The Company That Turned America's Hidden Fiber Into a Map You Can Actually Read

The internet is not a cloud. It is a stubbornly physical thing: glass threads buried under highways, steel towers bolted to rooftops, windowless buildings humming in industrial parks. Almost nobody can see it. GeoTel Communications has spent more than two decades making it visible. The Lake Mary, Florida company does not lay cable or light a single building. It does something more quietly useful - it knows where all of that infrastructure is, and it sells that knowledge to the people who have to make decisions about it.

That is the whole business, and it is a bigger one than it sounds. Every fiber acquisition, every rural broadband grant, every 5G rollout and every data center site search runs into the same question early: what is already in the ground here, and who owns it? Answering that from scratch is slow, expensive and error-prone. GeoTel's pitch is that you should not answer it from scratch. You should buy the map.

It helps to picture the alternative. Without a dataset like this, a network planner starts with a pile of carrier PDFs, some public filings, a few phone calls, and a lot of guesswork about whether a building three blocks away is actually lit. That work can take weeks and still be wrong. GeoTel compresses it into a lookup. The company frames itself as the leading provider of telecommunications infrastructure data in GIS and internet-based spatial technologies, and the claim rests less on any single clever feature than on the sheer patience of having assembled and maintained the picture for over two decades.

What it doesTurning invisible infrastructure into a layer on a map

GeoTel's raw material is telecom infrastructure data - carrier fiber routes, on-net and near-net fiber-lit buildings, cell towers, central offices, colocation and data centers, and the geographic boundaries that define service territories. On its own, that data is a pile of coordinates. GeoTel's work is to clean it, verify it, and layer it onto high-resolution street maps and aerial imagery so a route stops being an abstraction and becomes a line you can trace to a specific building on a specific corner.

The company sells that intelligence two ways. The first is as GIS data: files formatted to drop straight into a customer's existing mapping tools, including Esri's ArcGIS, where GeoTel is a partner. The second is TeleTracker, a browser-based platform that puts the same datasets behind a point-and-click interface, so a business analyst can study a market without ever opening GIS software.

The atlasWhat is actually inside the database

The scale is the point. GeoTel reports roughly 5.7 million carrier fiber route miles and about 15.7 million fiber-lit buildings, drawn from thousands of carriers. There are more than half a million cell towers and rooftop sites, plus central offices and data centers marking where equipment and colocation actually live. These are not static once and forget maps - the value depends on keeping them current, which is the unglamorous, never-finished part of the job.

Who buys itThe rivals who all need the same map

GeoTel's customer list reads like a directory of everyone touching a network. Carriers and internet service providers use it for competitive intelligence and to see where their fiber sits relative to a rival's. Wireless and 5G operators use it for network planning and site selection. Data center and colocation firms use it to understand connectivity around a facility. Infrastructure developers and investors use it during diligence, because you cannot value a network you cannot see. Commercial real estate teams use fiber-lit building data to tell tenants what connectivity a property already has. And government agencies use it for broadband planning and analysis.

That is the clever structural trick. Because GeoTel is a neutral data provider rather than a carrier, it can sell the same intelligence to companies that would never share a spreadsheet with each other - and to the investors circling them, and to the governments regulating them. The neutrality is the product.

Two ways to buyGIS files, or a browser you don't need training for

The split between GIS data and TeleTracker maps neatly onto two kinds of buyer. Some customers have their own geospatial engineers and want the raw layers to combine with internal data. Others just need answers and would rather not stand up a GIS stack to get them.

TeleTracker is the piece that widened GeoTel's audience. Its latest release, V7, added data visualization tools and expanded GIS access, letting users build accurate digital representations of telecom assets and share a clean map image with a stakeholder who will never log in themselves. The platform folds street-level detail, aerial imagery and analysis into a single view, so an underground route and the buildings around it show up together.

The operatorBootstrapped, boring, and still here

GeoTel was founded by Dave Drazen, who has run it as CEO since 2001. The team is small - about 15 people, built around in-house geospatial engineers rather than a large sales floor. There is no venture round in the story, no headline funding event. The company has simply maintained a hard-to-replicate dataset and sold it, year after year, for over two decades. Third-party estimates put annual revenue around the low eight figures, which for a 15-person data business is a healthy place to be.

A quick note that clears up a recurring mix-up: this GeoTel is not the 1990s call-center software firm named GeoTel Communications Corp that Cisco acquired in 1999. Same name, entirely different company. The GeoTel in this story is the Florida telecom-data house, privately held and independent.

The market fitThe data layer under everyone else's decisions

Where does a company like this sit? Underneath. GeoTel is not competing to build networks or to win subscribers. It occupies the data layer that other people's decisions rest on. The alternatives to buying from GeoTel are telling: assemble the data yourself from public filings and carrier disclosures, lean on a broader geospatial data vendor, or trust public resources like the FCC's national broadband map. Each of those works up to a point, and each is exactly the friction GeoTel removes by having already done the assembling and the upkeep.

The moat is not a clever algorithm. It is coverage and currency - having more of the assets, mapped more accurately, kept more up to date, than a buyer could justify building alone. In a niche defined by one plain question, that turns out to be enough. What people can copy from GeoTel is less a technology than a posture: find an expensive question a whole industry keeps re-answering badly, answer it once, keep the answer fresh, and license it to all sides. The conditions where it does not work are just as clear - if the data becomes free and current from a public source, or a buyer's need is a one-off small enough to Google, the map loses its premium. GeoTel's bet, sustained for 20-plus years, is that neither of those is true for telecom infrastructure at scale.

For anyone who has to decide where to dig, what to buy, or which building to connect next, that bet is the reassuring part. The fiber is invisible. The map does not have to be.