Dark Fiber Lease Rate Intelligence SaaS for Regional Network Operators
Lumen Technologies claims that replicating its 400,000-route-mile fiber network would cost $150 billion. Zayo is building 5,000 new route miles specifically for AI data center interconnection. BIG Fiber just raised $250 million from Stonepeak to lay metro dark fiber for hyperscalers. And the 1,800 regional fiber operators sitting on the other half of America's lit and unlit strands are pricing 20-year Indefeasible Rights of Use contracts the same way they always have: a phone call to a broker, a gut check against the last deal they closed, and whatever number the buyer's consultant suggested first.
The Problem
The U.S. dark fiber network market was valued at $1.18 billion in 2025 and is projected to reach $1.84 billion by 2031, growing at a 7.65% CAGR (Mordor Intelligence). The global market is larger and moving faster: $6.90 billion in 2025, projected to reach $21.88 billion by 2033 at a 15.9% CAGR (ResearchAndMarkets). These numbers describe a market in the early stages of a massive repricing event driven by artificial intelligence infrastructure, and the operators who own the underlying fiber assets have almost no tools to price it.
Dark fiber pricing is negotiated route by route, deal by deal, with virtually no transparency. A CTC Technology & Energy analysis of lease rates across multiple U.S. markets found that pricing for the same basic product (one strand of single-mode fiber, one mile) varies by more than 60x depending on geography and market type (though as we discuss in Limitations, the true within-category spread is more like 3-5x). Rural long-haul routes in Illinois trade at $3.44 per strand per month. Palo Alto metro routes command $177 to $591 per strand per month. A municipal program in Riverside, California charges $125 per strand-mile per month for short commitments and $87.50 for ten-year terms. Laurinburg, North Carolina publishes rates of $36 to $43 per fiber per mile per month. These are not edge cases cherry-picked to exaggerate the spread. They represent the normal operating range of an industry where every transaction is bespoke and no centralized pricing data exists.
The opacity is so thorough that even SEC filings redact the numbers. An exhibit from a fiber network master lease agreement filed with the SEC shows 27 long-haul routes totaling 13,500 route miles, with every single rate entry replaced by [**********]. The filing discloses the routes (Atlanta to Nashville, Denver to Salt Lake City, Sacramento to San Francisco) but treats the per-mile lease rates as confidential business information too sensitive to publish, even in a regulatory document designed for investor transparency.
This matters now more than it ever has because the demand side of the market is undergoing a phase change. Hyperscale data center operators like Microsoft, Google, Amazon, and Meta are ordering 12 to 48 fiber pairs per route, up from the four-pair standard of just a few years ago (Mordor Intelligence). They need this capacity for AI training clusters that require low-latency, high-bandwidth interconnection between GPU-dense facilities, and they need it on routes that didn't exist in anybody's network plan three years ago. The result is a land grab for fiber capacity that is repricing every strand in the country, yet only the operators sophisticated enough to recognize it are capturing the premium.
Market Size
Supply-side TAM: The addressable market is regional and independent fiber network operators in the United States. According to FCC Broadband Data Collection filings and Fiber Broadband Association membership data, approximately 1,800 entities operate fiber networks in the U.S. outside the top five carriers (Lumen, Zayo, Crown Castle, AT&T, Verizon). This includes independent fiber companies, electric cooperatives with fiber divisions, municipal networks, competitive access providers, and cable operators with metro fiber assets. Of these, roughly 900 operate networks with sufficient scale (50+ route miles or 500+ strand-miles) to justify subscription analytics. At $499/month for a Standard tier (route-level rate benchmarking and IRU valuation) and $1,499/month for a Premium tier (real-time demand signals, hyperscaler RFP intelligence, and contract negotiation support), with a 60/40 tier split, the blended ARPU is $899/month. At 900 addressable operators: $9.7 million in base ARR.
Demand-side expansion: The buyers of dark fiber (enterprises, data center operators, wireless carriers, content providers) have an equally acute need for rate intelligence. A mid-size enterprise leasing dark fiber for a campus network has no way to determine whether a quoted rate of $400/strand-mile/month is at the 25th or 90th percentile for its metro area. A data center operator evaluating a $2 million IRU for a 20-year term on a 50-mile route cannot benchmark that price against comparable transactions because no comparable transaction database exists. At $299/month for 2,000 active enterprise buyers of dark fiber, the demand-side layer adds $7.2 million ARR. Combined SAM: $16.9 million. Year 3 target: 350 operator subscribers + 800 enterprise subscribers at blended $760/month = $10.5 million ARR.
Transaction advisory upside: Dark fiber IRU transactions routinely exceed $500,000, and portfolio-level fiber asset sales run into the hundreds of millions. A transaction advisory fee of 50 basis points on closed deals facilitated through the platform, applied to an estimated $2 billion in annual fiber asset transactions among regional operators, adds a $10 million revenue opportunity for a mature platform with sufficient market share to influence deal flow. This is a later-stage revenue stream but directly analogous to how CoStar monetizes commercial real estate intelligence.
The Product
A fiber asset intelligence platform that aggregates anonymized lease transaction data, public rate filings, FCC deployment records, and market signals to produce real-time rate benchmarks and asset valuations for dark fiber networks. Core modules:
- Route-level rate benchmarking: The fundamental product. For any fiber route defined by two endpoints and a geography type (metro, suburban, long-haul, intercity), the platform returns the current market rate distribution: 25th, 50th, 75th, and 90th percentile rates per strand-mile per month, segmented by lease term (monthly, 3-year, 5-year, IRU/20-year). Data sources: anonymized contributed transactions from participating operators, publicly posted municipal fiber rates, state PUC tariff filings, NTIA/BEAD grant applications (which disclose proposed rates), and machine-parsed SEC exhibits from fiber company filings. The cold-start challenge is real, but the public data layer alone (municipal rate schedules, PUC tariffs, BEAD applications) covers approximately 400 distinct rate observations across 38 states before a single operator contributes a transaction
- IRU valuation engine: An Indefeasible Right of Use is the dominant long-term transaction structure in dark fiber, typically 15 to 25 years with annual maintenance fees. Operators signing IRUs today are locking in economics that will govern their revenue through the 2040s, yet most price these deals using rules of thumb inherited from the telecom bust era when fiber was a distressed asset. This module produces a fair-value estimate for any proposed IRU based on route characteristics, current market rates for comparable routes, projected demand growth (incorporating AI data center buildout plans from announced projects), and a discount rate calibrated to fiber infrastructure risk profiles. The output: a probability-weighted range of NPV outcomes that tells the operator whether they are leaving money on the table
- Demand signal radar: Real-time monitoring of announced data center construction, hyperscaler capacity expansion plans, 5G densification timelines, and BEAD grant awards that will create new fiber demand on specific routes. When Microsoft announces a $3 billion data center campus in central Ohio, every fiber operator within 100 miles of that site needs to know within 24 hours, because the procurement process for interconnection fiber starts 18 to 24 months before the facility goes live. This module scrapes planning commission filings, utility interconnection requests, DOE environmental reviews, and corporate press releases to identify demand signals that affect specific routes and regions
- Network asset valuation dashboard: A portfolio-level view that estimates the current market value of an operator's entire fiber network, strand by strand, route by route. Essential for operators considering asset sales, seeking debt financing against their fiber plant, or evaluating acquisition offers from infrastructure funds. The fiber infrastructure investment space is booming: Stonepeak, DigitalBridge, KKR, and EQT are actively acquiring regional fiber networks. Yet sellers consistently undervalue their assets because they have no independent market data to counter the buyer's valuation model
Unit Economics
| Metric | Value |
|---|---|
| Monthly subscription (Standard: benchmarking + IRU valuation) | $499/operator |
| Monthly subscription (Premium: full intelligence suite) | $1,499/operator |
| Demand-side subscription (enterprise buyers) | $299/month |
| Blended ARPU (supply + demand mix) | $760/month |
| Data infrastructure cost per subscriber/month | $42 |
| Data acquisition and parsing cost per subscriber/month | $31 |
| Customer acquisition cost | $4,800 |
| Expected LTV (36-month avg retention, 90% gross margin) | $24,624 |
| LTV:CAC ratio | 5.1:1 |
| Gross margin | 90% |
| Startup cost (18-month runway) | $3.4M |
| Break-even | 22 months |
Methodology note: The 36-month retention assumption reflects the inherent stickiness of benchmarking products in infrastructure markets: operators who use rate data for IRU negotiations become dependent on it for every subsequent deal. CAC of $4,800 reflects a B2B sales motion targeting a niche, technical buyer base through industry conferences (Fiber Connect, PTC), trade associations (NTCA, ACA Connects, Fiber Broadband Association), and direct outreach. The LTV calculation: $760 × 36 months × 90% gross margin = $24,624. Payback period: 6.3 months. Gross margin of 90% reflects a pure software-plus-data model where the primary costs are data acquisition (scraping, parsing, anonymization infrastructure) and cloud compute for the valuation engine.
Go-to-Market
Phase 1 (months 1-8): Build the public data foundation. Aggregate every publicly available dark fiber rate: municipal fiber lease schedules (hundreds of cities publish these), state PUC tariff filings, NTIA BEAD grant applications (which disclose proposed middle-mile rates as a condition of funding), FCC Form 477 and Broadband Data Collection filings, and SEC exhibits from fiber company IPOs, debt offerings, and lease agreements. Parse historical rate data from archived utility commission proceedings. This creates a baseline of approximately 400+ rate observations spanning 38 states without requiring a single operator to contribute proprietary data. Simultaneously, recruit 50 regional operators to contribute anonymized transaction data in exchange for free platform access, targeting members of NTCA – The Rural Broadband Association (850 member companies) and ACA Connects (700+ members).
Phase 2 (months 9-16): Launch the Standard tier at $499/month with route-level benchmarking and the IRU valuation engine. Target the 200+ BEAD subgrantees who are building new fiber networks with federal funding and need to price wholesale access for the sustainability plans required by their grant agreements. These operators are building brand-new fiber assets, often for the first time, and have zero historical pricing data to guide their wholesale rate-setting. BEAD's $42.45 billion in federal funding is creating the largest single cohort of new fiber network operators in U.S. history, and every one of them needs rate intelligence. Integrate with fiber network management platforms (Vetro FiberMap, 3-GIS, IQGeo) to pull network topology data and automate route matching.
Phase 3 (months 17-24): Launch the Premium tier with demand signal radar and the enterprise buyer product. Partner with fiber transaction advisory firms (Cartesian, Columbia Capital, Cowen) to embed the platform's rate data in M&A due diligence workflows. Approach infrastructure funds (Stonepeak, DigitalBridge, KKR Infrastructure, EQT Partners) as enterprise clients who need rate benchmarking across multi-network acquisition pipelines. Enterprise tier at $5,000/month for portfolio-level analytics covering 10+ networks.
Who Exists Today
| Company | What It Does | Rate Intelligence? | Pricing |
|---|---|---|---|
| GeoTel Communications | Fiber route mapping and network availability database | No: maps where fiber is, not what it costs | Contact sales |
| Vetro FiberMap | Fiber network design and GIS management SaaS | No: operational network planning, not market analytics | $500-2,000/mo |
| 3-GIS (SSP Innovations) | Fiber network engineering and outside plant management | No: manages physical infrastructure, not pricing | Enterprise pricing |
| Cowen Fiber Intelligence | Investment bank fiber sector research reports | Partial: macro market analysis for investors, not route-level rate data | $15,000+/report |
| TeleGeography | Submarine cable and WAN pricing research | Yes for lit services: IP transit, wavelength pricing. No dark fiber rate benchmarking at the local/regional level | $8,000+/yr |
| This startup | Route-level dark fiber rate benchmarking + IRU valuation + demand signals | Core product: anonymized transaction-level rate analytics | $499-1,499/mo |
The competitive gap mirrors what existed in commercial real estate before CoStar, in hotels before STR, and in apartment rentals before RealPage and Yardi Matrix. Every existing tool in the fiber space solves either a mapping problem (where is the fiber?) or an engineering problem (how do I design and manage the network?). Nobody solves the pricing problem: what is this fiber worth, and am I getting a fair deal? TeleGeography comes closest with its lit-service pricing research, but its coverage focuses on international transit, wavelength services, and IP pricing for carriers, not the route-level dark fiber lease rates that regional operators negotiate daily. The $15,000+ investment bank reports from Cowen and similar firms provide excellent macro analysis for PE fund managers but are useless to a 200-employee electric cooperative in Kansas trying to price a 40-mile dark fiber IRU for a wireless carrier.
Why Now
AI infrastructure spending is creating extraordinary demand for fiber on routes that were previously low-value. When a hyperscaler announces a multi-billion-dollar data center campus (Microsoft, Google, Amazon, and Meta collectively announced over $200 billion in AI infrastructure capex for 2025-2026) the fiber demand radiates outward from the site along every viable conduit route within 100 miles. Regional operators who happen to own fiber on those routes are suddenly sitting on assets worth multiples of what they were worth two years ago, but they have no market data to quantify the premium. Zayo is adding 5,000 new route miles specifically for AI-driven demand. BIG Fiber raised $250 million to build metro dark fiber for hyperscalers. The money is moving. The question is whether the regional operators who own the existing fiber will capture their share of the repricing or leave it to better-informed buyers and intermediaries.
The BEAD program, meanwhile, is creating 500+ new fiber network operators simultaneously. The $42.45 billion Broadband Equity, Access, and Deployment program is the largest federal investment in broadband infrastructure in U.S. history. Subgrantees are required to build open-access middle-mile networks and submit sustainability plans that include wholesale rate schedules. These are operators building fiber networks, many for the first time, who must set wholesale rates with zero market data and no historical pricing to guide them. They need rate intelligence from day one, and the BEAD program's timeline (construction beginning 2025-2027) creates a concentrated wave of first-time fiber network operators entering the market within a 24-month window.
On top of all this, infrastructure fund investment in fiber assets has exploded, creating a transaction market that demands valuation tools. Stonepeak's $250 million investment in BIG Fiber, DigitalBridge's fiber portfolio acquisitions, and KKR's digital infrastructure fund are part of a broader trend: private capital views fiber as a utility-like asset with predictable cash flows and long-duration returns. But unlike commercial real estate, where CoStar provides transaction comparables and cap rate benchmarks, fiber transactions happen in an information vacuum. Sellers routinely accept 20-40% below fair value because they cannot independently verify the buyer's valuation model. A rate intelligence platform that can produce independent fair-value estimates for fiber assets shifts the information balance and captures value from both sides of every transaction.
Original Contribution: The AI Route Premium Gradient
An estimate worth examining: Using the CTC Technology rate data as a baseline, we can estimate the magnitude of the AI-driven repricing event for dark fiber on routes proximate to announced hyperscaler data center campuses.
The CTC data shows that metro dark fiber rates in large urban markets range from $120 to $600 per strand per month, with a median of approximately $300. Long-haul rates run $3.44 to $13.75 per strand-mile per month, with a median of approximately $7. These represent pre-AI-boom pricing observed across a range of U.S. markets.
Now consider a 30-mile metro route connecting a regional operator's network to a new hyperscaler data center campus. At the pre-boom median metro rate of $300/strand/month, a 24-strand IRU on this route generates $86,400 in annual revenue. But hyperscalers are now ordering 12-48 fiber pairs (24-96 strands) per route and competing for scarce conduit capacity near data center campuses. If AI-driven demand compresses the supply of available strands on routes near data center clusters by even 30%, basic supply-demand dynamics suggest that rate premiums of 2x to 4x on affected routes are not only possible but expected, consistent with the premium structure observed in Palo Alto (up to $591/strand/month vs. the $300 median) where tech demand has historically concentrated.
Applied to the estimated 200 regional operators who own fiber within 50 miles of the top 30 announced AI data center campuses: if each operator has an average of 15 route-miles of fiber affected by the AI proximity premium, and the premium lifts rates from $7/strand-mile/month (long-haul median) to $14-28/strand-mile/month on a typical 48-strand cable, the incremental annual revenue opportunity across the cohort is $145 million to $290 million. That is revenue these operators are entitled to by virtue of owning fiber on the right routes, and will only capture if they have rate data showing that the market has repriced around them.
Limitations
Several weaknesses in this analysis should be stated clearly.
The CTC Technology rate data, which forms the backbone of the pricing spread analysis, was compiled from a limited set of markets and is itself over a decade old in its original publication. While we have supplemented it with more recent municipal rate schedules (Riverside 2024, Laurinburg 2022) and SEC filing data, the true distribution of current market rates across all U.S. fiber routes is unknown, which is precisely the problem this startup proposes to solve. The 60x pricing spread may overstate the variation that exists within comparable route types (comparing Palo Alto metro to rural Illinois long-haul conflates geography and network type), and the actual spread within a single route category (e.g., metro routes in mid-tier cities) is likely 3-5x, not 60x.
There is also the cold-start data problem. Dark fiber may be fundamentally harder than in the analogous markets (hotels, apartments, marinas) that we cite as precedents. Fiber operators are accustomed to treating their rate cards as proprietary intelligence, and the number of potential data contributors is smaller (1,800 operators vs. 12,000 marinas or 55,000 hotels). The data network effects that make benchmarking products valuable also make them hard to bootstrap. If the first 50 operators who contribute data represent only the smallest and least sophisticated networks, the benchmarks will be systematically biased downward and unattractive to the larger operators whose data would be most valuable.
Third, the AI route premium calculation assumes that hyperscaler demand will translate into rate increases for regional operators, but the actual procurement behavior of hyperscalers often involves direct fiber construction (build, don't lease), long-term dark fiber acquisitions at bulk discounts, or exclusive deals with Tier 1 carriers like Zayo and Lumen that bypass regional operators entirely. The proximity premium may concentrate in the hands of operators who already have conduit access to the data center campus, which is a small subset of all operators within 50 miles. The revenue opportunity could be both geographically narrower and more concentrated than our estimate suggests.
Strongest Counterargument
The strongest case against this startup is that dark fiber pricing opacity may be a feature, not a bug, for the operators who benefit most from the current system, and those operators will actively resist transparency.
Consider the incentives of a Tier 1 fiber operator like Zayo, which controls high-fiber-count routes across major corridors and negotiates hundreds of dark fiber deals per year. Zayo's competitive advantage is partly informational: it knows the market rate for every route it operates because it has transaction data from its own deals across thousands of routes. A regional operator negotiating with Zayo for an IRU on a shared corridor has no comparable data and is therefore at a systematic disadvantage. Rate transparency would erode Zayo's informational edge and shift negotiating power toward the smaller operators.
The same logic applies to the infrastructure funds acquiring fiber networks. Stonepeak, DigitalBridge, and KKR have teams of analysts who model fiber asset values using proprietary transaction data from their own portfolio companies. A seller who can independently verify the fair value of their network is a harder negotiation than a seller who is guessing. These funds have strong financial incentives to keep the market opaque, and they have the resources to discourage data sharing among operators they are courting for acquisition.
This isn't a conspiracy theory; it is standard information economics. The informed party in any market benefits from opacity. In commercial real estate, CoStar succeeded despite similar resistance because the market was large enough that transparent pricing created more total value (by expanding transaction volume) than it destroyed for any single participant. The question for dark fiber is whether the market is large enough for the same dynamic to hold, or whether a small number of sophisticated actors can maintain opacity because the total number of participants is manageable. With 1,800 operators vs. the hundreds of thousands of commercial real estate participants, the answer is genuinely uncertain.
There is also a legal dimension worth noting. RealPage, a rental rate benchmarking platform that operates on the same informational principle as this proposed startup, is the subject of a DOJ antitrust lawsuit (filed December 2024) alleging that its algorithmic pricing recommendations enabled landlords to tacitly coordinate on higher rents. Rate intelligence platforms that show participants where the market distribution sits can have the unintended effect of anchoring prices upward. Any dark fiber rate benchmarking product would need to be designed with antitrust compliance from day one.
Risks and Challenges
- Data contribution resistance: Operators may refuse to share transaction data even anonymized, particularly if they believe their rates are below market and don't want competitors to know. Mitigation: the public data layer (municipal rates, PUC tariffs, BEAD applications, SEC filings) provides a useful baseline even without operator contributions, and the BEAD cohort creates a natural seed group of operators who are required to publish wholesale rates anyway
- Regulatory complexity: Fiber lease rates are regulated differently across states: some are tariffed, some are market-based, some are subject to open-access requirements. The platform must account for these regulatory regimes without providing legal advice. Mitigation: partner with telecom regulatory attorneys and clearly scope the product as market intelligence, not compliance guidance
- Route comparability challenge: Unlike hotel rooms or apartment units, fiber routes are not standardized products. A 10-mile metro route in Denver is not directly comparable to a 10-mile metro route in Dallas because of differences in conduit congestion, competitive alternatives, anchor tenant proximity, and construction difficulty. The benchmarking engine must account for these variables or produce misleading comparisons. Mitigation: develop a route scoring methodology that normalizes for geography, competition density, and demand drivers, similar to how commercial appraisers adjust comparable property values for location-specific factors
- Market timing risk: If AI infrastructure spending decelerates or data center construction slows due to power grid constraints, permitting delays, or a pullback in hyperscaler capex, the urgency driving the AI route premium will diminish. The underlying rate intelligence need persists regardless of AI demand, but the premium pricing and urgency that would accelerate adoption may be cyclical
The Bottom Line
Dark fiber is a $1.2 billion U.S. market growing at 7.65% annually, undergoing a repricing driven by AI infrastructure demand that is compressing available capacity on key routes and pulling institutional capital into fiber assets at a historic rate. Lease rates vary by 60x across the country, IRU transactions worth hundreds of thousands of dollars are priced without market data, and the operators who own the underlying fiber have no tools to determine whether they are capturing fair value. Every comparable infrastructure market (hotels, commercial real estate, apartments, self-storage, marinas) has developed centralized rate intelligence. Dark fiber has not. The data cold-start problem is harder here because the participant pool is smaller and the incumbents benefit from opacity, but the BEAD program is simultaneously creating 500+ new fiber operators who need rate intelligence from day one and are conditioned by their grant requirements to share wholesale pricing. The timing window is narrow: BEAD construction starts in 2026-2027, hyperscaler fiber procurement is happening now, and the operators who don't know what their strands are worth in the next 24 months will lock in below-market IRU terms that persist for two decades.
What You Can Do
If you operate a regional fiber network: pull your last 10 dark fiber lease transactions and calculate the effective rate per strand-mile per month for each. Then request the published dark fiber rate schedules from three municipal fiber networks in your state (most are available through public records requests or published online). Compare your rates to theirs. If your rates are more than 20% below the municipal rates, which were set by government staff rather than market participants, you are almost certainly underpricing your private-sector leases, where the buyer's willingness to pay should be higher. If you are a BEAD subgrantee designing your wholesale rate schedule: do not set your middle-mile rates based solely on the cost model in your grant application. Those cost models were designed to demonstrate financial sustainability to the NTIA, not to optimize revenue. Contact three operators who sell middle-mile dark fiber in adjacent markets and ask what they charge per strand-mile. You will find that the range is wider than you expected, and the "market rate" your consultant suggested may be at the bottom of it. If you are an infrastructure fund evaluating fiber acquisitions: the seller's inability to independently value their assets is a feature of your deal pipeline today, but it creates adverse selection risk. The operators who accept your first offer without pushback are disproportionately likely to be the ones with low-quality networks and uncertain demand. A platform that gives sellers better information may raise prices, but it also increases deal flow from higher-quality operators who currently won't engage because they suspect they are being undervalued.
Related
📰 Cell Tower Lease Rate Intelligence SaaS — the same pricing opacity problem applied to wireless tower leases, another telecom infrastructure asset where landowners routinely undervalue their positions
📰 Billboard Yield Intelligence SaaS — rate benchmarking for another physical infrastructure asset class where pricing is opaque and fragmented across thousands of operators
📰 Broadband Permit Tracking SaaS — the regulatory and permitting layer that governs fiber construction, a complementary product for the same buyer persona