Greetings, and welcome to the Americold Realty Trust Third Quarter 2025 Earnings Call. [Operator Instructions] Please note, this conference is being recorded. I will now turn the conference over to your host, Rich Leland. Please go ahead.

Good morning, and thank you for joining us today for Americold Realty Trust's third quarter 2025 earnings conference call. In addition to the press release distributed this morning, we have filed a supplemental financial package with additional detail on our results. These materials are available on the Investor Relations section of our website at www.americold.com. This morning's conference call is hosted by Americold's Chief Executive Officer, Rob Chambers; and Jay Wells, our Chief Financial Officer. Management will make some prepared comments, after which we will open up the call to your questions. Before we begin, let me remind you that management's remarks today may contain forward-looking statements. Forward-looking statements are subject to a number of risks and uncertainties that may cause actual results to differ materially from those anticipated. These forward-looking statements are based on current expectations, assumptions and beliefs as well as information available to us at this time and speak only as of the date they are made. Management undertakes no obligation to update publicly any of these statements in light of new information or future events. During this call, we will also discuss certain non-GAAP financial measures, including NOI, constant currency, net debt to pro forma core EBITDA and AFFO, among others. The full definitions of these non-GAAP financial measures and reconciliations to the comparable GAAP financial measures are contained in the supplemental information package available on the company's website. Please note that all warehouse financial results are in constant currency unless otherwise noted. Now I will turn the call over to Rob for his prepared remarks.

Thank you, Rich, and thank you all for joining our third quarter 2025 earnings conference call. Before diving into the third quarter results, I would like to congratulate George Chappelle on his well-earned retirement after a long and successful career. He originally stepped into the Americold CEO role coming out of the disruptions from COVID and outlined the 4 key priorities that you have heard us talk about on our previous calls, a focus on providing excellent service to our customers, improving the retention, training and productivity of our workforce, growing our service margins and building out a robust pipeline of attractive development opportunities. I had the opportunity to work side-by-side with George along the way as we made significant improvements across all of these areas. They are now part of our company DNA and remain a foundational component of our strategy. Personally, I also benefited from having George as a mentor as he helped prepare me to lead Americold into the future as part of the Board's succession plan. Over the past 2 months, I've visited several geographic regions, both domestically and internationally, connecting with our teams and reinforcing our shared values and priorities. In addition, I've spent considerable time engaging with many of our top customers and strategic partners, most of whom I've had a relationship with for many years. The strength of these relationships, combined with our global scale and presence at all key nodes in the cold chain provides us with attractive and unique future growth opportunities. While I will continue to pursue many of the strategies we have established over the past 4 years, I believe we also have the ability to lean further into the areas of the business that we think provide the best long-term opportunities, such as growing our market share in…

Thank you, Rob, and good morning. First, I'd like to discuss the results for the quarter, then our capital position as well as our outlook for the remainder of the year. As Rob mentioned, third quarter AFFO per share came in at $0.35, which was in line with our expectations. Same-store economic occupancy was 75.5%, down year-over-year, reflecting the continued demand pressure that we have seen in the market and flat sequentially to the prior quarter. Same-store throughput increased slightly sequentially from Q2, largely due to the start of the annual agricultural harvest as expected. Same-store NOI contracted from the prior quarter, primarily due to the seasonal increases in power costs, in line with our guidance, and we continue to diligently control our expenses. While the fundamentals of the business remain pressured, the team continues to execute well. Despite the competitive pricing environment, our rent and storage revenue per economic pallet increased on both the sequential and year-over-year basis, as we continue to balance both price and occupancy. In addition, our services revenue per throughput pallet also increased both sequentially and year-over-year. Customer churn remains in the low single digits, while rent and storage revenue from fixed commitments held steady at 60%, maintaining the record level that we achieved earlier this year. As a reminder, we may see some quarterly fluctuations in this metric. However, 60% remains our long-term goal based on the fact that approximately 70% of our revenue comes from our top 100 customers, and most of them see the benefits of the fixed commitment contract structure. At quarter end, net debt to pro forma core EBITDA was 6.7x with approximately $800 million of available liquidity. We remain disciplined and prudent in our capital allocation decisions, focusing on customer-driven and strategic partnership projects that are lower risk and…

Thanks, Jay. While the current environment presents no shortage of challenges, the strength of our management team and diversification of our real estate gives us a strong competitive advantage in the market. Our value proposition remains strong, and we are managing the business to set ourselves up for the long-term success, leaning into opportunities and finding new ways to grow. The presence of the previously discussed headwinds does not diminish the importance and value of our operational excellence, deep customer relationships, industry expertise and mission-critical scale and diversification. I think it is important to highlight that Americold today is trading at a significant discount to our intrinsic value, and this is supported by several different measures. From a replacement cost perspective, it would be impossible to acquire the land and replicate the 5.5 million pallet positions in our real estate portfolio for anywhere near our current $8 billion enterprise value, not to mention the incremental value of our operating system and experienced team of associates. We are also currently trading at a historically high cap rate of around 10%, which is unusual for a business like ours that owns mission-critical infrastructure backed by long-term agreements, fixed committed contracts with high credit quality tenants. And finally, we have an enterprise value to EBITDA multiple that is well below valuations for most of our publicly traded industrial and commercial real estate peers. My job, along with our management team and all of our associates around the world is to operate this business to maximize the value of these assets for the benefit of our customers and shareholders, and I believe we are taking the right actions to ultimately deliver outsized earnings growth. With that, I'll turn the call over to the operator for questions. Operator?

[Operator Instructions] And our first question comes from Samir Khanal with Bank of America.

I guess, Rob, when I look at the KPIs in the quarter, occupancy and pricing did improve sort of when you look at it year-over-year, but throughput got a little bit worse. I guess how should we think about kind of throughput over the next 12 months? And maybe sort of expand on kind of what you're seeing on the ground over the last couple of weeks?

Sure. Thanks, Samir. So yes, from a throughput perspective, I mean, I think we still hear from our customers that the same thing that they're saying on their earnings releases, which is demand is challenged, largely because of lower and middle-income consumers that are still significantly under pressure from all of the factors that I mentioned in my prepared remarks. And so while there still should be some seasonal demand for Thanksgiving and for Christmas, it is muted. And that's largely what we had anticipated. And as we go forward into next year, we're not yet at a point where we feel like we can predict an inflection point. And so we think throughput will still be challenged as we go into next year. What we're hearing from customers on the ground is similar to what I just described. I think you've got customers that are hesitant, to be honest with you, to build inventory in the current environment until they really see a sustained increase in demand. And so as we're going through our discussions for next year, we factored all of that into some of the foundational elements that Jay talked about on the call in terms of what our expectations are for next year.

And if you look at sequentially, last call, I did talk that we'd see a little bit of lift sequentially in throughput, which we did. And that was really driven by the start of the harvest season and us starting to see those products come into our sites. And then next quarter, you will see we have a small lift in occupancy, about 100 bps, give or take, and that's really driven by the harvest season, too. So actually, throughput sequentially came in right around where we expected it.

Got it. And then, Jay, I guess, when I look at your guidance and also all the assumptions you have there, most of the items were unchanged, but interest expense did come down. So all else being equal, I mean, we should have probably seen an increase in AFFO, but that didn't go up. So maybe provide some color around this.

Yes. Sure. If you also look, it's a little bit, there was a move in classification from other income over to interest expense. So you'll see that the other income went down a similar amount. So overall, net-net, it didn't benefit AFFO.

And our next question comes from Greg McGinniss with Deutsche Bank (sic) [ Scotiabank ] .

This is Greg McGinniss with Scotia. I appreciate your ability to kind of project the business into the back half of the year. I'm curious on the margins that you're seeing quarter-over-quarter, some margin decline year-over-year as well. What are you doing to control the cost in the business? And what are your expectations there going forward?

Sure. So on the margin side of the business, with lower occupancy and lower throughput, obviously, that's going to challenge your margins a bit and you don't get the same leverage across your fixed cost base that you like to see when volumes go the other way. But we continue to do a really good job of controlling costs. We've been able to manage and match our direct labor to our throughput in a way that I think has really helped boost handling margins. We've delivered handling margins in excess of 12% and are on track for that, which was our goal when we came into the year and continues to be outsized relative to historical margins on that side of the business. I think that we are seeing really good progress and results out of Project Orion. And so we're continuing to implement that across the regions and Europe will be a big beneficiary of that as we go into next year. And then every single year, we have productivity targets that we set for our operations team. We have 2 great leaders of the P&L, like I mentioned on the call, and Bryan Verbarendse and Richard Winnall, who are very skilled and experienced at driving productivity through the Americold operating system and our technology platform. So I think we're going to be able to continue to control costs in a way that will allow us to deliver margins that we're comfortable with.

And on call, I discussed, we do have a very robust process of evaluating all of our low occupancy sites. We did remove another 3 sites this quarter with more targeted. And as we continue to do that, that's also taking cost out and will help us maintain our margin levels.

Great. And I just wanted to follow-up as well on the pricing impact expected from new occupancy -- sorry, new supply delivered over the last few years. Are you -- is Americold going to need to adjust fixed commitment pricing down as those contracts expire given the supply that's hit?

Well, I think you see that reflected in our prepared remarks in terms of what Jay outlined for our expectations as we go into next year. What I'd say is the team has done a remarkable job over the last few quarters. We've been in a tough demand environment now for a while, and you've seen us be able to maintain the fixed commitment levels at that goal of 60%. You've seen growth in pricing, both on the storage and the handling side over the last several quarters. But there are certainly some markets and some nodes. We called out the 4 distribution centers in particular, where there's pressure on both of those KPIs, both from a pricing standpoint and from a fixed commitment standpoint. So we've chopped a lot of wood in terms of getting through a lot of our contract renewals during this tough environment, but there is more to go. And in certain instances, we're seeing some of the fixed commitments get tightened up. We generally don't see our customers moving away from fixed commitments because they do want to protect the space. They understand the value. But in instances where their physical inventory has decreased to a point where they can bring down the fixed commitment a bit, we've seen some of that, and we've planned for that in terms of some of the building blocks that we outlined in the prepared remarks.

And moving next to Michael Carroll with RBC Capital Markets.

I guess, Rob, in prior quarters, you highlighted a pretty sizable sales pipeline that reflected roughly 8% of total revenues. I know your updated guidance range has assumed that this comes online in later periods kind of pushing out to 2026. I mean is that still the case? I mean, are these customers still going to bring product into your facilities? Or has that kind of pulled back and that sales pipeline kind of got smaller over the past few quarters?

Thanks, Mike. The sales pipeline has been a bright spot. I'll tell you; we're going to have a very good sales year this year. It will be a record for us in terms of new business wins. It's definitely been slower to materialize than we had originally planned. And in some cases, a lot of these programs are not immune to the same challenges that the rest of the business has had. So as they come into Americold, they come in, in lower amounts than what were originally anticipated or contracted for. So it's still a highlight for us. I think new business. The team has done a great job acquiring it. But in the end, we have seen some of that offset by both reductions in the base business and just traditional customer churn.

Okay. And then on the fixed commitment side, is that -- should we expect more of those contracts to be up for renewal in the beginning of the year? I mean is there kind of seasonality? Or is it kind of spread out throughout the year?

That really is spread out through the year, Mike, is the contract terms tend to be based on when they're signed. So it's contract years more than it is fiscal years. So you'll see, I would say, a relatively consistent renewal cadence throughout the course of the year versus anything outsized in one quarter or another.

Your next question comes from Michael Griffin with Evercore ISI.

Rob, I want to go back to your comments just on the fixed commits and how you're negotiating with them, realizing that maybe you're prioritizing the stability of those cash flows that we might consider traditional REIT income types, so to say. But would you say that you'd be willing to give a bit on pricing in order to secure a longer-term commit? Or maybe walk us through the push and pull of a longer fixed commit contract versus what the pricing might be there?

Yes. I mean we balance all of those things. I mean we're looking at existing profitability. We have all the tools to be able to understand what market rates are, what profitability is by activity. We have a great activity-based pricing model. And so any time you have conversations with customers about a contract renewal, there's going to be dialogue around price, around volume, around length of contract, around business across the network. So we balance all of those things to try to make sure that we're doing the right thing to maximize the value of those agreements and ultimately be able to defend our market share, while also maintaining the appropriate level of profitability. So there's not -- I think the most important thing to say is there's not a one-size-fits-all strategy there. You have to take each agreement kind of as they come and understand where the current profitability is and what levers you can push and pull to get the right outcome for both us and our customer.

That's some helpful context. And then maybe, Jay, you talked about the facilities that you're taking offline. What happens there from a P&L perspective? Are you capitalizing the costs associated with those facilities now? And what would be the ultimate plan for those? Would it be reposition it, sell it? Maybe walk us through that a bit.

Thanks for the question. Many of these are leases, and it really is end of lease term that we evaluate them. So overall, once we remove all the pallets from the facility, those types of costs will go below the line. They are capitalized, but they're generally pretty minimal at that point in time. But it is predominantly lease facilities that are either being repurposed by the landlord, removed for residential purposes, so a variety of different uses. So it's mostly a small amount moves below the line when they become inactive assets, but that's only for a very short period of time.

And obviously, the benefit from our standpoint, too, beyond reducing some of the costs and eliminating some maintenance expense is the fact that you get to move a lot of the existing customers from those leased facilities into owned infrastructure, and that provides a nice benefit.

And our next question comes from Blaine Heck with Wells Fargo.

Rob, you talked about spending considerable time engaging with customers and strategic partners. Can you just talk a little bit more about what you learned on the customer side, particularly how they're thinking about cost pressures on their businesses and what impact that's having on their inventory planning versus kind of the lower demand environment that has been mentioned several times as kind of the driver of that inventory management going forward?

Sure. So I mean every customer is looking at ways, obviously, to find opportunities to be efficient as they look out and they say that demand may be softer for a longer period of time than what anybody had originally anticipated. I think from the discussions with most of our customers, what they are really having their internal discussions and debates about are when is the right time to build inventory. This, as an example, would be the typical time of the year that you would see significant builds in inventory to support what are seasonal spikes in demand. What is a bit of a different approach at the moment are some of our customers saying, we're going to try to manage these shorter-term or seasonal spikes in inventory with the existing product that we already have in the system. So they're hesitant to build because nobody wants to be in a position where they have excess inventory like what happened, say, 18 months to 2 years ago, where many of the food manufacturers got their workforce rebuilt and back into their production plants, overbuilt and then took a long time to bleed down that inventory because they were in an environment where the demand just wasn't there. And so the internal conversations that most of our customers are having is looking out over the course of the next few quarters and saying, what are some of the indicators they can see to show that maybe any of the increases from a demand perspective are sustainable, and it would allow them to ultimately start building. The other big question that a lot of our customers on the food manufacturing side are having is when is the right time to introduce new innovation, new products, new SKUs. Those are things that we very much look forward to because obviously, as you see more innovation, more SKUs, that drives incremental safety stock. So I think our customers are trying to have those conversations. And then lastly, I would say it would be what levels of promotional activity are really going to drive volume. So all of our customers do want to continue to invest in their product. They have over the last few quarters in a variety of ways with mixed success, to be honest with you. I think when customers have historically made the investment in their product to support promotional activities, they've probably seen better results than what they've seen over the last few quarters. And that's simply because the cost of food has gone up at such a pace that even a slight discount off of that elevated price isn't enough to stimulate demand. So those are really the 3 questions that our customers are having every day with themselves and with the retailers.

Okay. That's really helpful context. Secondly, you talked about some of the newer competition in the industry that don't have a sustainable long-term business plan. I think you mentioned some of them already exiting. I guess when do you think you see those exits really accelerating? And how much of that product is likely to be the quality and potential price that you're comfortable with and maybe an acquisition opportunity?

I mean it's certainly something that we believe will become opportunistic over time. We're just not there yet. So most of the new market entrants that have come in really thought about, okay, let's get some scale. And then I think they were potentially encouraged by a lot of the acquisition activity that have been happening going back a few years and thought that, that would be a great exit strategy. With that not in the cards at the moment, it puts a lot of pressure on that business model. And so when you don't have the same level of, let's say, network that Americold has or you don't have the same operating system we have or the technology stack that we have, there's not a lot of levers to win new business. Prices may be one. But if you start deeply discounting space to fill up your buildings and you can't get to a point where you're at full occupancy, the P&L looks very bad. And I think that's where we are for a lot of these new market entrants. How much and how long folks want to deal with that is certainly not our call. But we're not in a position where we're going to be bailing any of the new market operators out. And so I think we're in a position where we can sit focus on driving our business, focus on growing our relationships with customers. And over the next few quarters, as some of that maybe results in more capitulation, then we're here to listen. But at this point, we're focused on driving our business.

And moving on to Michael Goldsmith with UBS.

You talked about how it could take a couple of years for the excess capacity to be absorbed. So what are the assumptions that you're using to arrive at that conclusion? And how can you best position yourself to navigate that sort of backdrop?

Sure. So we called out in our prepared remarks that we've seen what we believe is in excess of 15% of additional capacity that has been added. And this is an industry that, for a long time, had grown more with GDP and population growth. And so if you just do that math, it would be a few years for it to be absorbed. I think as we continue to gain market share, that certainly helps absorb some of the capacity. I think as I've said, with the business models or the business plans that a lot of these new market entrants had, if those business models don't work, and we really believe that a lot of them are struggling at the moment. That potentially accelerates the ability for Americold to play a role in absorbing some of that capacity. And then outside of that, I think the other thing that's important to keep in mind is Americold is not standing still in this environment. We have a lot of different ways to open the aperture in terms of how we drive new business into our portfolio, which isn't just waiting for the core kind of frozen food on the manufacturer side to grow. We're going aggressively after retail business, which is largely in-sourced. We're going aggressively after quick service restaurant business that today we play a very little role in. I think there's opportunities to look at triple net lease deals that historically we've been not as open to because we want to do both the storage and the operation. I think there are commodities outside of just food. So there's a lot of things that we're in early stages of exploring. And I think as some of those initiatives ramp up, we'll see our own capacity fill up, and we'll see the ability to potentially take some of those capacity in the.

Appreciate that color. And my follow-up is on pricing. You're telling low occupancy facilities. Can you talk a little bit about the pricing from like low occupancy facilities is something that may be more full?

Yes. I mean it really does depend on a lot of different things. It's -- our customers signing up for commitments, are they signing up for longer-term agreements? So it can be a pretty big variety across the board. I think we understand well, given our size and our scale, what market rates are in many of the different geographies. And so what we tried to articulate on the call was that when we look at it across the nodes of the supply chain, which we think is a great way to talk about this business, the ones that are under the most pressure are the 4 distribution locations. That's where most of the speculative capacity has been added. And so we are being more thoughtful about the way that we defend our market share and win new business in those geographies. And the net of that is the potential outcome that Jay outlined in his prepared remarks.

And Nick Thillman with Baird has our next question.

Rob, I appreciate all the commentary on all the different nodes, but one knock that generally is put on Americold is just the age of the portfolio relative to all the new builds and optimization of networks and kind of the effect there. I was wondering if you could break down or dig a little bit into -- you're talking about the new supply issues just broadly in the forward distribution node. But if you look at the portfolio age and you look at sort of your composition, how does that all break down? Is it pretty similar across all 4 of those? Or is it maybe a little bit more skewed one way or the other?

Yes. I don't -- to be honest with you, I don't have the numbers right off the hand. So I don't want to share anything without all the facts. But I do want to say that the first point we would disagree with vehemently. We spend a tremendous amount of effort, dollars, time maintaining these facilities. Our buildings are world-class. They provide a great service to our customers, and they're mission critical. If anything, other than that was the case, you would see Americold losing market share, not gaining market share. And so because we've got the team that we have in place that maintains these facilities, we're very proud of our network. It's led to Americold being able to lead the industry from commercial excellence in terms of the most fixed commitments and the pricing type gains that we've been able to achieve over the last few years. All of that is because of the mission-critical high-quality infrastructure that we have. And so we would vehemently disagree with anyone that says that the age of our network is a knock on Americold.

No, that's very helpful. And then I wanted to get your -- pick your brain a little bit on the comments. Jay, you mentioned sort of the hurdle rates for new developments. And as we look at your development schedule just over the last 3 years, haven't necessarily hit the stabilized yields yet even for like the 3-year vintage assets. So I want to kind of pair that with Rob's comments on driving shareholder value, thoughts about -- and the discount in the stock price. I guess where does stock share repurchases kind of rank in the deployment side of things as we look at capital allocation going forward?

So let me start, and then I'll hand it off to Jay. I think when we look at our development projects, the important thing to keep in mind is that these projects largely are not immune to the macro environment that impacts the broader network. So any time we're building a facility, whether it's dedicated or multi-tenanted, if our customers' volumes are going to -- are down, that's going to impact the current returns. We still have a very high degree of conviction that our development projects will meet our stabilized returns and underwriting. It just takes a longer period of time in this environment. And we're very encouraged by the significant improvements that we've made in our development platform over the last few years in terms of the team that we've been able to bring in. And you see that reflected in the fact that just over the last 2 quarters, as an example, we've delivered multiple projects on time and on or under budget. So feel very good about the development platform when it comes to some other capital allocation decisions. Jay anything else?

No. I said a couple of things on my prepared remarks. Number one, we have got to provide the growth requirements for our customers for our partnerships with CPKC, with DP World. That is a must do to maintain our customer base and our great partnerships that we have. And then second, I mentioned on the call, maintaining our dividend, maintaining our investment-grade profile our top priorities also. So really, we're balancing those 2 items in development pipeline and maintaining our dividend and our investment-grade portfolio based on our current leverage.

We'll go next to Mike Mueller with JPMorgan.

A couple of questions. So for the first one, for the 200 to 300 basis points of economic occupancy erosion for the year that you're talking about for '26, should we think of that as being ratable throughout the year or starting off worse and ending the year better, which is obviously better for '27 or just kind of vice versa? And the second question is, I apologize if I missed this. You talked about the economic occupancy down. But for '26, is it safe to say that you're expecting year-over-year pricing to be negative as well for services and storage?

Sure. On the pricing side, yes, what we called out there is we think that it could be a headwind next year of 100 to 200 basis points. We'll continue to do our general rate increases. But when we think about what it takes on the renewal side of the equation, when we take -- think about what it takes on driving new business and defending our market share, we think when we aggregate all those things, we could see it being a potential headwind for next year. So that's on the pricing side.

And that's across both storage and services to answer your question. And then when you look on the occupancy side, it really is going to come as Rob talked throughout the year as we redo our fixed commit. So there will be some headwind to start the year, but we also have a little wrap headwind from this year. So I would say we're not giving specifically quarterly guidance at this point on our occupancy. We feel at this point of doing our budget, very confident that the guidance I gave on the call is appropriate. But quarterly, you have a little bit of wrap because we've seen some headwinds to start this quarter and next quarter that will flow in. But hard to say exactly how to phase it throughout the year at this point.

Yes. I think the point, Mike, is we don't do annual resets of these. So it's not like, hey, on January 1, all of our contracts reset. We negotiate our agreements as they kind of come online throughout the course of the year. When we sign an agreement, that tends to -- that date that we sign the agreement tends to be the annual kind of check-in point for when we -- when contracts ultimately are renewed. So it's not on a calendar basis. It's more on a contract year basis.

Got it. So -- but that's 100 to 200 average. And if you're talking about the ratable contract, I guess, negotiations occurring throughout the year, it just seems that you would end the year possibly at a lower point than that 100 to 200. Is that a fair statement? Or am I kind of off on that?

No, no, that's not how we're thinking about it. We're not really thinking about ending next year below those -- the points that -- or the metrics that Jay called out in the script. I think the way to think about that is that's the annual impact of it.

Moving next to Todd Thomas with KeyBanc Capital Markets.

I guess I just wanted to follow up first on that line of commentary around economic occupancy. I guess, as we look at expirations over the next few years, is there a potential risk of further decreases in economic occupancy beyond 2026 if demand does not improve much in the quarters ahead or if conditions do not really pick up from here?

I mean we're really not at a point now where we're talking about anything beyond what we think is going to potentially happen in 2026. I mean you see the renewal schedule. So our agreements with the large customers. So again, 70% of our revenue comes from our top 100 customers. They tend to be the ones that sign longer-term agreements. Those agreements are anywhere between 3 and 7 years. So they average 4 to 5. So every year, there's going to be a tranche of contracts that come up for renewal, and they're all based on what the current market conditions are at the time. So if the environment improves, we think it becomes certainly a tailwind for us. And if the environment doesn't, it could become a headwind.

And keep in mind, we have been in a difficult environment for a while now. So we've already rolled through several renewals of agreements already, and we really see next year as being really the key year to get through the predominant amount of these type of agreements in the current difficult environment.

Okay. And then, Rob, you spent some time talking about the current portfolio mix today. There was a lot of commentary sort of back and forth around some of the different nodes. And I was just wondering if you can expand your comments around that in terms of emphasizing capital deployment, whether you plan to sort of reshape the complexion of the portfolio. I guess, how should we think about the portfolio mix going forward? And are there any significant changes that we should anticipate to that mix?

So we wanted to highlight that we put particular importance on those production advantage in the retail locations because those are areas where we feel like we've established leadership positions that are very difficult for anyone else to come in and replicate. I mean on the production advantage side, there's a tremendous amount of benefit there in terms of those agreements tend to be longer term, fixed commitments. And it takes relationships with the big key customers that have been built over decades to really get them to trust you to build or run their plant advantage or plant attached sites. So we think there's opportunity to continue to grow at that node. Retail is also an area where we're going to lean more into. It's very opportunistic from the standpoint of the fact that most of this business is in-sourced today. And there's a moat around it because you have to have a great operating platform to be able to deliver the type of service in retail that's required from that group of customers. So I think you'll see us probably lean more into those 2. Certainly, we're not very interested in adding speculative capacity in the 4 distribution locations right now, given what we've seen occur over the last few years. And then even in the port facilities, we're really going to focus our efforts if we're going to grow in that node by aligning to the strategic partnerships not just adding speculative capacity, but adding capacity that's in conjunction with our 2 strategic partners that creates a value proposition and an ecosystem that nobody else can match.

And moving next to Brendan Lynch with Barclays.

Maybe just following up on that last one. Rob, in your prepared remarks, you mentioned you're considering expanding into other food and nonfood categories. Maybe you could expand upon that a bit.

Sure. So again, I mean, there are certain categories that we're already in like retail and QSR that we want to lean more into. But we do hear from our customers that there's opportunities, as an example, to co-locate some of their dry product closer to where their frozen or refrigerated products are. So we're having dialogue about that to potentially absorb some capacity. There are other markets like floral, pharma, components that all need refrigerated that today, we essentially do-nothing in. Pet food is a fast and growing market that we view as opportunistic. So as we look at opening the aperture here to continue to drive occupancy, I think we have a lot of avenues that today, we're just dipping our toe into that could be very opportunistic and look forward to talking about more of that over the next few quarters.

Great. That's helpful. And then it looked like your power costs didn't really increase that much year-over-year in the same-store pool. Can you talk about any related risk that you see coming related to power cost increases going forward and what protections you have in place?

Yes. I think, look, on power, I think we're doing a lot to drive power costs down in the business. We have solar programs. We do a lot of the maintenance programs that we have are focused on driving down power, the LED lighting type of initiatives that we have are all focused on ways that we can take cost out. We -- some of the continued maintenance that includes the rapid open and closed doors helps to save on power. So we've got a lot of different initiatives that drive those down. And I think the other thing that we've done a nice job of over the last few years is making sure that to the extent that we do see power increase in certain markets, that would be considered a cost change that's largely beyond our control that we would look to pass on.

And ladies and gentlemen, thank you for your participation. This does conclude today's teleconference. You may disconnect your lines, and have a wonderful day.