Walled Garden or Open Road? Your POS Is About to Decide Your Future
Paul Molinari had just returned from FSTEC in Dallas when he opened a recent episode of Modern Solutions for Modern Restaurants with a scene from the ballroom. The room was standing-room only. Par's SVP of product clicked to a slide with a single number in giant orange type: 44%, the share of operators replacing their point of sale in 2026. Molinari said the room went quiet, because for a technology vendor that figure is either the biggest opportunity of a career or the thing that walks a best customer out the door. For operators, it means a decision they may live with for five to seven years is arriving for nearly half the industry at once.
Molinari's habit is to read the footnote, and this one matters. The statistic comes from Hospitality Technology's 2026 POS study, which found that 44% of restaurants plan to replace or significantly upgrade their systems this year. The slide said "replacing." He wasn't calling it a gotcha, since it's still a huge number, but it's a reminder that a vendor's job is to make a stat land. The same study explains the motivation. Ninety percent of operators said they want to strengthen integration across platforms, 80% named guest experience and loyalty, and 60% wanted deeper insight from their data. Nobody asked for faster card swipes. Every goal is about the POS connecting to something else, which means it is now judged as the foundation of the tech stack rather than as a register.
Molinari traced how it got that way. A few years ago the POS sat at the center of three connections: payments, kitchen, and reporting. Today it sits in the middle of a spiderweb of delivery, kiosks, online ordering, loyalty, labor, inventory, guest data, analytics, and, lit up in neon orange, AI. Operators bolted on each tool in response to a crisis, whether delivery during the pandemic, kiosks when labor got tight, or loyalty when third-party apps started owning the guest. Each choice made sense alone, but together they turned the POS into a switchboard it was never designed to be. Drawing on nearly a decade at Crunchtime, he described what happens when integrations are messy: managers exporting spreadsheets, a VP of ops stitching numbers together at midnight, and nobody trusting the dashboard. An AI layer makes that worse, because AI is only as good as the data it's fed. If sales, labor, and guest data live in three systems that sync once a night, he said, you don't have an AI strategy.
That pressure is what forces the fork. One path is the walled garden, or all-in-one, where a single vendor supplies the POS, payments, online ordering, loyalty, kitchen display, and increasingly payroll, marketing, and AI. Molinari called Toast the poster child, noting roughly 140,000 locations, a landmark Applebee's deal in 2025, a new drive-through product he saw at FSTEC, and a native AI layer called Toast IQ. The appeal is one contract, one support number, and data that is unified in theory, which is a real gift for an operator without a big IT department. The other path is the open ecosystem, where the POS is a core that plugs into best-of-breed partners through open APIs. Par built its identity here, and Molinari cited its 2021 claim of more than 200 integration partners for Brink. That route offers flexibility and no single vendor holding your operation hostage, which is worth a lot to an enterprise brand with a real tech team.
He also pushed back on the tidy framing, starting with who is on the stage. Par sells an ecosystem, yet it also bought the loyalty platform Punch for close to $500 million in 2021 and pitched the deal as a unified platform. Toast and Square call themselves open as well, but they also build their own versions of what their partners sell. In Molinari's words, the partner is invited into the garden while the gardener grows the same crop. So the real question isn't open or closed, since every serious POS is some kind of hybrid. He offered four sharper ones. Who owns the data, and can you get all of it out in real time without paying a toll? What does an integration cost the partner, and what does it cost you? Does the POS compete with the partners in its own marketplace? And what does the exit look like if you leave? For independents and small multi-unit groups, he said, all-in-one is compelling because simplicity is a feature, though you're betting on one vendor's roadmap. Enterprise brands with technology teams and a distinct operating model are better served by open, as long as they accept the integration burden and the 7 p.m. Friday refereeing between vendors. Emerging brands in the messy middle should look for a hybrid they can grow into, with a strong suite, a genuinely open API, a real marketplace, and data they can export whenever they want.
For solution providers, he offered three moves. Map your customer base to their POS so you know how exposed you are and which accounts might be in the 44%. Treat integrations as a product that is certified, documented, and fast to deploy, because the vendors that survive a POS switch are the ones that make the move easy. And make sure you're more than a feature, since anything a POS could ship as a checkbox puts you on borrowed time. He added a caution, from outside his own expertise, that POS companies often charge for data access and marketplace placement, so a vendor buried below the fold should assume someone else is paying to be seen. His closing advice was that neither side of the fork is universally right. The mistake is choosing by default, because it's what your last brand used, the demo was slick, or a big orange number made you feel late. Write down where your brand will be in five years, ask the four questions, and pick the foundation that fits who you're becoming.
Marketing help? Visit: https://popcorngtm.com
