probably that groceries are already a pretty terrible business. Whole foods in general--produce, meat, even stuff like eggs--are often already loss-leaders, even if you're selling them in person to a customer who can inspect the product. Grocery stores use these to lure customers and then sell them much more profitable packaged goods, and it works out because competition for retail space is fierce and demand for low-margin groceries can be a competitive advantage.
Remove that constraint and the grocery business no longer makes sense because you can cherry-pick the best parts of the business and build a focused delivery business. Packaged pantry goods: Amazon. Refrigerated or frozen packaged foods on demand: GoPuff. Even if you want to deliver fresh whole foods, you can just restrict the number of unique ingredients, hire a chef to turn the Cartesian product of those ingredients into a list of decent recipes, sell packaged recipes, and that's HelloFresh.
Bit by bit, other business models are going to eat away at everything that isn't a loss-leader, and you're not left with any viable business in delivering whatever is left. Even the cash-rich-time-poor segment is better served by some variant of the HelloFresh model, assuming they're too time-poor to shop for groceries but not too time-poor to cook (which is kind of a weird assumption, although perhaps a pre-prepared "just throw into a crock pot" meal delivery service would be viable).