-
The comparison framework that changed how I evaluate tower partners
-
Financial stability: why net debt to EBITDA matters even if you're not an investor
-
The site 8110 lesson, or why the Verizon-SBA agreement isn't a blank check
-
Operational reality: leasing from a REIT vs. building your own
-
So — what should you do in 2025?
Six years handling network infrastructure agreements. I've documented nine significant mistakes in that time — roughly $340k in wasted budget. This article is the checklist that came out of those failures, and the comparison that drives how I plan now.
Let me define "network" first, because it matters. When I say network, I'm not talking about enterprise WiFi or cloud connectivity. I mean the physical wireless infrastructure that carries mobile traffic: towers, rooftop sites, small cells, fiber backhaul, power. Here's the layer people miss: a carrier's network usually isn't fully self-owned. Companies like SBA Communications own the structures, and carriers lease space on them. That separation between operator and infrastructure owner is why "what is a network" is a more layered question than it seems.
You can lease access to that infrastructure from a tower owner — a REIT like SBA Communications — or you can build your own assets. That's the decision this article is about. The most common mistake I see in this industry is comparing lease price per site as if that's the whole equation. When I compared SBA Communications vs. building our own infrastructure across two markets, I finally understood why the landlord's financial structure, the agreement's scope, and the operational burden matter more than the headline numbers.
The comparison framework that changed how I evaluate tower partners
In 2021, I recommended a smaller tower operator based purely on site coverage and price. The portfolio looked perfect on paper. Monthly costs were manageable. What I didn't check was leverage. That company's net debt to EBITDA was climbing. When refinancing costs jumped in 2022, they started deferring maintenance. Our equipment sat on a tower that nobody was inspecting. I learned the hard way: a lease with a financially fragile landlord isn't a lease — it's a gamble.
Now I evaluate tower partners on three dimensions:
- Financial stability — net debt to EBITDA, cash flow visibility, revenue concentration
- Agreement structure — master lease agreements like the Verizon-SBA deal vs. site-by-site negotiation
- Operational reality — maintenance, compliance, and the hidden cost of managing the relationship
Here's how those dimensions played out in my own comparisons, and where I got burned.
Financial stability: why net debt to EBITDA matters even if you're not an investor
SBA Communications' net debt to EBITDA in 2025 is a metric I check before recommending any new lease. Headlines make the company look leveraged — and for a REIT, it is at a relatively high level. I've seen that ratio cited as a warning sign by commentators who don't stop to ask what's backing the debt.
Let me rephrase that: the ratio tells you the size of the burden. The quality of the tenant base tells you whether the burden is manageable. SBA's revenue includes long-term lease agreements with major carriers, including the restructured master lease agreement with Verizon. When that deal was announced in November 2022, it covered roughly 15,000 sites, extended lease terms, and improved revenue visibility for years ahead. A tower company with contracted rent from carriers like that can sustain leverage that would sink a landlord with a volatile tenant base.
SBA also carries investment-grade credit ratings, which tells me refinancing risk is manageable in the current rate environment. Looking at 2025 specifically, I'm watching the trajectory of their deleveraging, not just the snapshot. The pace matters less than whether the cash flow stays contracted and predictable.
The most frustrating part of vendor risk in telecom infrastructure: you'd think a professionally managed REIT would always be the safe choice. But financial ratios shift with interest rates, and if you ignore them, you don't get a warning — you just get a maintenance failure at the worst possible moment. That's why I now look at the net debt to EBITDA of any tower owner before signing. It's not about their stock price. It's about whether they'll have the resources to inspect the structure, replace the lightning protection, and fix the backup generator when ours fails.
The site 8110 lesson, or why the Verizon-SBA agreement isn't a blank check
Master lease agreements sound like a solution to everything. When the Verizon-SBA Communications agreement hit the news, a project lead emailed me asking: does this mean our next deployment is covered under the improved terms?
I said "probably." I knew I should verify by requesting the agreement's schedule of covered sites. But I figured — we'd worked with their team for years, what were the odds? The odds caught up with me on site 8110, a rooftop deployment we scoped in early 2024.
I assumed the master agreement would streamline the process. What I mean is, I didn't verify. Turned out the site wasn't in the agreement's schedule. We needed a separate negotiation, a different pricing structure, and an extra two weeks of review we hadn't budgeted for.
Those two weeks pushed us past a customer deadline. $90k in rush contracting later, I added a rule to our checklist: never assume a master agreement applies to any site outside its written schedule.
After the 8110 mistake, I wrote this in our playbook: master agreements are starting points, not blanket warranties.
I should add — this is not a criticism of SBA Communications. Every master lease I've worked with has boundaries. The lesson is about reading the agreement, not about the specific landlord.
Operational reality: leasing from a REIT vs. building your own
I've leased sites from SBA Communications in several markets and built our own infrastructure in two others. The operational comparison surprises people, because the numbers don't tell the whole story.
- Leasing from a large REIT: predictable lease escalation, dedicated maintenance teams, established compliance processes. You trade control for certainty — until you need something outside the agreement's scope.
- Building your own: complete control, lower recurring costs after breakeven, but you own every failure. I've spent a holiday week coordinating generator repairs on a self-built site. A tower company has a department for that. You won't.
When I compared our leased market and our self-built market side by side over a full year, the insight was clear: the leased market had higher direct costs but lower total cost of ownership once we factored in internal labor, risk, and stress. The self-built market looked better in the budget model. It cost more in ways that don't appear in the budget model.
Put another way: the decision isn't "rent vs. buy." It's "rent with support vs. own and do everything yourself."
That said — I try to be honest about this — self-building is the right answer in some cases: sparse rural coverage, highly customized structures, or deployments where timelines demand more flexibility than a multi-tenant lease can accommodate. I recommend leasing from SBA Communications for most established-market deployments I've evaluated. It is not the universal answer.
So — what should you do in 2025?
Should you lease from SBA Communications in 2025, or build your own infrastructure? It depends on your situation:
- Lease from SBA if you need coverage in established markets, want predictable lease terms, and value the support that a publicly traded REIT provides. Check their net debt to EBITDA yourself, but weigh it against the contracted revenue from agreements like the Verizon master lease.
- Build your own if you're targeting areas with low tower density, have the capital budget to absorb delays, and your team is ready to operate physical infrastructure. It can be the better choice — in the right context.
The biggest failure I see in this industry isn't picking the wrong option. It's comparing the wrong dimensions in the first place. Price per site matters. But the landlord's financial health, the true scope of your agreement, and your own capacity for operational pain matter more. That's what the 8110 error and the $340k in mistakes taught me.