A Maravedis analyst's view on the Q1 2026 reckoning for fixed broadband, and why the economics of multifamily are quietly becoming the most interesting line in the deck.
- Q1 2026 broke US broadband: Charter's stock halved on flat ARPU and 120K subscriber losses, Optimum lost 64K, and the network no longer wins customers now that speeds have converged.
- Convergence is mostly a discount that defends households without lifting the revenue ceiling, and even fiber is fragile, with take rates stuck at 45 to 47 percent.
- MDU bulk inverts the math: one multi-year contract replaces hundreds of churning subscriptions, and Smart Building as a Service turns flat per-user broadband into an expanding per-account revenue line.
The quarter that broke the spell
If you wanted a single image to capture where US fixed broadband sits in mid-2026, it is the Charter chart. The day before its Q1 2026 earnings, Charter traded around $238. In a reporting session, it was near $149, and over 12 months, the stock is down more than half. The trigger was not a revenue miss. Revenue was roughly flat at about $13.6 billion. The trigger was two words Wall Street has decided it can no longer tolerate: subscriber losses and flat ARPU.
Charter shed roughly 120,000 broadband subscribers in the quarter and guided residential broadband ARPU to stay flat through 2026, sitting around $70.72. CEO Chris Winfrey called the market reaction disappointing and argued the company has never run for short-term ARPU, pointing instead to customer lifetime value and converged metrics. The market did not care about the nuance. It cared that the growth engine of US cable for two decades, cable and broadband in general, is now in a defensive position.
Optimum, formerly Altice USA, reported a similar pattern on a smaller scale. It lost 64,000 broadband subscribers in Q1 2026, or about 56,000 excluding a one-time adjustment, worse than the roughly 47,000 analysts expected. New Street Research said it would take a "monumental effort" for Optimum to perform better in the second half of 2026 than it did in the second half of 2025. The company has about $7.4 billion in debt due in 2027 and is already restructuring assets to protect value. One notable bright spot was ARPU: broadband ARPU was about $74.95, higher than at Comcast and Charter, while converged ARPU was about $79.32. Optimum has begun reporting that metric because per-subscriber figures no longer capture the full picture.
A two-front problem with no easy front
The structural picture is not a cable problem or a telco problem. It is an access-layer problem with two sides pulling against each other.
On volume, the trend lines are set. PwC has modeled US cable losing on the order of 6.5 million subscribers by 2029, while fiber connections grow by more than 50 percent and fixed wireless access surges by close to 77 percent. Fixed wireless from T-Mobile and Verizon, and fiber overbuilders almost everywhere, have turned the single-family home into a contested account that did not used to be contested.
On price, the ceiling is just as real, and the deeper problem is that the network itself has stopped being a differentiator. On the latest Ookla data, the fastest US network runs at around 300 Mbps while the others sit at 200 to 220 Mbps, a gap that is real in the lab and invisible in daily life. When 200 Mbps already loads every app and streams every video, the customer experiences no problem to solve, and the tens of billions in annual network capex buy parity rather than separation.
That has a direct revenue consequence. The drivers of switching have shifted from the network to everything around it: billing clarity, support, and value perception. The category's net promoter scores are brutal, with telecom near 14 globally, compared with digital services in the 30 to 70 range, and US consumer scores of 82 for T-Mobile, 37 for Verizon, and 20 for AT&T. The most uncomfortable finding for an industry chasing ARPU: the highest-ARPU customers carry the highest churn risk, because the people paying the most expect the smoothest experience and leave when they do not get it. The revenue base and the flight risk are the same people.
So the two-front problem is really one squeeze. Volume is contested, price is capped, and asset operators keep spending on the network; it no longer wins the customer.
Even fiber's economics are a knife-edge, one home at a time
It is tempting to read all of this as a "fiber wins, cable loses" story. The more useful read, and the one that sets up the multifamily argument, is that even fiber economics are fragile when fought house by house.
US fiber now passes more than 100 million locations and reaches above 60 percent of households, but take rate is the pressure point, stuck around 45 to 47 percent. More than half of the homes passed generate no revenue at all. Capital per passing runs from roughly $800 in dense suburban builds to well above $2,000 in complex ones, and $6,000 to $10,000 in rural terrain. Around 16 percent of locations now sit in overbuild zones where two or three networks chase the same household, and once take rate drifts below roughly 35 percent, a large share of those builds cannot clear their cost of capital.
The framing that matters: the land grab for "homes passed" has become a knife-edge fight for "homes captured." Capital markets have noticed. Subscale platforms without density or bundling are clearing at mid-single-digit EBITDA multiples while scaled, bundled operators hold low to mid-teens. The market is no longer paying for a footprint; it is paying for certainty of monetization. Hold that phrase. It is the entire case for multifamily.
Convergence is the reflex, but mostly it is a discount
The playbook the operators reached for is convergence. Bundle fiber or fixed wireless with mobile, drop churn, lift lifetime value. On April 1, AT&T launched OneConnect, collapsing fiber and mobile into a single household subscription at flat prices of $90, $120, and $225. The behavioral logic is sound: roughly 40 to 45 percent of large-operator fiber customers already bundle mobile, more than 70 percent of users want a single bill, and converged households show both higher retention and higher revenue per account. Verizon is running the same play under Dan Schulman, explicitly pivoting from "lines" to converged accounts, managing to customer lifetime value at the household level, and citing converged households that churn about 30 percent less.
The problem is what convergence does to price. As Craig Moffett put it, convergence is an elevated name for discounts. You make the household stickier by making the relationship cheaper. The same conclusion holds on the architecture side: collapsing the subscription is the easy part, and OneConnect is a simplification of the commercial interface rather than a transformation of the underlying system. Pure commercial convergence is good but not a long-term magnet. It defends the household. It does not lift the ceiling.
That is the gap. Operators have correctly identified that the household, not the line, is the right unit. They have not yet found a household where the bill grows instead of shrinks.
Where the math actually changes: the building as the account
This is where multifamily stops being a niche and becomes the most rational answer on the table, because the MDU bulk model inverts almost every term in the single-family equation.
One signature replaces hundreds of sales motions. A property-level bulk agreement converts a building of individually churning subscribers into a single account on a multi-year contract. The take-rate knife-edge that defines single-family fiber, where more than half the homes passed pay nothing, largely disappears, because a bulk building delivers near-total penetration by design. Acquisition cost per door collapses, and revenue visibility extends years rather than billing cycles. For operators whose entire valuation problem is the unpredictability of net adds, contracted multi-year revenue across a portfolio of buildings is a different financial instrument, and it is exactly the "certainty of monetization" capital markets now pay for.
Churn moves from the resident to the asset. In a bulk arrangement, the customer is the property owner or operator, and connectivity is embedded in the operating model. Residents turn over; the account does not. That is the structural opposite of the single-family treadmill, where every move-out and every competitor promotion reopens the sale.
The account, not the user, becomes the unit of revenue. A converged single-family household is worth an estimated $2,200 to $3,000 per year, with convergence cutting churn 20 to 30 percent. An MDU operator gets the same household-level economics, but with contracted near-full take and a fraction of the per-door acquisition cost. And the deeper structural break now visible in fiber applies with full force here: ownership of the infrastructure and ownership of the customer have split. Whoever holds the property contract and the resident relationship captures the value, regardless of who owns the glass in the ground. That is precisely the position a managed-connectivity operator occupies in a building.
It is worth noticing that the converged-ARPU and converged-ARPA metrics the big operators have suddenly started reporting- Optimum at $79.32, Comcast near $85, Charter calling it the more relevant number- are all attempts to move the conversation to the account level. The MDU market lets them do it natively, because a building is an account, not a subscriber.
From ARPU to ARPA: Smart Building as a Service
The reason multifamily is more than a churn-defense tactic is what sits on top of the connectivity layer.
Once an operator owns the property-wide network, connectivity becomes the platform, not the product. The same infrastructure that delivers managed Wi-Fi also carries access control and smart entry, energy and HVAC management, leak and environmental sensing, security and video, amenity-space connectivity, and the resident app that ties it together. Vendors are already building the operating layer for exactly this purpose, with portfolio-wide views of network performance, device behavior, and the resident experience. Our PropTech Evolution in U.S. Multifamilyreport maps this exact shift, from fragmented smart-apartment point solutions to integrated building-intelligence platforms that unify managed Wi-Fi, access control, HVAC and energy, IoT sensors, and AI-driven resident services. The industry shorthand is Smart Building as a Service, and it is the mechanism that turns a flat per-user broadband price into a layered, expanding per-account revenue line.
This is the deployable version of the "network as a product" idea, now gaining currency across the telecom industry. The core argument is that operators have to stop selling speed and volume, because that path leads only to flat revenue, and start selling capabilities as distinct products: latency, SLAs, telemetry, multi-tenant slices, each with their own price point. A multifamily property is the cleanest environment in which to actually do that. It is a contained, account-level setting where the network anchors a stack of billable services, and where the buyer, the property owner, is motivated by something other than megabits. Our own Multifamily Rental Connectivity Market Analysis 2026-2031sizes the managed-service-provider opportunity in US multifamily growing from $8.9 billion in 2025 to $13.7 billion by 2031, and weights Financial Strength most heavily in the Maravedis Market Score precisely because the contracts that anchor this market typically run seven to ten years. That is the contracted, account-level revenue the single-family market cannot match.
For that owner, the appeal is net operating income. Reliable managed Wi-Fi has been associated with materially lower resident turnover, on the order of 8 to 15 percent, and bulk-plus-smart-building arrangements can shift connectivity from a cost center to an income-producing, revenue-shared asset that improves valuation and refinancing math. For the operator, that owner incentive is the doorway to a contracted, multi-service relationship with pricing power the standalone broadband market no longer offers.
This is the honest version of convergence. Not a discount to keep a household, but an expanding bill of services to a property. Commercial convergence removes surface-level friction, but the durable value lies in the layers above the pipe. In a building, those layers have names, budgets, and a buyer.
The Maravedis read
The strategic logic is clean enough that the gap between it and execution is the real story for the next eighteen months. Three things separate the operators who will capture this from the ones who will simply say "converged ARPU" on earnings calls.
First, the physical layer. Most older buildings lack the in-unit wiring to deliver seamless property-wide service, and that infrastructure gap, not demand, is the binding constraint for roughly one-third of US households living in multi-dwelling units. Second, the platform and the relationship. Owning the smart-building operating layer, the data, and the resident relationship is what defends margin, and that is a software and go-to-market competence cable and telco operators have historically outsourced. The consolidation now underway reinforces the point: capital is rewarding the players who can aggregate demand and own the customer rather than those who simply own footprint, and open-access structures like the AT&T and BlackRock Gigapower joint venture are splitting infrastructure ownership from customer ownership outright. An MDU operator is the most concentrated form of demand aggregation in the market, one contract at a time. Third, regulation. Bulk billing economics are under active scrutiny, from California's AB 1414 to Colorado's bulk Wi-Fi rules, and a model that delivers service to residents well below retail is exactly the model that draws policy attention. Any operator betting on multifamily is also betting on the policy outcome, and needs to build that into the thesis.
None of that changes the core conclusion. The single-family broadband business is now a defensive, flat-ARPU, high-churn, take-rate knife-edge that the market has stopped rewarding, and the network that operators keep funding no longer wins customers. The multifamily business offers the one combination that the rest of the access layer cannot: long-term contracted revenue, account-level rather than user-level economics, and a service stack that increases the bill rather than discounts it. The operators bleeding subscribers in Q1 2026 already know this. Watch how fast they move from saying "converged ARPU" to actually building for the account.
Maravedi's research and recognition
For the full data behind this analysis, see two Maravedis reports:
• Multifamily Rental Connectivity Market Analysis in the United States 2026-2031. Market projections, competitive intelligence on 20 managed service providers, SWOT assessments of five Wi-Fi infrastructure vendors, and the proprietary Maravedis Market Score, with the state-level regulatory landscape (California AB 1414, Colorado markup caps, New York Affordable Broadband Act) mapped in detail.
• PropTech Evolution in U.S. Multifamily: Building Intelligence, Connected Systems and Market Outlook 2026-2031. The full PropTech stack from connectivity foundations and MSP business models through access control, HVAC, and energy, smart-unit platforms and interoperability, value-added services and AI, with segment-level forecasts through 2031.
The operators and vendors building the account-level future of multifamily connectivity are also the ones recognized through the Maravedis MDU Connectivity Awards, which honor the most significant innovations and deployments across the multifamily connectivity ecosystem, with winners announced at a dedicated ceremony in Q4 2026.