And it’s a huge sensitivity driver in the overall financial performance of the investment. If you can double penetration rates in the first 18 months over what had been historic levels, it’s amazing what that does to pay back effectively and we’ve been really successful at doing that. I think to give you a little color on your question, as we hit the 40% pen level in a market, which we’re now getting more and more markets where we’re kind of at that 40% share, 40% penetration level, our tactics do switch. And so, as more markets hit the 40% level, we have to go to a little bit different set of tactics around how we do that. And I think frankly in many instances, especially where those are in region, they play into our strengths in terms of how we drive more value into the household, how we use the bundling lever in an effective way, how we use data differently to target, what distribution channels we use to contact those customers shift as a result of that.
So I would say we have a really fine-tuned set of plays that get us from 0 to 40%, and we’re pretty good at doing that. And then when we hit 40%, we kind of start to use a different set of playbooks in those particular markets as they mature. And I feel really good that the team has their handle on that and they’re doing it the right way. We spend the right money at the right time to unseat those customers and that’s why you see that business scaling so nicely in the way that it is.
Phil Cusick: Since you mention it, what is that first 18-month penetration at this point?
John Stankey: So next question.
Phil Cusick: Thank you. Bye, John.
Operator: Our next question will come from the line of Michael Rollins of Citi. Please go ahead.
Michael Rollins: Thanks and good morning. Two, if I could as well. First, could you discuss the factors behind the slowing wireless device upgrade rate, how it’s impacting the financials, and could this go even lower in the future? And then secondly, just an update on where AT&T is on the run rate cost cutting targets and if you could size the potential for additional savings over the next one to three years?
John Stankey: Sure, Michael. So I would tell you that I don’t think anything’s really changed in what we see in the device rate, and whether or not it slows dramatically over what its current rate is. Hard to say. I would say the bias is, pick your view of what the economic environment does, and that’s probably the higher correlation as to what happens to the upgrade rate. And, if there’s a more stressed economic environment, maybe it slows a bit. If it stays healthy and robust and kind of perks along where we’re at right now, then I don’t think it’s going to dramatically change. I think what we’ve been seeing in general is a couple things. I’ve mentioned this before in other calls and it doesn’t change my point of view. The devices, frankly, from generation to generation change a little bit less.
It’s harder to get differentiation in the hardware. I mean, they’re really good cameras on them and there’s really good modems and they all have really good speed because of the spectrum bands they handle and so customers aren’t necessarily hanging on a device evolution to say there’s such a dramatic uptick in functionality that I can’t use my device for several years. Two, I think people are — the human body, the commercial case industry that is responsible for protecting devices, people are dropping them less and they’re taking better care of them, and as a result of that, they last a little bit longer. You add onto that the fact that we’re very successful at selling insurance into our customer base. Because we sell insurance, customers are more prone to potentially take a replacement device within the terms of their agreement that they have rather than swap out to a new device.
And that has worked out well for customers and it works out well for us. And that tends to extend the life cycle a bit. And look, as devices get expensive — more expensive, and they are getting more expensive, for whatever reason, consumers are rational animals, and like any other more expensive thing, oftentimes you keep it a little bit longer, you try to squeeze a little bit more out of it, and I think there’s a cycle of that occurring. So I think that’s why we’re seeing the cycle we’re seeing, whether it continues to thrive its way down or it kind of flattens out remains to be seen, but it’s pretty explainable, and I don’t think it’s going to substantially alter kind of our point of view of momentum views for you and what we’re looking at as we move forward.
On the run rate on cost cutting that you ask about, look, the run rate is we did $6 billion over three years, right? And we did that in an inflationary environment, so it was really like, I would say it’s 130% of what we really wanted to do when you think about what we were able to actually work through and get done. We’ve given you $2 billion more over the next three years. You saw that we took an accrual this quarter. That accrual is set up through the course of next year. I think if you want to understand how we expect some of these costs to go, there’s probably a correlation to that accrual that you should think about. That’s obviously not all of it, but it’s a portion of it. It would give you some indication of how we think about feathering this in over the course of the next 12 months.
And I would say, as I indicated in my opening remarks, we feel really good about where we are in momentum right now and some of the things we have underway. I mentioned to you many quarters ago that we’ve been investing in our information technology infrastructure, it’s been painful, it requires a lot of work, it’s very, very detailed work every time you change out a CRM system or billing system and you have to carefully deal with your customer base and your different distribution channels. We’re now getting to the point where we’re starting to turn some scale up on those platforms. That coupled with the fact that more of our activity is built on fiber and wireless is giving us a different kind of cost structure in the business. We’re going to continue to ride that curve.
We’re going to continue to make sure that we streamline the business effectively for what we have as the new products moving forward. And that’s part of the legacy migration and what we’ve been doing in geographic footprint shutdown. And I feel really good about us being able to achieve that $2 billion over the next three years.
Michael Rollins: Thanks.
Operator: And our next question will come from the line of Craig Moffett of MoffettNathanson. Please go ahead.
Craig Moffett: Hi, thank you. I want to stay with the topic you were just discussing about upgrade rates and things and just try to get a sense of what you’re seeing in terms of the new iPhone launch and what that might mean for margins in the fourth quarter. And then just a second question if I could just to clarify your remarks earlier on the BEAD program. As it does ramp up in what now sounds more likely to be 2025, would you expect that that would be, to some degree, a substitute for some of your fiber builds that are currently thought of as competitive overbuilds, or would they be a supplement to competitive overbuilds?
John Stankey: So what I would tell you, Craig, is I’ve made some comments in my opening remarks that we had a pre-order rate in this cycle that was probably the best we’ve seen in a long period of time. And whether or not that’s unique to AT&T or unique to the industry, I don’t know. I have not heard others report at this juncture. I don’t know what their guidance is going to be, but I will tell you that our upgrade rate was a bit higher than what we have seen in the last several quarters, but it wasn’t what I would call out of pattern where it’s going to be anything that is inconsistent with the guidance we’ve given you on our margins for the year and inconsistent with what we expect for the business performing. So everything that we have articulated to you where we would guide in on service revenue growth, what we think the operating margins are going to be within the business, our EBITDA performance is all still very much in check relative to what we see going on there.
