Andrej Danis
Partner & Managing Director - New York
The opportunity in North American fiber is hiding in plain sight. More than 400 independent operators have built the networks that large carriers would like to own, but as capital gets more expensive and subscriber growth slows, the economics that financed those builds have turned against the operators holding them.
On paper, the conditions for consolidation appear ideal. Yet thus far, deals are not surfacing. The reason: Anyone can buy fiber, but almost no one can integrate it. Buying that which you cannot run effectively is a waste of money.
Assimilating an acquired fiber network is a slow, costly, and stubbornly bespoke process. Small operators have fewer moving parts and a cleaner footprint extension; the catch is volume. You have to acquire many small networks, and that only pays off if each integration is cheaper and faster than the last.
We believe a different approach is the solution to the problem. A standardized, three-month integration model can turn fiber M&A from a string of one-off projects into a systematic, replicable program. The state of a target’s network and its data decide whether this approach functions well, yet most diligence misses them. Below, we lay out how to determine your ability to absorb an acquisition target before you put pen to paper.
Three forces hold back the conventional approach
Each acquisition target has its own operational and business support systems (OSS and BSS), its own network vendors, its own inventory (with uncertain quality), and its own customer acquisition process. The conventional approach treats each of these as a novel problem to solve from scratch. Three forces keep that approach uneconomical at volume:
Diligence that does not scale: When network and product diligence are customized for each target, the cost per deal barely falls as the deal count rises.
Integration costs anchored to the target: When buyers absorb each acquired OSS / BSS, every integration becomes a custom systems-mapping exercise.
Timelines measured in years: Reconciling overlapping data models, systems, and vendors stretches each integration well past a year, delaying synergies and the next deal.
The cumulative effect is that integration costs never grow cheaper, and a model that does not grow cheaper or faster with repetition cannot underwrite a consolidation strategy.
Three requirements to successfully absorb a target
Buyers start by standing up a single, standardized tech stack and a cloud-native OSS / BSS. Target legacy systems are not integrated; they are switched off. Built on an open, standards-based network architecture, the clean stack is reused for every deal, so the endpoint for each integration is decided before targets are even identified. This cuts the most expensive part of integrations—the bespoke mapping of source systems into the buyers’ own—into a repeatable migration.
But before any of this works, the buyer must be ready to receive an acquisition. Carriers that have grown through decades of organic building and prior acquisitions carry layers of legacy architecture that were never designed to receive new networks at volume. The product catalog alone—accumulated plans, grandfathered pricing, regional variations—can be as complex on the buyer's side as anything found in a target. That means a rationalized product catalog, a defined OSS / BSS target architecture, and clean data infrastructure on the receiving end. Most diligence focuses entirely on the target. The harder question is whether the buyer's own systems can absorb what they are about to buy.
Three elements must be in place from the onset:
Networks merge seamlessly: Operators are wiring their control plane into someone else’s physical plant while it carries live traffic; the worst window is when both the old and new provisioning tools can reconfigure the same optical line terminal (OLT). Get that wrong and you risk massive scale problems.
Invoice accurately: From the first billing cycle post-integration, you need to bill the right amount to the right customer on their correct plan. An incorrect first bill at scale is a major cause of churn.
Ensure support has the answers: From day one, care agents need access to customer plans, charges, and the state of their networks across systems. If the agent is missing any piece, problems turn into complaints.
Implemented correctly, the business runs as one company within a quarter. The factory launches the clean stack before deal close, ingests the target’s network and customer data, routes new orders to the new stack, and migrates existing customers in waves. The long tail, any stray records, and legacy systems winding down clear through normal operations over the following months and never hold the organization hostage. Suddenly a multi-year, custom project becomes a sequenced, replicable drill.
| Dimension | Conventional integration | Clean-stack integration |
| Destination | The buyer's existing systems, mapped per target | One standardized OSS / BSS, reused on every deal |
| Team | New project team assembled per acquisition | Standing integration factory that compounds experience |
| Timeline | 12+ months, with prolonged dual-running | ~90 days, clean cut-over by workstream |
| Cost curve | Flat: Deal 40 costs about the same as deal 1 | Declining: Each migration reuses the last |
| Diligence | Run from scratch post-signing | Templated, begins before the deal is signed |
Why this is the unlock the industry needs—and how AI helps
Cost synergies are only half of the return. The larger half is revenue, and a sub-scale fiber base is almost always leaving it on the table. Take the acquisition of Lumen by AT&T. The acquired network runs near 25 percent penetration, against roughly 40 percent on AT&T's own fiber, and closing that gap is the first prize. The larger prize comes later, from pushing penetration past that benchmark.
Beyond penetration, companies that properly integrate new networks have ample opportunity to move customers up speed tiers, entice them with add-ons, and sell bundled services that raise the ARPU across their base. This is more than buying fiber; it's about buying homes you can sell (or sell more) to.
AI earns its place in managing the data. A regional operator's records used to be kept by hand, replete with undocumented plans, mismatched addresses, and inventory that did not match the network. Modern AI reads that mess, maps it onto your stack, and flags the records that do not add up, so your experts spend time on the exceptions, not the routine. And with no need to wait for close, you can start on the diligence data right away, and months of data work compress into weeks.
The larger opportunity is in execution. Agentic workflows embedded across data migration, system integration, cutover validation, and day one billing activation are what transform a three-month integration target from aspiration into repeatable reality. The continuous improvement loop compounds with every deal. Each migration makes the next one faster and cheaper.
The prize was never the fiber; it’s building the capability to take over a network and have it provisioning, billing, and supporting customers from day one. That capability starts with the buyer's own house. Carriers that industrialize the integration process and embed AI in the execution, not just the diligence, will set the terms of consolidation for all others that follow.
