On July 20, 2026, The Industry Group and the Kiosk Manufacturer Association released the TIGER 2026 Global Self-Service Kiosk Market Report in conjunction with RetailNOW 2026. The report sizes the global self-service kiosk market at $39.4 billion in 2024. By 2031, TIGER projects growth to $82.1 billion across 14 segments. Unattended POS alone is projected to grow from $1.254 billion to $3.418 billion at a 15.4 percent CAGR. Those numbers are significant. The definition the report uses to arrive at them is more significant.
Specifically, TIGER 2026 describes POS as a distributed, unattended, software-defined commerce layer. It spans self-checkout, self-ordering, EV charging, smart vending, healthcare check-in, and digital signage-enabled transaction points. The category is no longer a single fixed-lane terminal. The report concludes: POS is now the operating system of self-service commerce.
That definition change is not semantic. It is architectural. A terminal is a capital expense with a depreciation schedule. An operating system is a technology investment that requires connectivity, edge compute, software update cycles, data pipeline integration, and AI capability at the device level. Retailers who are still buying kiosks the way they bought furniture in 2015 are acquiring operating systems with a procurement process designed for furniture. The gap between those two approaches is where most self-service deployments fall short.
What TIGER 2026 Actually Said About Where the Market Is Going
The growth TIGER 2026 documents does not come from retailers buying more of the same device. Four forces are changing what the device is. First, AI and computer vision capabilities are moving to the edge. Second, digital payment infrastructure is becoming embedded in the device rather than attached to it. Third, accessibility and compliance requirements are forcing more sophisticated hardware configurations. Fourth, labor cost pressures are accelerating unattended commerce models in sectors that previously relied on staffed service.
The 14 segments TIGER covers illustrate how far the kiosk category has expanded beyond the retail checkout lane. Self-ordering in quick service restaurants, check-in at healthcare facilities, payment at EV charging stations, and transaction points in digital signage installations are all part of the same market. Moreover, all of them run on the same underlying technology stack: edge compute, AI inference, digital payment rails, and cloud connectivity for fleet management and remote updates.
In other words, the market message from TIGER is clear. The retailers and solution providers who will capture the $82.1 billion opportunity are the ones who understand that they are not buying terminals anymore. They are deploying intelligent commerce nodes that have to be designed, connected, maintained, and upgraded like software products, not like physical fixtures.
The Architecture Decision Most Retailers Are Still Not Making
I have been building and deploying self-service hardware across the United States and Latin America for over two decades. The conversation at the purchasing stage has changed dramatically in the last three years. However, one pattern persists: the first question most retailers ask is still about price per unit. That was the right question in 2015. For a software-defined commerce layer, it is the wrong question to ask first.
The right first questions are about architecture: what does the device need to know, how does it get that information, and how does that information make it smarter over time? Those three questions determine whether the kiosk a retailer purchases compounds in value over five years or becomes obsolete in two. As I described in the self-service ROI gap analysis, the gap between the vendor’s projected return and the retailer’s actual Year-1 result is almost always a deployment design gap. The architecture decision that precedes the purchase order is where that gap either opens or closes.
The Compute Layer Has to Be Ready for AI Before the AI Is Deployed
The TIGER 2026 report specifically highlights Intel Core Ultra and vPro architectures as the compute foundation enabling AI inference, remote fleet management, and reliable operation for enterprise-scale self-service deployments. That is not a vendor endorsement. It is a description of a minimum hardware standard for a device expected to run AI workloads at the edge without cloud dependency.
A kiosk that requires a cloud round-trip to process an AI recommendation or a fraud detection signal is not an edge AI device. It is a connected terminal with an AI subscription. Those are different things with different latency profiles, different reliability requirements, and different failure modes. In particular, LatAm store environments have connectivity conditions that make cloud-dependent AI inference an operational risk. The compute decision has to happen at the hardware specification stage, before the purchase order is signed. Retrofitting is not possible after installation. The device has to be replaced.
The Connectivity Layer Is Not an Add-On
The TIGER 2026 vision of kiosks, digital signage, and ESL as a single software-defined layer requires those devices to communicate with each other and with store data systems in real time. A self-checkout kiosk that updates a price without that update being reflected in the electronic shelf label creates a customer trust problem. Self-ordering terminals that ignore real-time inventory create operational failures. Digital signage that cannot receive pricing updates creates a compliance risk.
Consequently, the connectivity architecture that ties these devices together is not an add-on to the hardware deployment. It is a prerequisite for it. Retailers who deploy kiosks without designing the network layer that connects them to ESLs, digital signage, and inventory systems are buying individual devices. They are not building an integrated commerce layer. The device price is the same. The operational capability is completely different.
The Data Layer Determines Whether the Device Gets Smarter Over Time
The most important question to ask about any self-service device in 2026 is not what it does at installation. It is what it learns after installation. A kiosk that captures transaction data, interaction patterns, and customer flow signals and feeds that data to the store’s analytics layer compounds in value over time. By contrast, a kiosk that processes transactions and generates receipts without sending structured data anywhere is worth exactly what it cost on day one. It depreciates from there.
A device that generates data and a device that feeds structured, actionable data into your analytics infrastructure are not the same product. One is a source of insight. The other is a source of storage costs. As I described in the self-checkout loss prevention analysis, the retailers who manage shrink effectively are the ones measuring the right things at the device level. Measurement requires data. Data requires a pipeline. The pipeline has to be designed before the device goes on the floor.
What the $82.1 Billion Number Actually Means for the Retailers Buying Now
The TIGER 2026 projection of $82.1 billion by 2031 is not an abstract forecast. It is a description of competitive pressure. The market nearly doubles in seven years. The retailers who deploy intelligent, software-defined layers with the right architecture in 2026 and 2027 will have a seven-year head start in operational learning and AI capability compounding. The retailers who wait will not.
The unattended POS segment growing at 15.4 percent CAGR is the most relevant signal for traditional grocery and general merchandise retailers. That growth rate reflects the acceleration of a category still considered a pilot format three years ago. The retailers who treated self-checkout as a pilot in 2022 and 2023 are already behind on the architecture decisions TIGER 2026 describes as table stakes. They are scaling the old architecture, not the new one. Additionally, the retailers making those architecture decisions now will be running intelligent commerce layers rather than upgraded terminals when the market hits $82 billion.
What This Means for LatAm Retailers
The TIGER 2026 growth projections are global, and the LatAm self-service market has specific characteristics that make the architecture decisions more consequential, not less. Network infrastructure in many LatAm store environments is less reliable and more expensive than in U.S. or European markets. That makes the edge compute decision more critical, because a device that depends on cloud connectivity for AI inference will fail in ways that a device with local compute capability will not.
Furthermore, the labor cost dynamics driving self-service adoption in the United States are also present across LatAm markets. Currency volatility makes payroll costs unpredictable in ways that technology capital expenditures are not. LatAm retailers who invest in software-defined self-service infrastructure now are hedging against a labor cost structure that has historically been volatile. The device that processes transactions without a cashier does not require a wage adjustment when the local minimum wage changes.
As a result, the architecture decisions that determine whether a self-service deployment succeeds in a LatAm environment are the same ones TIGER 2026 describes for the global market. The stakes are higher because the margin for infrastructure failure is smaller.
Three Questions to Ask Before the Next Hardware Purchase Order
Can This Device Run AI Inference Locally Without Cloud Dependency?
The answer determines whether you are buying an edge AI device or a connected terminal with an AI subscription. Ask the vendor specifically: what happens to the AI capability if the network connection drops for 30 minutes? If the device reverts to a basic transaction mode, you are not buying edge AI. You are buying cloud AI delivered through a device at your location. Those are different products with different reliability profiles and different long-term cost structures.
What Structured Data Does This Device Send Back to My Systems?
Not what data it can theoretically generate. What structured data it actually sends, in what format, to which systems, on what schedule. Ask to see a sample data output from a live deployment before you sign the purchase order. That is the fastest way to determine whether you are buying operational intelligence or operational noise.
How Is This Device Updated When the AI Capability Changes?
A software-defined commerce layer has a software update cycle. The device you install today will need updated AI models, updated compliance configurations, and updated security patches over its operational life. Ask the vendor four specific questions before signing. How are updates delivered? Who manages the process? What is the downtime per update? What happens to a device two software versions behind the current release? The answers will tell you whether you are buying a managed lifecycle or a device that requires physical replacement every three years.
The kiosk market is growing to $82.1 billion by 2031. The retailers who capture that growth are not the ones who buy the most devices. They are the ones who buy the right architecture, deploy it with the right data pipeline, and update it with the right software cycle. Every other retailer is buying hardware in a software market.
If you are evaluating a self-service hardware deployment or reviewing your current device architecture for AI readiness, connect with me here or reach me on LinkedIn. I am happy to walk through the deployment framework we use across the U.S. and Latin America.
Adriana Rivas is a retail technology executive and AI strategist, and the founder of a U.S.-based hardware company specializing in self-service kiosks, POS systems, electronic shelf labels, and digital signage deployed across the United States and Latin America. She is the award-winning author of How to Implement Self-Service Without Failing (Amazon #1 Hot New Release, Silver Nonfiction Book Award 2025) and recipient of the Gold Stevie® Award, Thought Leader of the Year 2026. She is also recognized by Thinkers360 as the #7 Global Thought Leader in Retail and a Certified Master Expert in Retail.