Uniti Group Inc. Q1 FY2025 earnings call

Management Statement and Operational Highlights:

• Business Performance: Uniti had a strong quarter with core recurring strategic revenue growing approximately 4%, and capital intensity of fiber business (excluding GCI) declining over 50% year-over-year.
• Merger with Windstream: Received shareholder approval for the transaction, with 16 of 18 PUC approvals, on track to close in H2 2025. Kinetic expects to double fiber homes passed in 2025, reaching 2 million homes by end of 2025, 2 years earlier than expected.
• Regulatory Trends: Encouraged by FCC/NTIA changes on copper network retirement and government subsidies for fiber deployment.
• Cost of Capital: Improved significantly, debt yielding ~7.5% blended basis now vs. over 12% 2 years ago. ABS market seen as attractive financing tool.
• 2025 Outlook: Reiterated full year revenue, adjusted EBITDA, and AFFO guidance. Uniti Leasing expected revenues $902M, adjusted EBITDA $872M; Uniti Fiber expected revenues $304M, adjusted EBITDA $125M. AFFO range $1.40-$1.47 per share.

Segment Performance:

• Uniti Leasing: Segment revenues of $222 million and adjusted EBITDA of $215 million, with an adjusted EBITDA margin of 97% for the quarter. Net success based CapEx was approximately $170 million, including $175 million of investment relating to the Windstream GCI program.
• Uniti Fiber: Revenues of $72 million and adjusted EBITDA of $29 million during the first quarter, resulting in an adjusted EBITDA margin of 40%. Non-recurring revenue was lower than expected due to timing of a $4 million fiber sale. Uniti Fiber net success based CapEx was $18 million in the first quarter, a ~25% decline from prior year’s levels.

Guidance:

• Reiterated full year revenue, adjusted EBITDA, and AFFO guidance.
• Uniti Leasing: Revenues $902M, adjusted EBITDA $872M at midpoint, $185M success based CapEx (including $175M Windstream GCI).
• Uniti Fiber: Revenues $304M, adjusted EBITDA $125M at midpoint, $85M success based CapEx, capital intensity 28%.
• AFFO range $1.40-$1.47 per diluted share, midpoint $1.43, 6% increase from prior year.

Risks:

• Macro factors: Capital markets volatility, potential recession risk due to tariffs, impact on capital costs.
• Regulatory: Prolonged higher tariffs could affect CapEx.
• Merger execution: Integration challenges post-merger.

Question and Answer: Q: Concern on M&A environment and lease-ups shifting to training data centers/inference?

A: Good questions. Yes, on M&A, the short answer is no; definitely don’t see any slowdown in activity there at all. In fact, I’d say probably the opposite or at the very least no speed bumps related to the conversations that we’re aware of and the progress that people are making on various strategic fronts. And look, from our perspective, we’re very focused on integration -- putting an integration plan in place and getting our transaction closed and hitting the ground running on legal Day 1 without any disruption of service but also accelerating our insurgent fiber go-to-market strategy and really accelerating the Kinetic build but always have M&A in the back of our minds. That’s a gene that we’ve had at Uniti for many, many years and that’s never going to change. So we’re staying very engaged with the strategic market, both strategics and financial parties. And I just think there’s a lot of interest in the fiber space; I think it’s fueled by both the convergence themes that we’re seeing across the industry and it’s also, of course, fueled by the hyperscaler activity. And we happen to have a set of assets, certainly on a combined basis with Kinetic, that are right down the fairway on both of those. And so we’re in the middle of a lot of interesting conversations and I look forward to -- we look forward to continuing that on a go forward basis. Look on lease-up, yes, I think you nailed it, Greg. I think it ebbs and flows [ph] when you just have the quarterly check-in. But from the standpoint of hyperscalers, we’re definitely not seeing any slowdown in the investment required for the large language models. And I think that’s not going to change anytime soon. I think we’re in a 1 or 2 to 3-year investment cycle here for those models. I think you’re going to see some large -- probably some large greenfield type opportunities coming down the pike for us later this year, especially on a combined basis with Windstream wholesale. We’ve got some opportunities in the funnel that we’re very, very excited about and can’t wait to talk to you about. So you’re still going to see those large greenfield opportunities but I do think the inference phase is going to be upon us a lot sooner than expected, or at least sooner than we originally expected, Greg. And as you’ve heard us talk about many times, that’s the phase that we’re most excited about because that’s when I think you’re going to see the real ramp in recurring revenue for fiber businesses as these large language models start to fuel people’s usage of AI across all the different endpoints that we have. And when you listen to what the hyperscalers say publicly, they’re starting to have trouble discerning between AI work streams and cloud-based work streams. They’re starting to mesh into one single -- AI is starting to be infused in all the other work streams, cloud-based work streams and others which I think is an early indication that inference is already here. And at some point, we’re not going to be able to distinguish what’s AI versus not; and I think that’s very exciting. So -- and we certainly haven’t seen any slowdown in hyperscaler activity. And really bringing that back to your question about lease-up, a large portion of the lease-up that we saw this quarter actually was from hyperscalers. And another theme that we’ve mentioned previously was that some of these really high-strand count transactions that we’ve seen in the past 12, 18 months, we’re now seeing hyperscalers come back to us and double down on those high-strand count requests which again is an exciting trend because it validates their infrastructure investments from a couple years ago, a year ago and they’re now seeing that capacity be consumed and needing more, even when they’re initially asking for 400 strands or 1,800 strands and coming back for more; so exciting times. I think you’re going to always see a good healthy mix of lease-up in our model though. That’s a conscious effort on our part and that’s what helps keeps our free cash flow yields approaching between 25% and 30% when you’re really sweating the asset that way. So good call out, Greg, but good themes behind all of that.

Q: Unpack bookings nature, AI-related percentage, returns on these AI driven builds relative to some builds, you know, some of the other builds you’ve seen historically?

A: On the bookings, we -- so directly to your question, the percentage related to hyperscalers is probably around 20%, depending on how you measure it. So somewhere in the 15% to 20% range which by the way has been pretty consistent over the past 12, 18 months. It’s been a growing percentage but it’s been relatively consistent in that range over the past couple or 3 quarters which is great. And we always talk about one of the benefits of the wholesale fiber business is that we’re agnostic as to the winning use cases of fiber or the use case of the day in fiber. Right now AI is front and center for everybody but the reality is when you peel back the onion, all of the different use cases of fiber for us are accelerating. Last year and continuing into this year, our biggest customer segment is actually the Fiber-to-the-Home providers across the country procuring backhaul to support the Fiber-to-the-Home buildout. And we’re seeing that again this year; so very excited about that. We’re excited about the AI theme. I mentioned in my prepared remarks that the wireless carriers are starting to spend again, bookings for wireless was double quarter the first quarter what it was -- the first quarter of last year -- which again, we sort of foreshadowed that at the end of last year that wireless was picking up. So all that to say, AI bookings are growing, it’s just the rest of our bookings are growing as well. And so continuing to be in that 15% to 20% range. Also, just to call out, Rob, on AI. And we’ve mentioned this before but because we’re still in this large investment period for the learning models, many of those deals don’t get reflected in bookings in traditional way because these are greenfield builds that have very high NRCs and get treated as either IRUs or strategic fiber sales. And so the activity with the hyperscalers is a little bit understated by the bookings number, when in reality I think it’s a lot greater. And back to my point about inference, I think that’s going to change once we really get into the inference ramp later on. With respect to the returns on these deals, look, I think that we treat them in the same way that we treat all other anchor lease-up models that we look at. And for the most part, the hyperscaler deals are generally anchor deals for us. And so as a reminder, our strategy is to target 5% to 10% yields for the anchor with a really clear path to lease-up beyond 10% after the anchor deal. And that’s why we track and report each quarter to show that across the portfolio, we’re nearing 30% [ph] blended yields on our initial anchor deals. When you put the hyperscaler opportunities and you look at it through that lens, we’re nearing 20% yields on our hyperscaler deals. So inclusive of anchor yields plus lease-up over the past couple of years, we’re already approaching 20% yields. So, we don’t like to talk about specific customers and specific customer deals but on a blended basis our hyperscaler deals are tracking frankly ahead of our traditional anchor lease-up model.