Where to Grow Next: The Franchise Site Hunt gets a Tech Upgrade

Paper maps, pushpins, and “pin studies” were the industry norm with field teams flagging down strangers in strip mall parking lots to ask about lunch habits or childcare preferences.

In today’s franchising world, those tactics have been replaced by data layers and predictive algorithms designed to determine where to grow next.

Across industries, from pizza and trampoline parks to early education and disaster restoration, franchise brands are reinventing how they analyze and select new locations. 

“We tell people that you need the technology, and that’s going to narrow down what your option is,” says Robert Morris, vice president of franchise development at Indoor Active Brands, which owns Altitude Trampoline Park and The Pickle Pad, “but real estate is still tactile at the end of the day. You’re going to have to walk around that space.”

Michelle Ryman, vice president of real estate and strategic market planning at Papa Johns, echoes the shift from old-fashioned techniques to large amounts of consumer data. “Instead of 26 responses in a day,” she says, “I now have 2,600 data points.”

Pushpins to AI

When Ryman began her career, there were no apps, no live dashboards, and certainly no AI. Instead, there were paper maps on office walls, color-coded pushpins, and field teams with clipboards trying to flag down strangers outside restaurants.

“We called them pin studies,” Ryman says, describing the process of conducting in-person interviews.

Back then, if a company wanted to know if a new restaurant location made sense, they hired a market research firm to stand outside and ask passersby. The data was limited, the sample sizes were small, and the respondents were often skewed demographically.

“The only people who would stop were retirees,” she says. “You’d never get someone like me to pause mid-errand.”

Despite its obvious flaws, pin studies remained in use for decades. Even 10 years ago, Ryman’s team at Wendy’s still occasionally deployed them.

Ryman now operates in the era of real-time decision-making powered by tools like Kalibrate and Google AI. Papa Johns has access to information about customer origin, visit frequency, transaction details, and movement patterns passively collected through mobile devices and digital behavior.

“The only data that I’m missing is the people who don’t have their cell phone on,” she says.

Papa Johns recently completed a full rebuild of its predictive analytics platform, Kalibrate 2.0, in partnership with its data science vendor. The new system can layer competitor locations, employee counts, and other data points that are rendered on a digital map.

“We use it every day,” Ryman says. “Not just the real estate team: Operations, marketing, even pricing teams rely on it now.”

While Kalibrate helps Papa Johns look outward, a partnership with Google AI is helping to look inward. The company is building an internal dataset on all 3,500 North American restaurants, tracking sales history, manager performance, and more.

“Each department used to keep its own version of history,” Ryman says. “Now, we’re bringing it all together in one place.”

The human element hasn’t disappeared. Every proposed site still gets a set of eyes on it when real estate directors and operations partners physically visit. But there’s no retreat from the digital transformation. Decisions that once relied on intuition now lean heavily on data.

“The technology’s here,” she says. “The challenge now is building teams that know how to use it.”

The wall map with pins has been gone for years. Ryman remembers the map store she used to frequent during her early days as a real estate professional. She occasionally thinks about buying one for her kids.

“I actually don’t even know if they have a concept of what a paper map looks like,” she says. “I’m going to order them an Atlanta map and tell them to put it in their car and see what happens.”

Mapping the territory

Tim Courtney, vice president of franchise development at PuroClean since 2017, also remembers the old ways.

Courtney says the tools were rudimentary when he joined the company. “It wasn’t even web-based yet. They were a little behind the times when I got here,” he says. “Even managing leads, they had a big giant whiteboard and magnets of where each candidate was in the process,” he says. “I walked in and said, ‘Well, that’s got to go.’”

That setup has been replaced by a layered data strategy that reflects how much the business of franchising has evolved. Today, PuroClean uses a customizable mapping tool called GBBIS to build its territories by ZIP code.

Unlike sandwich shops or gyms, which can rely on traffic and visibility, restoration services like PuroClean require a different calculus. “We’re not selling sandwiches here,” Courtney says. “You can’t throw a commercial out there: ‘Hey, the next time you have a kitchen fire, call PuroClean.’ It’s not how this works.”

Instead, PuroClean builds territories by layering demographic data like population and median income with professional networks (insurance agents and adjusters, roofers, and plumbers).

“We want to see what of those other services and contractors and insurance people are also in those areas,” he says, “because that’s how the franchisee is going to execute their local marketing strategies.”

The company’s standard protected territory includes 100,000 people with an additional 150,000 in a “halo.” Each market is evaluated for reach and serviceability. Rural areas must be weighed carefully.

“As we start to go out into more rural areas, is a service area really serviceable by a franchise?” Courtney says.

PuroClean draws on multiple data sources, including AM Best, to assess how many insurance claims exist in a given market and whether current franchisees are already absorbing that demand. The system helps the company determine when to open new territories and where.

“We can look and see what my existing franchise owners are getting from claim work and how much opportunity is there from claims that PuroClean franchisees aren’t getting,” he says.

The company also ranks open territories based on claims data, population, and contractor networks. It uses those rankings to decide where to focus marketing dollars. “Feed success, starve failure,” Courtney says.

The technology is layered, but the mindset remains human. “In the end,” he says, “it’s the same old franchise development. It’s just got new lipstick on.”

The more you know…

For franchise brands like Altitude Trampoline Park and its parent company, Indoor Active Brands, finding the perfect location requires layers of data, tech tools, and a healthy dose of human instinct, Morris says.

“When most people think of franchising, they think of restaurants. And rightfully so,” Morris says. “We have to find unique things because we are a trampoline park and an entertainment building.”

Unlike fast food, entertainment brands aren’t impulse stops; they’re destinations. That changes the real estate equation.

“Most people plan their trip to a location,” Morris says. “You’ll drive by a Starbucks and go, ‘Well, I do need a mocha right now.’ We’re not necessarily a spontaneous purchase.”

To refine their site-selection approach, Morris and his team lean on geospatial and demographic tools like Placer.ai.

“Placer.ai is huge because it narrows down a neighborhood,” he said. “They’re looking at cell phone traffic. If someone goes into the shopping center, some people might go to Target and leave. That’s great. But three blocks down, a shopping center might have a Burlington, a Kohl’s, and a Ross, and everyone goes to all three stores. That might actually be better for me, even though it has less traffic, because the traffic’s more spread out.”

Demographics are crucial, especially for child-focused attractions. “If you find a place with 100,000 people and 25% kids, it could be a great demographic for us,” he says. “We’re not looking at it in the same way that a Starbucks looks for a location.”

Altitude also collects unique data from its existing locations. Everyone who comes into a park fills out a waiver, which includes a section for ZIP code. That’s valuable information in the right hands.

“We can literally reverse the math and find out what demographics all of our parks are pulling from,” he says

He adds that today’s mapping software allows the company to model realistic drive times, which is the true radius of a location’s reach.

“The majority of our guests are driving within 15 to 21 minutes to our parks,” he says. “Proximity to an interstate can be very valuable. We really look at the drive time.”

When Morris consults his system, a location could have access to 486,000 people, and when a landlord consults their system, it could show 482,000. People have different ways of accessing data, but it generally creates a range that allows Altitude Trampoline Park franchisees to move forward with confidence in the numbers.

“If you’re going to sign a 10-year lease based on data,” Morris says, “that data should have a value the moment after you signed your lease.”

Some landlords use Matterport, a mapping tool that provides a virtual look at the inside of the building, but it’s not in widespread use. “We could actually go into it and move attractions around inside the space to see what our park would look like,” Morris says. “I would say very few landlords do it, but for the ones that do, it’s a good asset for us.”

Massive amounts of data help to find potential locations, but human experience always comes into play before any final decision. Cost is always a factor, and Morris says some landlords understand trampoline parks while entertainment venues are new and untested for others. The condition of the building is also central to the decision. Someday, AI might be able to take all those elements into account, but for now, feedback from flesh-and-blood professionals is required.

“Technology allows us to find the right neighborhoods to go to, but there’s still the feel and the emotion,” he says. “Even with using the technology, there’s still the art of it.”

Smart, safe, specific

When Matt Zaia describes site selection for The Goddard School, he sounds like someone playing a long game.

“If you’re going to go build a Goddard school, it’s probably going to cost you $4 million to $8 million, so it matters,” he says, adding that there are different types of builds depending on the sites available. 

As senior vice president and chief development officer, Zaia knows how important it is to get the location right the first time. Childcare isn’t like sandwiches or fitness centers. With 15-year franchise agreements and premium tuition rates, every build has to deliver on accessibility, safety, and long-term viability. Zaia notes that Goddard often builds in markets with average household incomes that range from $100,000 to $150,000.

To navigate that complexity, Zaia and his team lean hard into data. Their go-to tool is SIMMS, a CBRE-backed market planning platform created by Forum Analytics.

“It uses census block information,” Zaia says, “and then what SIMMS does is it procures data, purchases outside data, and wraps it into one user interface so that it’s convenient for us to use.”

The platform combines basic demographics (household income, population density, education levels) with predictive modeling.

“What’s being created is what’s called a regression model forecast,” he says. The model takes known performance data from existing schools and looks for national matches. If a school in one town is thriving, the model looks for other places where the same conditions exist.

But algorithms don’t know everything.

“The machine says, go here, and then the human goes and validates whether or not that is where we want to go,” he says.

No matter how well drawn it may be, some things can’t be identified on a map. Any new location needs to be convenient for families to reach. Which side of the street gets morning traffic? Where are new schools and subdivisions coming? Does the site have a calm, controlled environment?

“We want a very safe area for them to be,” Zaia says. “That doesn’t necessarily mean that it’s sequestered. We prefer to be prominent with quality buildings, visible signage, and safe parking.”

While he appreciates access to all of the data points, Zaia says he relies on local intelligence before sealing a deal. 

“The market planning software is a snapshot in time,” he says. “What my real estate team can tell me is there are 1,200 houses being built on this north side, there’s a brand-new school coming up over here, and, oh, by the way, a brand-new retail development is being planned for this area and should break ground at the end of 2026.”

In the end, it’s about what Zaia calls “permission to believe.” The tech creates confidence, the team confirms it, and students get a safe, stable space to learn.

Technology meets intuition

For Potbelly, selecting the right franchise location is informed by what a potential franchisee will need to operate their business. Sites have clear requirements. 

“We’re an A-site brand,” says John Beckley, Potbelly’s vice president of franchise and corporate real estate. “Our standard prototype is 1,800 to 2,200 square feet. We love end caps. We don’t necessarily need to have a drive-thru. Anchored centers are preferred, but we can do unanchored centers where it makes sense.”

When looking for locations, the real estate teams use Sitewise, Potbelly’s primary mapping and market planning platform. 

“I see Sitewise as the bridge between the art and science of what we do,” Beckley says. “There’s a lot of forecasting, and it identifies trade areas.” 

It helps the company plan intentionally. Information is forwarded to franchisees, but it’s not just handed over. Potbelly teaches franchisees how to use the information.

Raymond Phillips, director of market planning and analytics at Potbelly, says he enjoys getting into the weeds with all the possibilities modern technology offers. 

“We just finished evaluating new tools within Sitewise that will allow us to basically build a map view that has the data layers we would want the franchisees to see,” Phillips says. “They can go in and pan around the map zoom. They can only see the territory we share with them and the data layers we share with them, but it is a live, interactive map.”

Those data layers include information about competitors, trade areas, visibility, parking, access, and daytime population. 

“We’re a lunch-centric brand,” Beckley says. “Most of our transactions happen before 4 p.m., so we’re looking at daytime population first.”

All of that data is layered together as different models are combined to create a single forecast. As for artificial intelligence, Potbelly is experimenting.

“It is a new tool in this specific field,” Phillips says. “I would say it hasn’t been fully vetted out.”

AI is being used to evaluate variables in Potbelly’s forecast model, but it hasn’t proven its usefulness yet. “It’s not something we’re ignoring,” Phillips says, “but it’s not fully integrated either.”

Even the most advanced analytics can’t replace years of experience. Beckley says there’s always a need for a human component because franchising is a relationship business. 

“In franchise development, the relationship and the time that you spend together are critical,” he says. “When you’re having these discussions, you always need to lean on experience.”

Phillips echoes the sentiment: “The technology is a tool to give you guidance, and then humans make the final decisions.”

Technology and Site Selection

Technology is transforming how franchise brands identify and evaluate new locations. Here are five key ways data and digital tools are influencing the site selection process:

• Predictive mapping tools. Franchise brands now rely on specialized mapping platforms that combine demographics, competitor data, and performance metrics to identify optimal growth areas.
• Custom territory design. Modern systems allow real estate teams to build tailored territories using ZIP code-level inputs, population thresholds, and service range criteria.
• Data layer integration. Site selection models pull from multiple data layers, such as income, foot traffic, co-tenancy, and labor patterns, to forecast site performance.
• AI and machine learning applications. AI is used to improve forecasting and identify patterns in consumer behavior, trade areas, and competitive saturation, but full integration is still evolving.
• Physical validation required. Every potential location undergoes a physical site visit to confirm on-the-ground conditions and uncover factors that data can’t capture.