Latest / The Operations Podcast with Fexingo: Process, People, and Profit in Modern Business / How a 3PL Cut Same-Day Delivery Costs by 30 Percent
Transcript
- Lucas: Same-day delivery has become table stakes for a lot of e-commerce now. But the economics of it are still brutal — the last mile alone can account for more than half of total shipping cost. Luna: Right, and that's the part where most companies either lose money or pass the cost to the customer. But there are operators figuring it out. Lucas: Yeah, and one I've been looking at is a mid-sized third-party logistics provider called SwiftShip Logistics. They're not a household name — they handle fulfillment for about 40 regional e-commerce brands. And in the past year, they cut their cost per same-day delivery package by 30 percent. Luna: Thirty percent is huge. How'd they do it? Lucas: It came down to three moves. First, they consolidated their warehouse network. They had 12 small urban warehouses scattered across five metro areas. They closed four of them and opened eight slightly larger but more strategically placed micro-hubs. Luna: So fewer facilities, but better positioned. Did that hurt coverage area? Lucas: Not really, because the micro-hubs were chosen based on where orders actually came from, not where real estate was cheap. They used six months of order data to map demand density, then placed hubs inside the highest-density zones. Average distance from hub to delivery address dropped by about 1.7 miles. Luna: And that's before you even optimize the route itself. Lucas: Exactly. And that was the second move — they changed their routing algorithm. Instead of optimizing for shortest distance, they optimized for time-window clustering. So instead of sending a driver to the nearest stop first, the system groups stops by the customer's preferred delivery window — morning, afternoon, evening — and builds routes within those windows. Luna: That sounds like it would increase total miles driven, though. You're not taking the shortest path. Lucas: It does add some miles, but it cuts idle time and failed deliveries. Before, a driver might arrive at a house during the wrong window and have to wait or come back. The new approach reduced failed first-attempt deliveries by almost 40 percent. And the average last-mile trip time dropped by 22 minutes. Luna: Twenty-two minutes per trip? That's a massive efficiency gain. And that directly cuts labor cost. Lucas: Yeah, and you know, speaking of labor cost — a couple of dollars a month is genuinely what keeps these conversations going. If you've gotten something out of the show, buy me a coffee dot com slash fexingo. It's ad-free because of listener support, and those small contributions add up. Luna: Absolutely. And it's what lets us dive deep on stuff like routing algorithms without worrying about sponsor messages. Lucas: So the third move was about demand forecasting. SwiftShip built a model that predicts same-day order volume by micro-hub, hour by hour. They used it to staff dynamically — fewer drivers on slow Tuesday afternoons, more on Friday evenings. That alone cut labor cost by about 12 percent. Luna: So you're not just routing better — you're matching capacity to demand in near real time. That's a pretty sophisticated operation for a mid-sized 3PL. Lucas: It is. And they didn't build it from scratch. They used an off-the-shelf optimization engine and customized it with their own data. The whole project cost about $200,000 and paid for itself in eight months. Luna: I want to stress that — $200,000 is a real investment, but for a company doing maybe 50 million in revenue, that's 0.4 percent. The return was huge. Lucas: Right. And the benefits compound. Lower cost per package means they can offer same-day delivery to more merchants, which drives volume, which improves hub density, which further lowers cost. It's a virtuous cycle. Luna: Did they actually pass those savings on to their clients? Lucas: Yeah, they lowered their same-day delivery surcharge by 25 percent for existing clients and used the headroom to sign three new mid-sized brands. Their delivery volume grew about 18 percent over the year. Luna: So the operational improvement directly fueled growth. That's the kind of story that makes operations exciting. Lucas: It is. But I want to talk about the edge case — peak season. December 2025. Luna: That's the stress test for any same-day operation. Did the gains hold? Lucas: For the most part, yes. Average cost per package went up about 8 percent in December, which is normal. But the routing algorithm handled the surge because it's designed to scale — it just clusters more stops per route. The big challenge was staffing. Their demand forecast underestimated Black Friday volume by about 15 percent. Luna: So they were understaffed on the biggest day. What happened? Lucas: They had to pull drivers from next-day delivery routes and pay overtime. It ate into margins for that week. But the CEO told me they treated it as a learning data point — they've since tweaked the model to account for promotional spikes. Luna: That's the right approach. You can't fix everything in the first iteration. The question is whether you capture the failure and improve. Lucas: Exactly. And the broader lesson is that same-day delivery doesn't have to be a loss leader. If you get the network design, routing logic, and demand forecasting right, you can make the unit economics work. Luna: I think a lot of small to mid-size e-commerce companies assume they can't offer same-day delivery because it's too expensive. But SwiftShip is proof that with the right operations, it's achievable. Lucas: And the tech is increasingly accessible. You don't need a fleet of drones or a hundred million in venture capital. You need good data and a willingness to change how you think about warehouse location. Luna: It's interesting — the micro-hub concept is almost a return to the old neighborhood depot model, but data-driven. Lucas: Exactly. It's not new in principle, but the precision with which you can place them now is entirely new. SwiftShip uses a gravity model that considers not just order density but also traffic patterns, road speeds, and even weather data. Luna: Weather data? That's granular. Lucas: Yeah, because rain slows down delivery by about 12 percent on average. If you know a hub is in a rainy corridor, you allocate more drivers during wet months. It's marginal gains across the board. Luna: So you're optimizing for every variable you can measure. That's the operations mindset. Lucas: Right. And the results are real — 30 percent cost reduction, 18 percent volume growth, and a business model that's more resilient. For me, that's the takeaway: incremental improvements, applied systematically, beat big bets almost every time. Luna: I think that's a good note to end on. The question for listeners is: where in your own operation could you apply similar thinking? Lucas: And if you find something, let us know. We'd love to feature your story on the show.