The call came as a surprise. Hongwei Liu's company had spent years perfecting the unglamorous art of helping shoppers find the Gap in sprawling suburban malls. Then the Department of Homeland Security reached out. They weren't interested in retail foot traffic.
Mappedin, a Waterloo, Ontario-based startup, just closed a $24.5 million growth round—announced on April 7, 2026—and the money tells a story about how indoor mapping has quietly migrated from the marketing department to the emergency operations center. What began as a wayfinding tool for Simon Property Group malls now counts among its use cases something considerably more urgent: getting accurate floor plans into the hands of first responders who, until recently, often arrived at crisis scenes with little more than a street address.
The funding, led by Edison Partners and Betatron Venture Group, lands at a peculiar moment. A 2025 survey by geospatial firm DATAMARK counted at least 17 states with active initiatives or requirements for digital indoor school maps: Florida, Texas, New Jersey, Wisconsin, Georgia, among others. The FCC is tightening rules around vertical location data for 911 calls—knowing which floor someone's calling from matters when seconds count. And the technology itself has changed. AI can now parse a PDF floor plan and generate a 3D model in under three minutes. LiDAR sensors, once exotic and expensive, ship standard in iPhones.
Liu frames it as infrastructure. "We believe accurate, accessible indoor maps should exist for every building," he's said, "not just the ones with marketing budgets." It's an expansive vision, and one that might finally have the regulatory tailwinds to become something more than a pitch deck aspiration.
The Messy Reality of Indoor Space
Here's the problem that exists right now, today, in most American buildings: the maps are a disaster.
Commercial real estate teams commission glossy 3D renderings to woo tenants. Facility managers keep CAD files buried in maintenance folders. Security personnel print floor plans, laminate them, stick them in binders. None of these systems talk to each other. Most buildings—offices, apartment complexes, smaller retail centers—have no digital map at all.
The market has splintered accordingly. Enterprise GIS giants like Esri fold indoor data into sprawling spatial intelligence platforms aimed at government agencies. Specialized players like MazeMap, which recently partnered with education software provider Ellucian, focus on campus navigation—solving the eternal freshman question of "where is Econ 101?" Digital twin companies, notably Matterport (now owned by CoStar Group), create photorealistic 3D scans primarily for real estate listings and insurance documentation. Matterport reported $99.6 million in subscription revenue for 2024.
Mappedin occupies something of a middle lane. The company offers AI-powered tools that don't require specialized surveying equipment, paired with software development kits that let app builders embed navigation. An iOS app leverages Apple's RoomPlan framework and device LiDAR to scan spaces when formal floor plans don't exist. The web-based Maker tool, which got an AI upgrade mid-last year, aims to put map creation within reach of property managers rather than reserving it for trained GIS specialists.
Sizing the opportunity depends on where you draw the boundaries. Precedence Research projects the broader digital map market growing from roughly $31 billion this year to $116.6 billion by 2035. The LiDAR hardware and software segment hit $2.8 billion in 2025 and is expected to reach $3.32 billion in 2026, per Mordor Intelligence. GIM International has pegged the LiDAR mapping market specifically at over $4 billion for the current year. SLAM systems—simultaneous localization and mapping tech critical for autonomous robots—are projected to grow from $1.81 billion in 2024 to $9.88 billion by 2034 with an 18.5% CAGR.
Those are big numbers. Whether they translate into defensible businesses is another question.
When Legislators Get Involved
The regulatory shift is harder to quantify but possibly more significant than the technology itself.
A 2025 survey by geospatial firm DATAMARK counted at least 17 states with active initiatives or requirements for digital indoor school maps: Florida, Texas, New Jersey, Wisconsin, Georgia, among others. West Virginia allocated $4.7 million for mapping grants. South Carolina's education superintendent requested $5 million for statewide digitization a few years back.
Some of this is moving from aspiration to statute. Kentucky's HB 652, which cleared committee in mid-March, would consolidate school mapping under the state's 9-1-1 Services Board at an estimated cost of $15 per student. Washington's HB 2328 proposes standardized digital mapping protocols with verified walkthroughs. New York introduced S2588 in the current legislative session, requiring up-to-date floor plans overlaid on aerial imagery as part of school safety plans.
These aren't about helping students find the library. They're emergency response infrastructure.
The DHS collaboration with Mappedin dates back to 2019. A feature article published by the agency's Science & Technology directorate on May 2, 2024 spotlighted AI that generates 3D indoor maps from floor plan images in 30 seconds to three minutes. The use case: first responders who previously showed up at shootings, fires, or medical emergencies with zero spatial intelligence beyond the street number on the mailbox.
The FCC is pushing complementary mandates. A notice of proposed rulemaking released in late March 2025 seeks enhanced vertical location data for 911 calls—essentially, z-axis accuracy within three meters for 80% of calls in major metro areas. Existing deadlines already required dispatchable location or floor-level data by April 2024 in top markets.
Technology, meanwhile, is converging from several directions simultaneously. Apple released RoomPlan for developers, turning every LiDAR-equipped iPhone into a basic floor plan scanner. Google announced Gemini-powered features for Maps in mid-March, mainstreaming AI-assisted navigation. Microsoft's workplace management platform, Places, requires indoor maps formatted in IMDF—Indoor Mapping Data Format, a standard Apple originally developed—creating overnight demand for conversion services. Companies like Archilogic now charge $329 per floor plan conversion, which is either a lucrative niche or the kind of tedious work that gets automated away.
Academic research is accelerating in parallel, perhaps more than most observers realize. Recent papers demonstrate real-time SLAM with 3D Gaussian Splatting, AI-powered planning for active mapping systems, semantically enhanced maps that respond to natural language queries ("show me the nearest accessible restroom"). Industry conferences like Geo Week, held in February, featured sessions on bringing reality capture indoors, with vendors demoing handheld LiDAR rigs and AI workflows accessible to non-specialists.
The Pivot Nobody Planned
Mappedin didn't start here. The company emerged from the University of Waterloo's Velocity accelerator and delivered its first wayfinding kiosk in late 2011. A $1.51 million seed round came in 2015, followed by an undisclosed Series A led by Channel Equity Partners in early 2023. The original pitch centered on shopping malls and entertainment venues—branded navigation apps without the cost of hiring engineering teams.
The Security Center product line, unveiled at ISC West in mid-March, represents something of a reinvention. Features include annotated maps showing camera locations, access control points, emergency equipment. Integrations with video management systems let security personnel click a camera icon on a floor plan and pull up live feeds instantly. It's fundamentally the same 3D mapping technology, repackaged for customers who care about response time rather than retail conversion rates.
That's a different buyer, with different budget cycles.
Competitors are making comparable shifts with varying emphases. MazeMap's Ellucian partnership embeds indoor navigation directly into student information systems at universities—solving the perennial "where is my classroom" problem. GoodMaps focuses on accessibility, deploying LiDAR backpack rigs to create navigation infrastructure for blind and low-vision users. Pointr's MapScale AI engine, demonstrated at Geo Week, automates large-scale map production from CAD and PDF files.
Esri's ArcGIS Indoors integrates with the company's sprawling GIS enterprise stack, adding AI assistant capabilities (currently in beta) and conversion workflows from BIM modeling software. It targets organizations already running Esri infrastructure who want indoor data in the same system. HERE Technologies and TomTom maintain indoor map offerings, though both companies are better known for automotive GPS.
The digital twin category overlaps but serves different priorities. Matterport captured 503 BMO bank branches in three months during 2024, generating photorealistic 3D models primarily for virtual tours and facility documentation. NavVis, with its VLX 3 mobile mapping system, courts architecture and construction firms needing survey-grade precision. Both have sophisticated SLAM and AI processing pipelines; neither positions itself primarily as a navigation platform.
What separates these approaches is less about underlying technology—most blend LiDAR, computer vision, and AI in some configuration—and more about distribution strategy and end-user workflows. Mappedin's wager is on tools simple enough for property managers to deploy directly, bypassing specialized surveying contractors.
Scale claims require a footnote. A company information packet from December 2025 cited 755 million square feet across 1,100 venues in 40 countries. The April funding announcement claims over 10 billion square feet across 86 countries—an order-of-magnitude jump that suggests either explosive recent growth or a change in measurement methodology. Edison Partners, the lead investor, uses the 10 billion figure in its press materials. Independent verification of either number isn't readily available; consider them directionally indicative rather than audited facts.
What Happens Next (And Why It Matters)
Indoor mapping sits at the confluence of several simultaneous technology shifts, and it's not entirely clear which will matter most.
Spatial computing—Apple's Vision Pro, Meta's Quest headsets—will eventually require indoor positioning that doesn't depend on GPS signals. Autonomous mobile robots navigating warehouses, hospitals, and airports need real-time maps. The EU AI Act, which entered force in August 2024 with compliance deadlines phasing through August 2027, may classify certain indoor mapping AI systems as high-risk when deployed in critical infrastructure, triggering transparency and governance mandates.
Privacy concerns remain oddly underexplored given the sensitivity of the data. Detailed floor plans of schools, government buildings, corporate campuses—these are assets you don't necessarily want floating around public cloud platforms. Storing them remotely is necessary for first-responder access, but introduces exposure risk. NIST publishes privacy risk assessment guidance, and some municipalities restrict public access to facility floor plans through geospatial data policies, yet standardized security frameworks for indoor mapping data don't really exist. An industry article from early April flagged consent and secure storage as open questions for 360-degree scanning workflows.
The venture capital thesis has evolved. Early indoor mapping investments centered on retail analytics and customer engagement—heat maps of foot traffic, store layout optimization. Current funding rounds emphasize infrastructure and regulatory compliance. Edison Partners' investment rationale, articulated in the Mappedin announcement, focuses on "opening access for first responders, safety organizations, and large public venues." That's a fundamentally different buyer than a shopping mall marketing director, with different budget authority and procurement timelines.
Founders watching this sector should pay attention to the regulatory momentum. When state legislatures allocate millions for school mapping and the FCC mandates 911 vertical accuracy, enterprise customers suddenly have budget line items that didn't exist 24 months ago. The challenge shifts from convincing someone they need indoor maps—elected officials are doing that work—to navigating interoperability and standards. IMDF, OGC IndoorGML 2.0 (published mid-2025), and GeoJSON are all proliferating, but vendor lock-in remains frustratingly common.
The technical frontier is migrating from "can we map this building" to "can we keep it current." Buildings change constantly. Tenants relocate, walls get added, equipment moves. Matterport's BMO project captured 503 branches once; maintaining that data requires either scheduled re-scans or integrations with building management systems that update maps automatically as changes occur. Academic research on continuous SLAM and AI-powered change detection suggests this is solvable in principle. Few commercial products handle live updates gracefully in practice.
Mappedin's stated use of proceeds—expanding toward city-scale indoor mapping and opening the platform to third-party developers—hints at ambitions beyond selling software to individual venues. Something closer to infrastructure. Not a SaaS product a stadium licenses, but a foundational layer that other applications consume.
Whether that vision materializes depends on execution, obviously, but also on how quickly regulatory mandates force standardization. If every school, airport, and sports arena in North America needs a compliant digital map within the next couple years—and that's increasingly plausible—somebody has to build the pipes. Liu seems to think he's got a head start.
The harder question is whether indoor mapping becomes a durable business or gets absorbed into larger platforms the way early navigation startups eventually got acquired by Google and Apple. DHS partnerships and state legislative mandates change the calculus somewhat. Infrastructure plays typically have more defensibility than consumer apps.
For now, the company that started by helping shoppers find Nordstrom has a rather different problem set: how do you digitize the interior of every building that might someday require emergency response? It's an infrastructure challenge masquerading as a mapping problem, which is possibly why it took a hardware shift (cheap LiDAR), a software breakthrough (AI floor plan parsing), and a regulatory push (state mandates) happening simultaneously to make it feel suddenly, unexpectedly viable.
