4 ms·
Their largely manual, layered-over-retail fulfillment model has inefficiencies that cause higher costs and stocking issues vs. the high-efficiency warehouse app
by octernion 9y ago
Their largely manual, layered-over-retail fulfillment model has inefficiencies that cause higher costs and stocking issues vs. the high-efficiency warehouse approach that Amazon has perfected.
[citation needed] you mean the "perfected" approach they are shutting down in 9 states? (https://www.recode.net/2017/11/3/16601488/amazon-fresh-cancellation-shut-down-grocery-delivery-whole-foods https://www.recode.net/2017/11/3/16601488/amazon-fresh-cance...)
noting this as an engineer on the team that does order fulfillment @ instacart and who actually knows our model...
- abalone 9y agoCitation? Ok, you asked for it... "Instacart is infuriatingly bad. Replacing your main entree meat item with nothing, no call/notification, negates all convenience it offers" "My 1st @Instacart experience will be my last. 26 of my 49 items have been replaced but mostly refunded so I still have to go to the store." "i once had them tell a store had no cheese. completely out of cheese. no cheese was available to replace. i was naturally pretty cheesed off" "@instacart Why should I continue to use you all if every time I order, a huge % of items are out of stock and replaced? Update your database" "Same happened to me! 14 items, and 8 ended up being replaced. I'm going to give them another chance, hopefully it was just a bad shopper." ...and more.[1] Clearly the lack of integration with realtime inventory systems is a weak point. But perhaps you could clarify how the model is supposed to work. [1] https://techcrunch.com/2017/10/01/instacart-needs-to-deeply-invest-in-grocers-and-user-personalization-to-survive/ https://techcrunch.com/2017/10/01/instacart-needs-to-deeply-...
- octernion 9y agoI was noting our model in comparison to Amazon, which has it's own set of inefficiencies -- capital for warehouses far outside of city boundaries, the dedicated labor force/trucks, spoilage and selection issues, etc. In no way am I noting that our model is perfect or without inefficiencies, just that the idea that "Amazon has the perfect and most efficient grocery delivery system" and "to attempt to use local retail grocery stores as a collective warehouse is not profitable" is just a fallacy.
- abalone 9y agoI didn't say either of those things. Those are fake quotes. It is odd that you characterize warehouse real estate, labor and delivery fleet costs as "inefficiencies" in comparison to retail stores. Precisely the opposite is true. Retail stores have vastly higher costs per square foot, higher headcount, and are not at all optimized for delivery vehicle operations. I did not say Amazon has a perfect grocery solution in place today. They just closed the Whole Foods deal a couple months ago. What I am observing is that all these grocers calling up Instacart in a panic is not at all reflective of a strong long-term position. Instacart's unintegrated, layered-over-retail model carries all the overhead of retail and then some. Amazon is building a fully integrated supply chain, warehousing, delivery and retail system and so has a lot more operational efficiencies it can bring to bear.
- octernion 9y agoThose are not "fake quotes" -- they are directly in the parent I was replying to... I am not sure you can matter-of-fact state that "Precisely the opposite is true"; this is in fact, not true. I am not going to respond further as it's not entirely clear where you are getting any of your information ("layered-over-retail model carries all the overhead of retail and then some" as an example of a totally unsubstantiated and untrue statement).