Latest / The Operations Podcast with Fexingo: Process, People, and Profit in Modern Business / How a Bakery Cut Delivery Delays in Half
Transcript
- Lucas: So, there's a regional bakery chain in the Midwest called Bakehouse Collective. They run about forty stores, and their delivery network was a mess. Twenty-three percent of their deliveries were late, and customers were starting to notice. But here's the thing — when they dug in, the root cause wasn't what anyone expected. Luna: What was it, then? Traffic? Weather? I feel like those are the usual suspects. Lucas: Nope. It was order batching and a static delivery schedule. They had a fixed schedule that didn't change with demand, and orders were batched in a way that forced drivers to zigzag across town. So they were doing a lot of miles, burning fuel, and still missing windows. Luna: So it wasn't the drivers — it was the plan. What did they do about it? Lucas: First, they did a proper root cause analysis. They mapped every delivery for two weeks, tagged each one with the reason it was late — traffic, loading delays, routing, whatever. And they found that forty percent of the late deliveries were because the route didn't match the order volume. They were sending three trucks to the same neighborhood when two would have done the job. Luna: That sounds like a classic overcapacity problem. But how do you fix that without a fancy algorithm? Lucas: That's the interesting part. They didn't buy any new software. They used a simple spreadsheet and a map. They started grouping orders by zip code and time window, then batch-assigned them to drivers at the start of the day. They also added a rule: no batch smaller than five stops, unless it's a rural route. Luna: Okay, that's smart. But what about the drivers? Didn't they resist the changes? Lucas: Initially, yeah. Some drivers liked having a set route because it was predictable. But once they saw they were doing fewer miles and getting back earlier, they came around. They also cross-trained the drivers so any of them could load the truck or handle the paperwork — that cut the loading time at the bakery by fifteen minutes per truck. Luna: Fifteen minutes a truck — that adds up. Over a fleet of, what, twenty trucks? Lucas: About twenty, and they do two shifts, so it's forty runs a day. Five hours of saved loading time daily. That's like an extra half a truck per day, just from cross-training. Luna: So they cut the delivery delay rate from twenty-three percent to eleven percent in six months. That's a huge operational win. But what was the cost side? Lucas: They saved on fuel and overtime. The fuel bill dropped twelve percent because they were driving fewer miles. Overtime dropped too, because drivers weren't stuck in traffic at the end of their routes. And customer complaints about late deliveries went down by half, which meant fewer refunds and discounts. Luna: That's a nice bundled benefit. But I want to dig into the root cause analysis a bit more. How did they actually categorize the delays? Was there a framework? Lucas: They used a simple category tree: internal vs. external. Internal was loading, routing, order picking. External was traffic, weather, customer access. And then they broke it down further. What surprised them was that external factors — traffic, weather — only accounted for fifteen percent of the delays. The rest was internal. That's when they knew they had a process problem, not an environment problem. Luna: That's a powerful moment — when you realize the problem is inside the four walls, so to speak. Did they share that data with the drivers? Or was it kept at the management level? Lucas: They shared it. They had a weekly huddle where they'd review the top three causes of late deliveries and brainstorm fixes. The drivers were the ones who suggested the batching rule — they knew that the morning route was always heavy in one suburb, so they said, why not combine that with the next neighborhood's lighter load. That kind of frontline input was gold. Luna: It's classic: the people doing the work have the answers, you just have to ask. But then how did they keep the momentum? Six months is one thing, but did it stick? Lucas: They made it into a habit. Every Monday, they review the previous week's on-time percentage — they call it the OTP, on-time performance — and if it slips below ninety percent, they do a mini root cause session. It's baked into their weekly rhythm now. That's what keeps it from backsliding. Luna: So it's not a one-time project, it's a continuous loop. That's the secret sauce, right? The discipline of reviewing and adjusting. Lucas: Exactly. And they didn't stop at delivery. They applied the same framing to their production schedule — they started baking in waves that matched the order forecasts, which cut the idle time between batches. The whole plant became more responsive. Luna: It's amazing how one corner of the operation can ripple through the whole system. You fix the delivery routing, and suddenly you're improving baking schedules too. Lucas: Right. And it's worth noting that the total investment was basically zero — just time and a willingness to look at the data. That's the kind of operational win that doesn't require a big tech budget. Luna: And that's a message that resonates with a lot of folks listening, I bet. By the way, we should mention — we're able to keep this show free of ads because of listener support. If these kinds of operational breakdowns have helped you think differently about your own work, you can support us at buy me a coffee dot com slash fexingo. It's a small gesture that means a lot. Lucas: Yeah, we really appreciate anyone who chips in. It keeps us independent and lets us dig into stories like this one, where the fix isn't about buying something shiny, it's about thinking smarter. Now, back to Bakehouse — the other thing they did was change their delivery windows. Luna: Oh? I thought they kept the same windows to avoid upsetting customers. Lucas: They did for the big accounts — the grocery stores and coffee shops that have set receiving hours. But for smaller customers, they moved to narrower windows — like a two-hour window instead of a four-hour one. That actually improved perceived reliability, because even if a delivery arrived at the end of the window, it was within the promise. Customer satisfaction scores went up, not down. Luna: So it's a classic operations trade-off: you might lose a little flexibility, but you gain trust. And trust is harder to build than flexibility. Lucas: Exactly. And they also started communicating proactively — if a driver was running late, they'd send a text to the customer with the new ETA. That alone cut the 'where is my order' calls by a third. Luna: That's a nice touch. It's not just about being on time, it's about managing expectations. So what's the takeaway for a listener who runs a similar operation — maybe not a bakery, but a service business? Lucas: The biggest lesson is to separate the symptom from the cause. They thought they had a fleet problem — too few trucks, slow drivers. But the data showed it was a planning problem. The fix didn't cost a cent. So my advice would be: before you buy more capacity, look at how you're using what you already have. That's where the low-hanging fruit usually is. Luna: And the second lesson is to involve the frontline. The drivers knew the answer all along — they just needed someone to ask and listen. Lucas: Yeah. And I'd add a third: make the change a routine, not a project. They review the numbers every single week. That's what turned a one-time improvement into a sustained one. Luna: I love that. So if someone wants to dig deeper into root cause analysis, what's a good starting point? Lucas: Start with the simplest tool: the five whys. Ask why five times, and you'll usually get to a systemic cause, not just the immediate one. That's what they did — they kept asking why deliveries were late, and the fifth why was 'because we don't batch orders based on geography.' That's the kind of insight that changes everything. Luna: And it's a tool that costs nothing. You just need a quiet room and an honest team. Lucas: Exactly. And that's the beauty of operational improvements — they're not always about big tech or big budgets. Sometimes they're just about asking better questions. Which, I guess, is what we try to do here every week. Luna: Indeed. And on that note, we'll wrap up this one. Thanks for joining us — we'll see you next time with another story of how smart operations can move the needle.