Latest / Franchise Conversations with Fexingo: Buying, Running, and Scaling Franchise Businesses / Franchise Location Clustering How Density Drives Profit
Transcript
- Lucas: So you're a multi-unit franchisee with capital to open two more locations. Do you put them in two different cities to spread risk, or do you drop them both into the same neighborhood, five minutes apart, and let them compete with each other? Luna: Most people I talk to say spread risk. They don't want to cannibalize their own sales. Lucas: Right, and that's the intuitive answer. But there's a growing body of data that says the opposite — clustering locations close together actually yields higher per-unit profit than scattering them. We're talking about a strategy called location clustering, and it's one of the most counterintuitive findings in multi-unit franchising. Luna: I remember reading about this with Domino's in the early 2000s. They basically carpet-bombed neighborhoods with stores. Lucas: Exactly. Domino's was the pioneer. In 2003, they shifted from a strategy of one store per large territory to clustering stores within a two-mile radius in dense metro areas. The logic was simple: delivery speed is their competitive advantage. If you have a store every two miles, the driver can get the pizza to your door in under 30 minutes. That drives repeat orders, which drives unit volume. Luna: And that volume more than compensated for the cannibalization? Lucas: According to their franchisee disclosures, yes. A 2024 study from the University of Texas looked at 1200 quick-service franchise units across five brands. They found that clustered units — defined as locations within a three-mile radius of another same-brand store — generated 18 percent higher average unit volume than non-clustered units. More importantly, the profit margins were 22 percent higher because of shared overhead. Luna: Shared overhead meaning what exactly? A manager covering two stores? Bulk deliveries? Lucas: Both. When you have two stores within three miles, you can use the same regional manager, the same supply truck route, the same training schedule. You can cross-train staff between locations. One franchisee I spoke with in Chicago runs four Subway stores within a two-mile stretch. He told me his food cost percentage dropped by 3 points because he could buy in larger quantities for all four stores at once. That's a direct margin lift. Luna: Three points on food cost is huge in a sub shop. That's like an extra 10 percent net profit. Lucas: It is. And it compounds. The study also found that clustered stores had 14 percent lower labor costs as a percentage of sales because managers could float between locations during slow periods. Instead of having two fully staffed stores, you have one and a half, essentially. Luna: But there's a limit, right? Too many stores too close and you start eating your own lunch. Lucas: Absolutely. The study identified a tipping point. When stores were within a one-mile radius of each other, the cannibalization effect started to outweigh the operational benefits. Volume per store dropped by 9 percent, and margins flattened. So the sweet spot seems to be two to three miles apart. Luna: That's a narrow band. How do franchisees figure out where their stores land? Lucas: You have to model it. Most franchisors provide a territory map, but those maps are often drawn for the brand's benefit, not the franchisee's. They want to sell as many franchises as possible, so they might draw territories that are too small. The smart franchisee uses their own point of sale data — or if they're buying a resale, they ask for the previous owner's delivery records. Luna: That assumes you have the data. What if you're a new franchisee with no existing stores? Lucas: Then you use public data. Census block groups, household income, population density. One tool I like is Maptive — it's a simple GIS platform where you can overlay your planned locations and run a drive-time analysis. You want to see where your delivery radius overlaps. If the overlap is more than 30 percent of your trade area, you're probably too close. Luna: So clustering works best when the market is dense enough to support multiple stores without overlapping too much. That sounds like a metro-only strategy. Lucas: It is. You wouldn't cluster in a rural county. But in a city like Houston or Atlanta, it makes perfect sense. Subway has 12 stores in a six-mile stretch of Houston's Westheimer Road. That's an extreme example, but those stores have been there for years and they're all profitable. Luna: I think the reason this works is that clustering builds brand density. When customers see three Subway signs on the same road, they think 'Subway is everywhere, I'll just go to Subway.' It's a mental shortcut. Lucas: That's exactly right. There's a behavioral economics term for it — the availability heuristic. If a brand is top of mind because you pass it twice on your commute, you're more likely to choose it. Domino's understood this intuitively. They didn't just want to be convenient; they wanted to be unavoidable. Luna: And that's why they beat Pizza Hut in delivery market share in the mid-2000s. Pizza Hut had bigger stores with dine-in, but Domino's had tiny stores everywhere. Lucas: Precisely. Domino's average unit volume was actually lower than Pizza Hut's per store, but their system-wide sales grew faster because they had more stores in better locations. The franchisees who clustered made more money per store than the ones who didn't. Luna: Let's talk about the flip side. What about brands that tried clustering and failed? Lucas: McDonald's tried clustering in the 1990s with their 'satellite' stores. They opened small outlets near existing full-size locations, thinking they'd capture overflow. But the satellites ended up cannibalizing 40 percent of the parent store's sales, and they closed most of them within three years. The difference was that McDonald's stores are much larger and have higher fixed costs. You can't just drop a small store next to a big one and expect the economics to work. Luna: So clustering works better for brands with smaller footprints and lower overhead. That makes sense. Lucas: And for brands with a strong delivery component. If you're a sit-down restaurant, clustering doesn't help because customers won't drive five minutes to the next location if the first one is full. But if you're delivery-based, the customer never sees the store — they just see the driver arrive quickly. Luna: What about the franchisor's perspective? Do they encourage clustering? Lucas: It varies. Some franchisors explicitly prohibit it in their franchise agreements. They'll say you can't open a second unit within a certain distance of an existing one. That's usually to protect existing franchisees. But other franchisors, especially younger ones trying to grow fast, will allow it because they want density in key markets. Luna: So the franchisee has to read the fine print. If the agreement says 'no store within five miles of another,' then clustering is off the table. Lucas: Right. But even if it's allowed, you need to do your own due diligence. Don't rely on the franchisor's projections. I've seen franchisor provided pro formas that assume zero cannibalization, which is unrealistic. A good rule of thumb is to assume a 10 to 15 percent cannibalization rate for a new store within three miles of an existing one, and then see if the combined volume still makes sense. Luna: Let's run a quick example. Say you have an existing store doing $500,000 in annual sales. You open a new store two miles away. If it cannibalizes 15 percent of the old store's sales, that's $75,000 lost. But the new store might do $400,000 in its first year. So combined you're at $825,000, compared to $500,000 before. That's a win. Lucas: Exactly. And if your fixed costs per store drop because you can share management and supplies, your margin on that $825,000 might be higher than the margin on the standalone $500,000. That's the clustering math. Luna: It's a different way of thinking about expansion. Instead of 'how many stores can I open?', it's 'how dense can I make my existing territory?' Lucas: And that's a fundamental shift. Most franchisees think in terms of number of units. But the real metric should be profit per square mile. If you can operate four stores in a three-mile radius and each does $400,000 with a 15 percent margin, that's $240,000 in profit. Compare that to four scattered stores that each do $350,000 with a 10 percent margin because you have higher logistics costs — that's only $140,000. Luna: So the clustered franchisee makes $100,000 more a year from the same number of stores. Lucas: And they have less travel time, less management headache, and more control over their market. It's a compelling argument for clustering in the right conditions. Luna: I want to ask about something that came up earlier — the shared overhead. You mentioned cross-training staff. How does that work in practice? Lucas: It's a big deal. One franchisee I interviewed runs three Dunkin' locations within a five-mile radius. She told me she has a 'floating crew' of five employees who work at whichever store is busiest on a given day. That keeps her labor costs at 28 percent of sales, compared to the industry average of 32 percent. That four-point difference is huge. Luna: And that only works because the stores are close enough that employees can walk or take a short bus ride between them. Lucas: Exactly. If the stores were 20 miles apart, the floating crew would waste an hour commuting. The proximity is key. Luna: So if someone is looking at a franchise opportunity and they're considering multi-unit, what's your advice on clustering? Lucas: First, check the franchise agreement for territory restrictions. Second, do a GIS analysis of the market to see where your potential stores would overlap. Third, model the cannibalization conservatively. And fourth, talk to existing franchisees who have clustered. Ask them for their actual numbers. Luna: And if the franchisor won't share contact info for franchisees in a cluster, that's a red flag. Lucas: Big red flag. A confident franchisor will let you talk to anyone. If they're hiding something, assume the worst. Luna: You know, Lucas, conversations like this are exactly why listeners tell me they appreciate this show. We dig into real strategy, not just surface-level advice. Lucas: I'm glad it resonates. And if these conversations have moved your work forward in some small way, listener support is what keeps this show ad-free and independent. A lot of you have asked how to help — the simplest way is to buy me a coffee at buy me a coffee dot com slash fexingo. That's all lowercase, no spaces. It goes directly to keeping the podcast going. Luna: Yeah, it's a small gesture that makes a big difference for us. No pressure, just if you find value here. Lucas: Exactly. Now, back to clustering — one more nuance I want to mention. There's a difference between clustering within a single franchise system and clustering across multiple brands you own. Some franchisees own units from different brands in the same area, like a Subway and a Taco Bell. That can also generate overhead savings, but the risk is different because you're not competing with yourself directly. Luna: But you might compete for the same customer. If someone wants fast food, they might choose between your Subway and your Taco Bell. Lucas: True, but that's a portfolio diversification play. If one brand has a bad quarter, the other might pick up. The clustering logic still applies — shared management, bulk purchasing — but the cannibalization risk is lower because the brands are different. Some franchisees build entire empires this way. Luna: So the bottom line: clustering is a powerful tool, but it's not for every brand or every market. The math has to work. Lucas: And the data supports it when done right. The University of Texas study I mentioned earlier — they also found that franchisees who used a clustering strategy had a 30 percent lower failure rate over five years compared to those who scattered their units. That's not just about profit; it's about survival. Luna: That's a compelling stat to end on. Thanks, Lucas. Lucas: Thanks, Luna. And to our listeners — if you're thinking about your next location, maybe don't go too far. Stay close.