Latest / Small Business Success with Fexingo: Local Companies, Main Street Stories, and Community Commerce / How a Local Pizza Shop Used Data to Predict Customer Orders
Transcript
- Lucas: So there's a pizza shop in Portland, Maine — Rosario's Pizzeria — and they figured out that if it's raining on a Thursday, they're going to sell about forty percent more pepperoni pizzas than on a sunny Thursday. Luna: Forty percent? That's a huge swing. How do they even track that? Lucas: They started with a simple spreadsheet. The owner, Maria Rosario, just logged her daily sales by slice and whole pie, and then later she added weather data from a free API. After about six months, she noticed patterns. Luna: I love that. And it's not just weather, right? I'm guessing day of the week, local events... Lucas: Exactly. She cross-referenced her sales with the local high school sports schedule, payday weekends, even the farmers market calendar. What she built was basically a demand-forecasting model, except she did it on Google Sheets with no coding. Luna: Look, I know we talk a lot on this show about the tactics behind these small businesses. And I think it's worth pulling back the curtain a little — this kind of reporting takes time, and we're able to keep the show ad-free because a small group of listeners chips in through buy me a coffee dot com slash fexingo. It's not a big ask, just that if these conversations have moved your work forward in some small way, that's what keeps us going. Lucas: Yeah, it really does. And it's humbling. But back to Rosario's — the data didn't just help her order smarter. It changed how she staffed her shifts. Luna: So she's using the same spreadsheet to schedule people? That's genius. Lucas: Right. She noticed that rainy Fridays meant more delivery orders, so she'd bring on an extra driver. And on sunny Saturdays in summer, foot traffic was higher, so she'd have an extra person on the counter. By the end of the first year, she told me food waste dropped twenty-two percent. Luna: Wow. And I'm guessing her margins improved a lot too, since pizza ingredients are perishable. Lucas: Yes, her food cost percentage went from thirty-two percent down to twenty-seven. That's a direct profit improvement. And revenue went up eighteen percent year-over-year, because she was actually stocking enough to meet demand without guessing. Luna: I want to talk about the tools. She used Google Sheets and a free weather API — those are accessible to anyone. But a lot of small business owners I talk to are intimidated by data. They think they need a data scientist. Lucas: That's the biggest misconception. Maria started with literally a paper notebook. She'd write down 'Thursday, sunny, 75 degrees, 12 pepperoni slices sold.' After a few weeks, she typed it into a spreadsheet. The hard part isn't the tool — it's the discipline to log it consistently. Luna: And what about the weather API? Did she have to pay for that? Lucas: No, there are free tiers. She used OpenWeatherMap's free plan, which gives you current weather and forecasts. She'd just pull the high temp and precipitation for each day, paste it into her sheet. Total investment: zero dollars. Luna: Okay, but I'm a skeptic. Is this replicable for, say, a hardware store or a bakery? A pizza shop has a limited menu and predictable demand around events. Lucas: Great question. Maria's approach actually works better for businesses with more SKUs. A hardware store could log sales of, say, paint by color and cross-reference with new housing permits in the county. A bakery could track sales of croissants versus bagels by day of week and local running club events. Luna: So the principle is: find one external variable that correlates with your sales, start tracking it, and adjust your ordering and staffing accordingly. Lucas: Exactly. And here's the thing — Maria didn't overcomplicate it. She didn't build a dashboard or buy software. She just looked at the correlation coefficient between rain and pepperoni sales, and it was about 0.6. Strong enough to act on. Luna: Wait, she calculated correlation coefficients in Google Sheets? That's more advanced than most small business owners would do. Lucas: She did, but she told me she learned it from a YouTube video. It's a built-in function — CORREL. You highlight two columns of numbers and it spits out a number between negative one and one. Anyone can do it in two minutes. Luna: Okay, that's actually empowering. So the barrier is not technical skill — it's just deciding to start tracking. Lucas: Right. And Maria had one more insight I thought was brilliant. She said the data helped her with her suppliers, too. She could go to her distributor and say, 'I'm going to need thirty percent more mozzarella in October because of high school football games.' The distributor gave her a small discount for the reliability. Luna: That's leverage. So data gave her bargaining power. I love that. Lucas: And it's not just about ordering. She also used the data to test new menu items. Before adding a gluten-free crust, she looked at how many customers asked for it over three months. The data said about eight requests per week. She launched it as a limited trial and within two months it was ten percent of sales. Luna: So she's de-risking decisions. Instead of guessing, she's using a small sample to validate. Lucas: Yes. And she told me the biggest mistake small businesses make is trying to predict everything at once. She started with just one question: 'What drives pepperoni sales?' Once she answered that, she added more variables slowly. Luna: That's such a good lesson. Start narrow. What's one thing a listener could do tomorrow to start this process? Lucas: Pick your bestselling product or service. For a coffee shop, it's a latte. For a salon, it's a haircut. Then pick one external factor — weather, day of week, local event. Start a log in a notebook or a spreadsheet. Write down the factor and the sales number for that day. After thirty days, look for patterns. That's it. Luna: I love how low-tech this is. It's almost a throwback — like a captain's log for your business. Lucas: Exactly. And if you want to take it further, Maria's advice is to share your findings with one employee. Get them involved. She said her delivery drivers started telling her, 'Hey, it's raining, we should prep more dough.' They became part of the system. Luna: That's culture-building through data. It's not just about numbers — it's about everyone understanding the business better. Lucas: Right. So the takeaway is: you don't need a million-dollar analytics platform. You need a spreadsheet, a weather app, and the discipline to log a few numbers each day. Rosario's Pizzeria did it with pepperoni and rain, and it changed her business. Luna: I'm going to try this with my local bakery. I bet they'd be surprised by what they find. Lucas: And if you try it and find something interesting, let us know. We'd love to hear how it goes.