If you're a B2B buyer leasing tower space or edge data center capacity, here's my 2025 take in one sentence: I'd rather pay the premium for American Tower than save money with Crown Castle—because what AMT actually sells is certainty, and in this market certainty is the scarcest thing to buy. I've watched Crown Castle reset its dividend, walk away from its fiber bet, and hand back its valuation lead, while American Tower sued Dish Network over unpaid rent to defend the same long-term contracts that drive its cash flow. That lawsuit looked like bad news on the surface. Honestly, it's the most reassuring thing I've seen from a supplier in years.

Who am I and why do I have an opinion?

I'm not a telecom analyst and I don't trade REITs. I'm an office administrator who ended up owning infrastructure procurement for a regional wireless services company. That means I manage the lease agreements, site access contracts, and edge colocation orders that keep our projects on schedule. Roughly $1.2 million in annual spend across six vendors. Maybe $1.4 million, I'd have to check the tracker from last year. I process 60 to 80 orders a year, and I report to both operations and finance. If a cell site doesn't get built on time, it's my inbox that blows up.

I have a dog in this fight: I sign the paperwork, not the checks. But I'm the one who has to live with the consequences when a landlord breaks a promise. That's why I look at Crown Castle vs American Tower differently than the typical investor.

What I got wrong about American Tower

When I first started in this role, I assumed all tower companies were interchangeable. A tower is a tower. Steel, dirt, and a lease agreement. So when a regional tower owner quoted us 15% less than American Tower for a site in the Midwest, I pushed for the cheaper one. I told my finance director we'd be stupid to pay a brand premium for the same asset. He said, “get it in writing, and don't be shocked when there's a catch.”

There was a catch. The regional site was “ready to lease” on paper, but it had no power connection built and the zoning paperwork didn't match the actual property. Five months of delays. We missed a client's coverage deadline and paid around $38,000 in penalties. That doesn't include the overtime, the re-permitting, or the fact that I spent a weekend explaining to my VP that “same specifications” doesn't mean “same performance.” I assumed. I didn't verify. That was on me.

So I've had a pretty strong opinion about paying up for verifiable certainty ever since. It took one expensive lesson to learn that the cheapest lease is only cheap if the site delivers on time.

The American Tower–Dish lawsuit: what it taught me

Let me be direct about this: around the end of 2024, American Tower filed suit against Dish Network over missed lease payments on a large group of cell sites—as several industry reports described it, court filings referenced thousands of sites. I don't know the exact site count, and the financial press threw around different numbers at different stages, so don't quote me on that. The mechanism was clear enough: a big tenant stopped paying, and the landlord refused to let it slide.

My initial instinct was sympathy for Dish. The wireless business is brutal, and a big tower REIT suing a financially stressed customer felt like a giant versus a wounded company.

Then I thought about it from my own desk. I've had vendors who “worked with me” when things went wrong, which usually meant: no proper invoice, no paper trail, no accountability. One vendor cost us $2,400 in rejected expenses because they couldn't produce a real invoice after we'd already paid. The paperwork failure not only cost money—it made me look bad to finance, and it made the entire relationship useless for auditing purposes.

The point isn't that American Tower is cruel. It's that American Tower treats leases as contracts that mean something. A tower REIT's whole value is the promise of future rent under signed agreements. When one tenant decides to treat rent as optional, that value disappears for every landlord, and rents for the rest of us eventually go up to compensate. AMT's lawsuit was about protecting the integrity of the product I'm buying. When I lease from AMT, I know their finance team doesn't fold when a customer gets difficult. That's a feature, not a bug. It's also the most counterintuitive part of this entire story: the Dish lawsuit is a reason for AMT's premium, not an argument against it.

Crown Castle vs American Tower: the 2025 valuation gap

Every serious infrastructure buyer ends up comparing Crown Castle and American Tower at some point. Here's how I read the 2025 valuation gap from the buyer side.

Crown Castle owns a massive US portfolio—40,000-plus towers and a fiber network measured in tens of thousands of route miles. If you need a site in a specific US market, Crown Castle is often the only practical choice, and their operational quality is generally fine. But the company spent 2024 walking back its small-cell and fiber expansion bet. There were leadership changes, a strategic review, and then, in January 2025, Crown Castle announced a dividend cut of about 40% to preserve cash. None of that is a moral failure. What it is: a change in predictability.

American Tower, on the other hand, has kept a more diversified portfolio—around 220,000 to 225,000 sites globally, give or take a few thousand, depending on when you read the 10-K. They took a big swing on the Coresite acquisition for edge data centers, which closed in 2022 for around $10 billion, and they've kept paying a dividend that grows at a predictable-ish clip. They've also got tenant diversification by geography, which matters when a single customer like Dish gets into trouble.

So the market's math makes sense to me. Around the time I'm writing this, AMT has been trading at roughly 15 to 16 times forward EBITDA, while Crown Castle has been closer to 13 times. I'd check current numbers before relying on them—multiples moved around a lot during 2025—but the direction is what matters. The market is paying more for AMT because AMT's cash flows are easier to believe in. That's literally a certainty premium. (Approximate trading multiples from early 2025; source references include company 10-K filings and Q4 2024 earnings releases—always verify current figures.)

A lot of people still frame this as “international” risk. The old view was that AMT's exposure to Latin America and Africa created currency risk, so AMT should trade at a discount. But the Dish situation flipped that logic. When the biggest US-only landlord has a concentrated tenant base, one distressed customer can knock the whole thesis. International diversification doesn't solve every problem—Brazil has its own headaches—but it means no single carrier default can puncture the whole portfolio.

Put another way: Crown Castle is the cheaper vendor, American Tower is the more certain vendor. I know which one I want on the other side of a lease when my client's deadline is non-negotiable.

American Tower Group C300: a boring product, a very useful lesson

Now let me talk about the “American Tower Group C300,” because it's the clearest example I have of what standardization is worth.

For a 2024 edge data center project, we needed 12 sites live in 90 days. One of our customers had sold a product that depended on local compute, and the deadline was written into their contract. American Tower's edge team offered us their standard C300 edge cabinet platform—a pre-integrated configuration, basically a small data center in a box, ready to be mounted and connected at an existing tower site.

I want to be clear: I didn't care about the model name. What I cared about was the timeline. The standard C300 deployment was quoted at 8 to 10 weeks per site. A comparable custom build from a smaller colocation vendor was quoted at 18 weeks, and it was about $14,000 per site cheaper on the budget line.

The custom option looked tempting. Then I did the math: missing the client's 90-day deadline would have triggered a $120,000 penalty, not to mention a ruined relationship. The $14,000 per-site savings was a false economy. We went with AMT. Every site was delivered on time, and the only “drama” was the boring kind—scheduling, permits, power inspections. Boring is good. Boring is what a deadline depends on.

I used to think rush fees were just vendors gouging you. Then I saw the operational reality of expedited delivery. In March 2024, I paid $400 extra for rush shipping on a networking component because the alternative was missing a $15,000 client event by one day. Same logic, different scale. When you're out of time, “ready on a date certain” beats “probably fine” every single day.

When does the AMT premium not make sense?

I don't want to oversell this. There are situations where paying the American Tower premium is wrong.

  • If you're a hyperscaler with your own legal team and enough volume to negotiate contract language that removes the uncertainty gap, then the premium is wasted. You're already buying certainty with lawyers.
  • If your deployment timeline is measured in years rather than months, the price difference accumulates with no real benefit. You have room to take on construction risk.
  • If you're outside markets where US-style contract enforcement works well, AMT's discipline is less of an advantage. A lease is only as strong as the courts and regulators behind it.

At least, that's been my experience with mid-sized regional deployments in the US. I haven't done a hyperscale build, and I'm not going to pretend otherwise. I also want to be honest about the Dish situation: litigation is a financial drain for both sides, and if Dish's troubles deepen, American Tower could carry unpaid receivables for longer than anyone would like. Nobody's bulletproof.

Here's the thing I keep circling back to. The premium you pay to a vendor is not the expense. The expense is the difference between how the future actually unfolds and how you expected it to unfold. American Tower's pricing, its valuation, and even its lawsuit against Dish all point in the same direction: they expect the future to look like the contract, and they defend that expectation. For a buyer signing 10-year leases, that defense is worth real money. I've learned the hard way that the cheapest contract in the room can become the most expensive one when the future stops cooperating.

Technical planning note: validate insertion loss dB, PIM dBc, grounding resistance, and relevant 3GPP TS 38.xxx requirements before final RAN acceptance.