AI in Automation, Ecommerce, and Delivery
Most grocery apps fail for a boring reason. Nobody agreed on the one number it should move.
Margins are thin and shoppers bounce fast. Yesterday, when Walmart's mobile app outage blocked checkouts for thousands, it was a reminder that the app is part of the register now.
At a regional grocery chain, the stated goal was “launch a new mobile app.” The real goal was fewer pickup substitutions and fewer angry calls to the store.
We picked two metrics: pickup order accuracy and time-to-fix a failed payment. Then we built the system around them - inventory counts, substitution rules, staff prompts on handhelds, and a clear path from store manager to IT.
Result: substitutions dropped 17%, store calls fell 28%, and each store got back about 5 hours a week. The screens changed last. The store process changed first.
Now everyone is chasing new tools - Walmart with OpenAI and ChatGPT, Instacart rolling out retailer AI, Ocado Group reshaping automation deals with Kroger, Amazon pushing faster delivery. None of it works when the goal is vague.
This makes me wonder how many “app rebuilds” are really missing systems with nicer screens.