Max HenryFounder & Executive Director, Global Supply Chain Council (GSCC) · Founder & Executive Director at Global Supply Chain Council (GSCC)

Last Mile Delivery Failure: Lessons from a Unique Career Case #shorts

14 Aug 2026 · 2 min

Watch on YouTube

2:21

In this first‑person monologue, the speaker recounts a failed experiment at Flipkart where the company attempted to increase vehicle utilization by combining two very different last‑mile delivery verticals: large‑item deliveries (air conditioners, electronics, furniture) and grocery deliveries. Flipkart operated dozens of hubs across India, each with 18‑20 vehicles, but about half of the fleet was under‑utilized due to low density in many areas. The grocery team, which had higher order density, proposed merging its orders with the large‑item fleet to improve efficiency. Using route‑optimization tools such as Locus, the merged operation was launched. However, the experiment collapsed because the two product streams had vastly different customer profiles, order values (₹100,000 vs. ₹300‑₹600), and service expectations (installation and longer delivery windows for large items versus rapid grocery delivery). High‑value AC customers experienced delays and unmet expectations, while grocery timelines suffered. The speaker concludes that while combining non‑uniform supply chains may appear to boost utilization, it introduces upstream trade‑offs that can degrade the overall last‑mile experience. Companies handling mixed product profiles must carefully assess cost benefits against potential service degradation.

Chapters

This is a unique case which I faced in my career, okay? So I was looking for a large last mile delivery. So large is basically, uh, the ACs and the electronics and furniture delivery, okay, in Flipkart. And that point of time, and we used to run a lot of vehicles in each city. So every day morning, maybe one single hub, and we had around 40-45 hubs across India, and one single hub used to have 18-20 vehicles which used to run. And we found 50% of vehicles are actually not utilizing because we don't have density everywhere. So then the grocery team, which was a very different vertical, they came and they told we have a lot of density. Why not you merge grocery along with the furnitures and electronics? So we said it is a good, good proportion, right? We actually utilize the vehicles and then it will be more optimized. Then we tried it. We use some of the tool like Locus and other thing which give the route optimization points also. Now, um, you will not believe we failed immensely, okay? And the reason is that when we merge 2 supply chain for density benefit, the customer profile between both of them are very different. One product I am sending which is at a lakh rupee, okay, to a customer And another product which is grocery, maybe the ASP was around ₹600 or ₹500 or ₹300, right? So what was happening, it was killing the expectation of an AC customer who was very high-profile customer for us because maybe he's facing some delays or my delays was increasing or their expectation of waiting time at the doorstep because there is an installation and other thing which was killing my grocery timeline. So many cases that has shown last-mile delivery optimization or speed may be inefficient because we also need a competency of a uniformity of the product and customer profiling. So it is very easy to say that we can utilize a non-uniform supply chain, but in practicality there are a lot of upstream impacts also which impact the last mile delivery. So in case any company is handling with 2 profiles of product, they need to be very careful to understand what is beneficial How much cost and what are the things they can lose also. It is just my personal experience I thought to share.