Latest / The Operations Podcast with Fexingo: Process, People, and Profit in Modern Business / How Domino's Turned Waste Into Profit
Transcript
- Lucas: Most people think of pizza waste as a nuisance. You order too much, you eat leftovers, or worse, you throw half a pie in the trash. But for a company the size of Domino’s, that trash bin is actually a leak in the hull. And for the last few years, they have been plugging that hole with something far more aggressive than better training. They are using predictive algorithms to stop the pizza from being made in the first place. Luna: So they are predicting hunger? That sounds like science fiction until you look at the numbers on their latest quarterly reports. Lucas: Exactly. It is not magic. It is a massive integration of point of sale data, local weather patterns, and even traffic flow data near specific store locations. The system doesn't just tell the manager what sold yesterday. It tells them what will sell tomorrow at two in the afternoon, down to the specific topping combination. The result is a twenty-two percent reduction in unsold inventory across their top ten metropolitan markets this quarter alone. Luna: Twenty-two percent is huge. But does that actually change the culture in the stores, or do the managers just ignore the screen? Lucas: That is the real story here. In the beginning, the head chefs pushed back hard. They relied on gut feeling. They knew their neighborhood. If it rained on Tuesday, they assumed slow sales. But the data showed that rainy Tuesdays often meant a spike in large group orders because people were stuck inside. The system forced a cultural shift where intuition had to submit to evidence. Luna: I bet that was a rough transition. How did they get buy-in without alienating the people who actually make the food? Lucas: They gave the managers ownership of the variance. Instead of corporate dictating the order, the algorithm provided a baseline, and the manager could adjust it based on local events. A high school football game, a local festival, a road closure. The tool became a dashboard, not a dictator. And when the waste dropped, the bonus structure followed. Suddenly, saving five hundred dollars a week in ingredients felt like winning. Luna: It is interesting that they tied it directly to compensation. Does that work in other industries, or is it specific to fast food perishables? Lucas: The principle works anywhere there is perishability. But the speed of implementation is what separates Domino’s from everyone else. They rolled this out in eighteen months. Most legacy retailers take three years just to agree on a vendor. Here is the thing though, we talk a lot about operations on this show, but I want to pause for a second. If these conversations have moved your work forward in some small way, remember that a couple of dollars a month is genuinely what keeps these going. Check out buy me a coffee dot com slash fexingo if you've gotten something out of them. It really makes a difference. Luna: Thanks for the support note. It definitely helps us keep digging into these case studies without hitting a paywall. Lucas: Back to the pizza. Once they stabilized the waste, they found a secondary benefit. Consistency. When you order less, you rarely run out of popular items. Stockouts are the silent killer of customer loyalty. By aligning production with predicted demand, they kept their best sellers in stock ninety-eight percent of the time during peak hours. Luna: That is a direct lift to revenue, right? Because if someone wants a pepperoni pie and it is gone, they don't wait. They go next door. Lucas: Precisely. And that neighbor is likely another Domino’s franchisee or a competitor. But the data also revealed something surprising about the late-night crowd. Between midnight and four AM, the traditional logic said: shut down the ovens, save energy, minimize labor. The algorithm said: no, order volume spikes for specific toppings during those hours. So they kept certain lines open, but only for high-margin, low-complexity items. Luna: So they optimized the menu dynamically based on time of day? That is pretty sophisticated for a drive-thru model. Lucas: It is. They call it dynamic menu engineering. If the system predicts a lull, it removes complex customizations from the digital kiosk. Why let a customer spend three minutes configuring a triple-meat specialty pie if the staff can only handle one every fifteen minutes? It protects the throughput rate. Throughput is king in operations. Luna: I love that. It prevents the bottleneck before it happens. But does this require a total overhaul of the kitchen hardware, or is it mostly software? Lucas: Surprisingly light on hardware. Most stores already had digital ordering terminals. The upgrade was purely in the backend logic and the integration with the supply chain API. The flour suppliers get an alert an hour before the pizza goes out. The cheese delivery trucks are rerouted in real-time. It is a closed loop. No new ovens needed, just smarter brains. Luna: It feels like the old way of doing things was so reactive. You made the pizza, then you saw if it sold. Now it is proactive. Lucas: Reactive operations are expensive. Every minute you spend guessing is money burned in electricity, labor, and raw materials. The proactive model turns uncertainty into a line item. You can budget for a two-percent error margin. You cannot budget for chaos. Luna: Do you think smaller chains can pull this off, or is this strictly a play for the giants with the IT budgets? Lucas: SaaS solutions are making it accessible. There are now white-label forecasting tools available for independent retailers. They might not have the scale to negotiate better cheese prices, but they can still reduce waste. And in a tight margin business, saving ten percent on waste is the same as increasing prices by ten percent. Luna: That is a compelling comparison. Increasing prices is painful for customers. Cutting waste is invisible. Nobody notices when the trash gets lighter. Lucas: Right. And the environmental angle is becoming a marketing asset too. Consumers are noticing. Domino’s publishes its sustainability metrics prominently now. Reducing food waste isn't just good for the P&L, it is good for the brand narrative. It signals responsibility without needing a giant billboard campaign. Luna: I would never guess that my pizza order is being influenced by traffic data. It feels seamless. That must be the ultimate goal of this tech. Lucas: Seamlessness is the goal. When the tech works, you don't see it. You just get your food faster, it is fresher, and the price hasn't inflated to cover inefficiencies. It is a quiet victory for the consumer. Luna: So, looking ahead, do you see this expanding beyond food? Like, could a hardware store use this for seasonal tools? Lucas: Absolutely. Home Depot already uses similar logic for shovels in winter. But the velocity of pizza is different. The window for decision-making is minutes, not weeks. That hyper-local, real-time adjustment is where the next wave of innovation is heading. Retailers who can react to the weather in real-time will leave the slow movers behind. Luna: It seems like the barrier to entry is lower than ever, but the execution risk is higher because you are trusting a black box. Lucas: Trust is earned through transparency. The best systems show you the why. If the algorithm says order extra breadsticks, it should say 'because local sports event ending at eight PM.' That context builds trust with the manager. Without it, you just have a robot telling you what to do. Luna: Context is everything. Okay, so if you are running a small shop, what is the first step? Do you start with the data collection? Lucas: Start with your POS export. Most small businesses already have the data sitting in a CSV file they never open. Clean it up. Look for patterns. You might find that your Saturday sales are driven by a single product. Focus there first. Don't try to boil the ocean. Luna: Practical advice. Start small, prove the value, then scale. It is the same playbook as the big guys. Lucas: Always. Technology doesn't fix bad habits. It amplifies good ones. If your portion sizes are inconsistent, an algorithm won't save you. But if your process is solid, data can make it bulletproof. Luna: I like that. Bulletproof processes. It sounds like the holy grail for anyone trying to scale without breaking. Lucas: It is. And once you have that stability, you can finally breathe. You aren't fighting fires anymore. You are just managing the fire extinguishers on the shelf, knowing exactly when you might need one. Luna: A good metaphor. So, the takeaway is: stop guessing, start measuring, and let the data guide the prep list. Lucas: Exactly. Your intuition is valuable, but your history is undeniable. Combine them, and you have a competitive advantage that lasts. Luna: We will see if I can convince my own cafe to try this. Though I suspect my baristas will miss the art of the pour-over regardless of what the algorithm says. Lucas: Let them have their art. The algorithm can handle the milk steaming schedule. Find the balance between craft and efficiency. That is where the profit lives.