Thank you for standing by, and welcome to the Altice USA Q3 2019 Earnings Presentation. [Operator Instructions] I'd now like to hand the conference over to your speaker today, Nick Brown, with Altice USA. Please go ahead.

Hello, everyone, and thank you for joining. In a moment, I'll hand over to our Altice USA's CEO, Dexter Goei; and our new CFO, Mike Grau, who will take you through the presentation, and then we'll move to Q&A. As today's presentation may contain forward-looking statements, please read the disclaimer on Page 2. The slides are available on the company's website, and we'll make a replay of the call available. And now I'll hand over to Dexter.

Thanks, Nick. Hello, everyone. Before I begin, I just want to take a moment to welcome Mike as our new Chief Financial Officer of Altice USA. He most recently led our financial planning control organization, and prior to that, held various leadership roles in finance at Cablevision for more than 15 years. So welcome, Mike. I also want to take the opportunity to thank Charlie for his exemplary leadership at Altice USA, and I'm delighted for him as he takes on his new role at Sotheby's. I look forward to Charlie continuing as a member of our Board of Directors and to his ongoing valuable insights as adviser to me on strategic initiatives. Starting with the summary on Slide 3. Revenue growth slowed to 0.9% in Q3 as we lapped the prior year rate event and had the absence of political advertising revenue as expected. Adjusted EBITDA was flat year-over-year, although grew 0.7% ex wireless losses, which stepped up a bit since we launched Altice Mobile in September. Based on the initial contribution we've seen from Altice Mobile, we now expect slightly lower growth for 2019, approximately 2.5%, primarily due to the slower ramp-up in sales of effectively 0-margin handset equipment, as we had previously anticipated making those sales available on our e-commerce online channel in Q3. We expect to open this in other sales channels in the next few weeks to result in faster revenue growth in 2020, since it's mostly phasing for the ramp-up of Altice Mobile. Note that our guidance for margins, free cash flow and leverage remain unchanged. Given our confidence in the anticipated acceleration of revenue and free cash flow in 2020, we were happy to take advantage of available liquidity and increases in our buyback in Q3 to reach a total of $1.7 billion…

Thank you, Dexter, and good afternoon, everyone. It's a pleasure to join these calls, and I look forward to meeting, hopefully, many of you in the coming weeks. Resuming our presentation on Slide 10, we show how Altice USA's adjusted EBITDA margins have expanded over the past few years, reaching 44% in Q3 2019, which was flat year-over-year, excluding about $10 million in mobile losses. These mobile losses were slightly higher than the first half, reflecting some launch costs. It's worth noting that excluding the impact of political advertising revenues from both periods, our ex mobile EBITDA margins would have grown about 40 basis points year-over-year. We will continue to look for ways to drive efficiencies in all facets of our business as a means to further enhance margins and cash flow. These higher margins, in turn, allow us to continue to be aggressive in investing in all of our growth initiatives that Dexter highlighted earlier. Turning to Slide 11, we can see a breakdown of capital expenditures, which increased year-over-year as planned due primarily to our growth investments in fiber and new home builds. Our total CapEx intensity was 15.4% in Q3. But without fiber and new home build investment, this would have been approximately 11%. And remember, following the fiber build, we will be able to reduce CapEx significantly, and we'll also have opportunity to take operating expenses out of the business. The new FTTH network will enable us to reduce long-term costs to improve customer experience, including reduced customer interactions, lower technical service visit requirements and lower plant maintenance costs. And finally, our mobile CapEx dropped as expected in Q3 since we've already completed the core network build and initial store upgrades, and we'll be very CapEx-light from here. We may roll out a few additional stores…

[Operator Instructions] Your first question comes from John Hodulik with UBS.

Great. Maybe for Dexter. Can you give us any more detail on the promotional roll-off you talked about in your prepared remarks? Maybe give us a sense for the size of that impact and whether it will affect both high-speed data and video? And then in the quarter, the 24,000 high-speed data adds, sort of unadjusted for the -- or I guess, adjusted for the OSS issues, was stronger than what we've seen over the past few quarters. Could you talk about the underlying environment there? Or what drove that number above where it has been typically?

Sure. Listen, on the promotional roll-offs, as you know, historically, depending on when we're going through promos, the shape and the size of the promo looks a little bit different. So you sometimes see a 1- or 2-year or an 18-month or a 3-year promo. It just happens to be that we're hitting a vortex of a lot of the promos rolling off in the second half of this year. We saw some coming in Q3, but we're seeing an acceleration just in Q4. It's a significant percentage increase relative to last year. But going into 2020, we're going back to a normalized level that we saw in 2018. So this is really a onetime effect that we expect to see. I think, last year, in Q4, we were at like plus 7,000 unique customers. So we're not seeing a massive degradation expectation, but we do expect to see a net loss in Q4 relative to last year, which we were at plus 7. So it's a -- it's really just a one-off there. In terms of broadband, listen, we continue to see very, very strong broadband demand. We continue to deliver the availability of higher 1-gig speeds in the Suddenlink footprint, and we continue to upgrade nicely on the Optimum footprint: One, for 1 gig on an interim basis, which we'll be able to make available very shortly; and then secondly, continue to drive the build-out of the fiber-to-the-home project. I think we're just seeing continued very, very strong demand in both Suddenlink and in Optimum for broadband and higher speeds. We're not seeing a significant slowdown in that even though we have very strong penetration levels in Optimum. But particularly, the Suddenlink footprint, the penetration levels continue to rise nicely, and we continue to creep and take some market share in the Optimum footprint. So nothing special there. We're not going very aggressive on pricing. There's not aggressive pricing mechanisms there in any shape or form. I think some of our peers are a little bit more aggressive on the promotional side. So I think we're outperforming here relative to expectations.

Your next question comes from Philip Cusick with JPMorgan.

Following up on mobile. We've seen that marketing has picked up. You said you haven't ramped up the online yet. What's the sort of timing on that? And did you see October adds running faster than September? And then how should we think about EBITDA next year? Is that -- could that be positive? Or you think it really depends on growth, could be probably still negative if you're growing quickly?

Yes. So on mobile, first, week-over-week or almost day-over-day, we're seeing better trends. So that really is a function of 2 things. One is better performance of the online platform; and two, better training of our personnel. And so that will continue to drive, in our expectations, a continued increase of performance week-over-week going into year-end. In terms of the launch of online sales of handsets and other distribution -- opening up other distribution platform, we're going to hopefully be ready in the next 2, 3 weeks on the handset sales side, which will start driving a lot more volume, in our estimation. We see a tremendous amount of traffic on our sites, but they literally stop ordering when they can't get any handsets. So we know that that's going to drive incremental volume. And then secondly, we're in the process, and we have been already opening up some inbound calls -- seats, but we're going to open up a lot more inbound call seats in the near future and open up some of our other distribution channels as training increases. So we're really getting ready for a year-end push here for both on the marketing side and the volumes. In terms of EBITDA next year, I think this is really going to be a function -- we're going to make these decisions on a quarter-by-quarter basis as to how much money we're going to pump into our distribution channels and our marketing based on performance. If we think that we can go out there and get attractive volumes, even though we're going to lose money in the near term, then we'll go after that. If we think that it's more important to get to an EBITDA breakeven as soon as possible, then we'll make that adjustment as well. I think we've given guidance externally that we think that we're going to lose probably around $100 million of EBITDA in the first year after launch. So we'll update you accordingly, if we think that's going to be better or worse, depending on how we're seeing the volume trends.

Good. And good luck to Charlie.

Why don't you call him up?

Your next question comes from Craig Moffett with MoffettNathanson.

Dexter, I wonder if you can just dig in a little more into the economics of wireless. You've said repeatedly that you can make money at the $20 price point, which I presume means gross margin positive and contribution to overall fixed costs at scale can be positive. Does -- how much of that is a function of you've got a very good contract and how much offload? Can you just give an estimate of how much traffic you think you can offload on to the strand-mounted small cells? And then a related question is, as you think about the fiber build, can you just talk about the role of the -- I guess, what I'd call the freed-up coax? And is part of the magic of the fiber build actually to use the coax network precisely for that purpose, for strand-mounted small cells as a transmission medium because it's cost effective to put the radio equipment on?

Sure. So on the economics of wireless, listen, I think you probably asked me the question, some of your peers many times. We've been relatively mute on giving more precision. But yes, clearly, we believe we've got a very attractive contract on a relative basis to our U.S. peers, in terms of being able to -- if you were to simplify it to a cost-per-gig type of a metric. And we think that we are gross margin positive on our unlimited packages, right? So -- and the way we think about overall traffic is -- we obviously know what people's trend expectations are in terms of data usage on wireless. That goes between anywhere from 6 to 8 gigs. I think from our perspective, we're seeing numbers that are in line, if not better than that. So less usage than 6 to 8 gigs. And we're probably offloading onto our out-of-home WiFi network about 1 gig. So if you just do the math on that and you try and back-solve into what you think we're paying to our partners over at Sprint and AT&T, we feel that we are gross margin positive on every subscriber. And going forward, as you mentioned, on the volume side, at some point, we're going to contribute to the fixed costs related to OpEx here and be EBITDA positive as we get higher volumes. On the fiber builds -- sorry, go ahead.

No, I was just going to -- if I could just clarify, just to make sure I understand though, Dexter. You said that you're offloading about 1 gig onto your out-of-home WiFi network. But if I understand the contract correctly, what gets offloaded onto the strand-mounted small cells that Sprint put up also doesn't count against your usage levels. Is that right? So is that included in that 1 gig? Or is that...

No, the 1 gig is our WiFi network, right? So it is the Optimum WiFi network that we are not sharing with anyone, and it's our economics that are being saved.

But don't you save the economics when it goes over the small cells that were on your network?

No, no. That's just a densification in the lighting up of the spectrum. It has nothing to do with the WiFi itself.

Got it. Okay.

So just to go to your second question. Yes, I mean, clearly, we have still a very attractive coax network that we've overbuilt and continue to overbuild. One of the usages would be to accelerate small cell mounted strands onto it. That's been a very successful experiment that -- experience that we've had with Sprint. I think we would be open to having that dialogue with third parties. Going forward, there's a lot of things that we can do with that network as we free up capacity more and more.

Next question comes from Brett Feldman with Goldman Sachs.

Yes. And I just want to clarify on the subs. When you were talking about a potential decline in the fourth quarter, I think you were talking to your total customer base, not necessarily the broadband base because it does sound like there's considerable momentum in there. So if you could just maybe help us get that right? And then, typically, when you do see people come off promos, you get a little bit of an ARPU lift. So I was hoping maybe you could just let us know what the moving parts are in ARPU as you think about the product mix, exclusive of mobile?

So on the subs, I mean, listen, we don't -- we still are 2 months away from quarter end. But we're just kind of using a basic rule of thumb as we look at the volume of roll-offs that we're facing, that we'd expect just based on experience, what the churn numbers could look like, right? So I don't think it's right for me at this point, Brett, to tell you what I think our RGU performance looks like. But I think it's fair to say that almost every single one of our unique customers takes broadband today. So it's almost a one-for-one correlation. So if we're losing unique customers, the anticipation, in my view, today would probably be that we will lose also some broadband RGUs on a year-over-year basis. In terms of ARPU, listen, the mix continues to shift, right, in terms of the product mix. And so where we've seen maybe some pressure on ARPUs just driven by the video performance, whether it be cord shaving or people taking smaller packages, we continue to see good strength and resilience on the gross profit numbers because the mix is shifting more to broadband and people upselling on the broadband side. So I think Nick would be -- would love to spend a lot of time with you on your model going forward and giving you more granularity, but I think that's about as much I'll tell you on the phone call.

Your next question comes from James Ratcliffe with Evercore ISI.

Two, if I could. First of all, can you talk about the status for the Lightpath process and where we are on that? And secondly, on Altice One, you highlighted that. Can you talk about relative churn you're seeing for Altice One versus customers who don't have the Altice One platform? Just one housekeeping follow-up to the previous question. Just talking about losing broadband RGUs on a year-over-year basis or a sequential basis?

So just sequential, just to take that last question off the ticker. But on the first question regarding Lightpath, listen, the update is very consistent with what we've said over the past several months, which is it is an ongoing process, which is we're getting quite a bit of inbound calls from third parties. There are several parties who've done a lot of work who are putting together interesting offers, let's call it, but we don't feel compelled necessarily until we see a really compelling offer, right? So I think we're in the short strokes with a couple of people in terms of them presenting us proposals. And so we'll continue to have those discussions. But given that we've had a nice run-up of the stock, given that we don't need capital per se, and it's really about the quality of the partner and the value arbitrage that we could extract, I think we have to be just very, very thoughtful about how we proceed in this process and react accordingly, right? So there's nothing more to say. There's no update here other than to say that we continue to have dialogues with people. And we'll make a decision at some point in time, whether or not to pull the trigger. On the Altice One relative churn, I don't think we've been public about talking about what we think the relative churn is between the legacy boxes and our existing boxes, but we can tell you that it's down. We could also tell you that the live TV viewership is higher on the Altice One box than it is on the older boxes. We could also tell you that the usage of the apps are -- about 40% of the Altice One customers use the apps. And similarly, the usage of the voice remote continues to increase very, very nicely. So all of the things that we would anticipate with the investment in Altice One are coming to fruition. I just -- we haven't been public about talking about relative churn numbers.

Your next question comes from Michael Rollins with Citi.

Two questions, if I could. First, on the wireless side. Could you just expand in terms of what you're seeing on the mix of wireless -- interest in wireless activations between those that might be in the region where you have cable footprint versus those adjacency areas where you can serve the customers, but you don't actually have the cable footprint there? And then secondly, if customers are coming off promotion and are leaving your service, where do you see them going the most?

Well, listen, on the first question, listen, the mix is today, given the volume numbers are relatively small, is heavily weighted towards existing customers. Those are the ones that we can reach very easily, where the brand recognition is very easy as well. But as we ramp up advertising and all of our sales channels get opened up, and our e-commerce site is fully operational, we'll start targeting, let's call it, less dense residential areas such as, let's call it, Manhattan, right, as a place where we have Lightpath, but we don't have any residential customers. And so we already have billboards and digital ads and some broadcast ads in the Manhattan area. But we'll start accelerating a lot more of that advertising once we have all of our sales capabilities up and running. In terms of the promotions where people are rolling off to, I think it really depends, obviously, in terms of what regions we're talking about. But today, someone who is rolling off by nature in the Optimum footprint is tending towards going towards FiOS, if that's an option for them. For those people who are in non-FiOS zones in the Optimum footprint, it is really a retention effort, a change of the way he is subscribing to his business and what package he's taking, right? So that's more of an economic discussion as opposed to a pure churn discussion.

Your next question comes from Jonathan Chaplin with New Street.

Dexter, I'm wondering -- I know it's early days on the wireless product and volume hasn't been huge so far. But I'm wondering if you could give us some feedback on what the broadband attach rate has been for people coming in through a wireless channel. So for the guys that don't take broadband from you, what kind of pull-through are you seeing because of the opportunity for them to drop from $30 to $20 on broadband? And how do you think that will evolve when you push this more aggressively in -- across your entire footprint, through all of your channels?

I think you're right, Jon. It's a little too early, given that we're just about 2 months into the launch. And we still haven't opened up all of our channels, and we still haven't really targeted noncustomers per se, right? I think as we go broader, which is really, let's call it, Black Friday into Christmas season and going into the first quarter of next year, that's going to be something that I'm happy to share with -- I'm sure Nick will be able to give you some insights as to what we're seeing. But the anticipation is right, right? I think your question directionally is correct. We'd expect the attachment rate to broadband to be high. We expect churn rates to come down, whilst at the same time being profitable on a stand-alone basis in the wireless product. I think what will be interesting to see is, as we think about maybe playing around with our price points, is to try and incentivize noncustomers to become customers. And do we become more aggressive or less aggressive in terms of the disparity between the price points? I think that's something that we'll consistently think about.

Dexter, if you look longer term, how do you think this could change where terminal penetration for broadband in your footprint ends up? Do you think it could move where you would have sort of naturally stalled out in terms of broadband penetration materially?

Well, assuming static competition, you'd expect us to take market share consistently, right? I mean the product performance has been very good. It continues to get better. There are some software-related issues that we're working through with some of our core network and other OEM providers, but they're minor. But the performance overall has been very, very strong. And so it's really going to be about price points and making -- getting mind share and brand awareness out there. But we're going to have -- just on the Optimum footprint, we'll have a fiber-to-the-home -- a true fiber-to-the-home network with the best performance possible, and probably one of the stronger networks in the Tri-state area, whether it's the Sprint one today or even better with the new T-Mo network going forward, if that deal goes through. So we'd expect us to be able to have very, very strong attachment rates between wireless and broadband going forward.

Your next question comes from Marci Ryvicker with Wolfe Research.

Are there expenses that cross over from mobile into the core business, whether it's marketing or anything? Or are you able to segregate 100% of the wireless cost to wireless? And then secondly, as you're adding the streaming services like Amazon Prime and Netflix, what impact are you seeing on the video side of the business and on the broadband side of the business? Presumably, video maybe coming down and broadband going up. Just curious about that.

On the first one, listen, there's clearly crossover expenses. I mean there is a dedicated mobile team. There is a dedicated residential fixed line team, but we do spend, let's call it, marketing dollars and branding dollars and Wifi CapEx dollars somewhat in unison. We know we need to continue to densify our WiFi network, and we do that. We were going to do that even if we didn't do mobile, but some of that CapEx gets allocated by definition over to our mobile business. I don't know if MyTime, per se, if some accountant is using MyTime allocated, but I'm assuming that there's going to be some type of head count allocation hours used amongst corporates that are associated with mobile and get just allocated there. So as you know, Marci, we run a very tight ship with very few layers. So I think in many respects, the cost that we are spending on mobile per se probably are too much fully reflected, right? We would have been spending a lot of those costs in any case. And so we're just allocating some time from personnel and some CapEx, which we would have spent anyways on other endeavors. So it's probably to say that the loss numbers that are associated with it are probably higher than what they would be truly if you ran it on a stand-alone basis. I think on the effect of some of the OTT players, what we are seeing is clearly that people are using the Altice One platform to access Netflix as opposed to changing the input button on their TV to access it either through their Apple TV or directly through their smart Samsung. That is allowing people to stay within our ecosystem much longer. That has really been also the benefit of the Netflix button. It's been so easy for people to shift to it. It's something that, I think, intellectually, we resisted because we thought that it would potentially promote a brand that's not ours and take away from what we were trying to achieve from a branding standpoint, but it's been a phenomenal success from a consumer standpoint. So people like it. And so we think the usage of OTT platforms that are tethered to our Altice One platform has been higher than it has been if you got it in 2 separate platforms or you had to basically shift to another input button. In terms of our broadband, you're right, right, which is that the basic statistic is if you are a cord shaver and a broadband-only, you're using about twice as much data download than if you were a bundled customer with a linear cable bundle, right? So people are using, obviously, broadband more and more for video. And all the incremental usage that you're seeing even from bundled customers, a lot of it is video related.

Your next question comes from Doug Mitchelson with Crédit Suisse.

I wanted to ask you, Dexter, about the balance sheet strategy. But first, I guess, I'm feeling a little bit dense on the promotional roll-off. So the promotional sort of sub base that you have now different than a couple of years ago. I'm looking back through, I guess, broadband net adds, in particular, just sort of seasonally and over the last, like, 3 to 4 years. And I'm not really sure where I see the promotional benefit that you got. So I'm trying to figure out when the promotional subs came in that are now going to have such a dramatic impact on the fourth quarter of '19. And then I'll ask you the balance sheet question.

Yes, I think it's really -- the time frame is back-to-school, right? So -- and in the back-to-school periods, we're running various different promotions depending on the years. I can't remember, and shame on me, just on terms of what the promo was exactly last year and the year before, but we are seeing an overlap, unfortunately, of the promos, and that's what's really driving this Q4 effect.

So it's kind of a combination of 1-year promos and 2-year promos rolling off at the same time?

Exactly. We're at kind of the vortex of a bunch of promotional offers that happened in the back-to-school period and going into the Black Friday period and into Christmas, which changed in reaction to competition or just in terms of us coming up with a new product idea that was different from the year before. And so that is just -- we're at a onetime effect because as we fast forward to next year, and we look at what we think the volumes of promotional roll-offs are, they are perfectly normalized with numbers that we saw in 2018.

Got it. And then on the balance sheet, you've built a lot of flexibility in the next sort of 5, 6 years by pushing the stacks out. And I imagine there's a little bit of a cost to pushing the stacks out that far. And I guess, my question for you is, why are you building that much financial flexibility in your balance sheet?

Well, I mean, listen, I think that cost of capital on the debt side continues to be very cheap. And the availability of long-term capital, effectively almost permanent capital, is attractive at very cheap rates. I don't think we are building, let's call it, capacity for anything per se than -- other than optimizing our cost of capital and trying to maximize return for our shareholders. There is, obviously, a desire by a subset of our shareholders and maybe the credit agencies for us to delever. But in this type of interest rate environment and where we think our stock continues to be undervalued, we do like the fact that we're going out there and borrowing very cheap debt to go out and retire expensive equity.

Yes, that was my back door way of trying to ask if there was other Suddenlink opportunities out there or whether our focus should be on buybacks, but...

Well, listen, I think there is -- I've been open, which is -- the best use of our capital continues to be M&A, but there needs to be a seller for us to be able to buy something. And so we'll look at stuff. There are small systems we're looking at. We'll hopefully be announcing stuff at some point on small systems. But when something larger comes out, I'm sure you're going to know it at the same time we're going to know it. And you'll call us and ask us whether we're interested. And by definition, we are interested in most systems out there.

Your next question is from Bryan Kraft with Deutsche Bank.

I wanted to ask you 2 questions. One, can you talk about the economics of the $70 for life, double-play offer that you've had in the market, I think, since September? Specifically, just how do you manage the profitability of a permanent promotional price point that's that low, given the annual inflation in programming costs? And then secondly, I just wanted to ask a couple of numbers questions. Can you tell us what the right interest expense run rate is going forward? And also, any update on when you expect to become a full cash taxpayer or a material cash taxpayer?

So I'll take the first one, and I'll hand it over to Nick and Mike on the second. Listen, on the double-play offer, it obviously excludes price adjustments relating to equipment and sports rights and broadcast fees. So those are excluded. So as we look at -- to your point exactly, on the programming adjustments that we see an inflation, we're going to continue to be able to pass those through to them. What we won't do is change the makeup or the price points related to the packages itself. So if you have 200 megs, you're not going to be charged more for your 200 megs ever for the rest of the fact that you stay in that exact bundle. Same thing with your package. Your package itself will not change. Now the logic behind it is pretty simple, which is, one, this promotional roll-off situation is somewhat unsustainable, where you're attracting customers for 1 to 2 years at attractive promotions and then jacking up prices by $15 to $20 after the roll-off. And not only does that lead to a retention effort where you're probably spending quite a bit of money in retention, but the customer satisfaction or dissatisfaction and the amount of phone calls that you get into your call center really drives a lot of customer contact, which is negative and, in many respects, unproductive. And so we're more of the thought process on that particular promotion, which is at some point in time, this customer is going to want to change the package, whether that be on broadband speeds, whether that be on picking up other channels or reducing the amount of channels. At that point in time, it resets obviously the package. And that's really the anticipation. That's the experience we've had with these types of promos is they are attractive. They give people peace of mind. There's a subset of customers who'll keep this offer forever and be very, very happy with it, and they'll never call us, which is fantastic. So NPS scores go up and customer contact numbers go down, which helps our OpEx and our EBITDA and our cash flow. And then there are those that are sitting there with very good peace of mind, but at some point, they're going to say, listen, I'm not at 12 connected devices anymore, I'm at 24, and I'm not downloading 280 gigs anymore, I'm downloading 500 gigs, so I want to change. And at that point in time, we switch them into some type of other type of product. So that's really the genesis behind it. I don't know whether we'll keep it for long periods of time. But the reception has been very good from consumers, which is why we're -- we feel good about our gross add numbers. It's really about disconnects in Q3 and Q4.

As far as your second question goes, Bryan, I think you were asking kind of an annual interest expense run rate going forward. We talked about having a weighted average cost of debt right now of 6.0%, and that's on about $22 billion in debt. So I think if you do the math from there, you'll be in the right place. On taxes, we have NOLs right now that will take us through the end of 2020. Somewhere in 2021 at the current time is when we would expect to become a full cash taxpayer.

Your next question comes from Ben Swinburne with Morgan Stanley.

Dexter, just a couple. First on the OSS migration. Do you have -- I think you've talked about sort of singles and doubles from here, but are there other kind of chunky synergy opportunities beyond this one that you look out over the next couple of years? Or is it more incremental? And the way I understood it -- obviously, I might be wrong. I thought you lost sort of a week of customer adds because you had systems shut down. I'm sure that's wildly oversimplifying the situation. But I guess I would have thought those would have just slipped into the fourth quarter, which doesn't seem to be the case. So I was just wondering if maybe you could help explain a little bit the adjustment that you guys are making, so we can understand.

No, listen, I think you're spot on, Ben. So just to answer your second question first, which is, there was a full week of installs that were not able to be made. And yes, that shifts into -- a percentage of that shifts, so you do lose customers on a time delay factor, right? So...

Right. Don't want to wait, yes.

Yes, they don't want to wait, right? So they'll go somewhere else, and they won't want to wait. But what we're talking about in Q4 has nothing to do with gross adds, it has all to do with disconnect volumes, right? So we are seeing the gross adds as expected in Q4. It's just that we anticipate, based on the first month and the expectation as we've seen historically on volumes of promotional roll-offs, to see a much higher churn rate related to promotional roll-offs in Q4.

Anything about your synergies?

Yes. Listen, I mean, we keep on getting asked that question because you want to know how to run your margins. I think there's 2 things you need to really focus on. Obviously, the mix is changing on products. And so let's call it, there's some fake margin improvements, which has nothing to do with OpEx related. It's just about customer mix. And then secondly, on the OpEx side, we are continuously looking to optimize and allocate more efficiently our capital, right? So you can imagine we're in the middle of budget period right now, and that is a discussion as to how do we continue to allocate capital primarily to ways to run our businesses better, which could help us disintermediate costs. And so there will be a program in place, for sure, for 2020.

And then I just want to ask on the third quarter results, for you or for Mike or for either of you. ARPU -- residential ARPU, I think, was down about $1.30 or so Q-on-Q. I know you comped last year's price increase, but it wasn't obviously why it would have been down sequentially. I don't know if there's anything to call out there, pay-per-view or something else weird. And then I didn't know ...

So there is some pay-per-view and there is -- and then there is -- I think there was a fight comp in Q3 last year that was not here this year. I also do think, just by definition, you're not seeing it in our gross profit numbers, but you do see a little bit on the revenue numbers as -- even though our subscriber numbers on video RGUs look better, you do continue to see a deterioration of the video product in general, which is a disproportionate amount of your revenue and your ARPU. And so when you're onboarding people at $110 to $120, but you see people cord shave $70, $80 on video, it's just a mathematical equation that the whole sector effectively is going through.

Yes. I just didn't -- this is kind of housekeeping. I don't know if you guys knew how much acquired revenue you had in the third quarter. I think Cheddar -- you had a full quarter of Cheddar this quarter, and if you knew how much that contributed?

Probably around 8% to 10% -- 7% to 8%, somewhere in there. We also had an outage if -- I don't know if you knew this, Ben, and I know you -- because, right -- I think it was a Friday of the U.S. Open, we had a massive outage here, which led to quite a bit of credits that we gave back to customers. So that's also -- that hit ARPU numbers in Q3.

And the last question that we have time for today comes from Kannan Venkateshwar with Barclays.

So I have a couple. First, on the promotional side, Dexter, if you could just help us understand, what is so different about the promotions that you ran over the last couple of years, which gives you confidence that as you go into next year, the promotions that roll-off from this year don't really have the same impact in 2020 and beyond? And secondly, when we think about your broadband revenue growth, you guys have been growing faster than both Comcast and Charter, and price has been a big part of it. But I think some of it is basically just speed upgrades. If you could just help us understand the different levers to keep this revenue growth going at this pace. And how much of your price -- or the ARPU growth in broadband is due to promo roll-offs versus upgrades? That would be very helpful.

Sure. Listen, on the promotional side, roll-offs, we know in our database, in our CRM, every single package and when a customer comes off a promotional roll-off, right? So it just happens to be, again, that historically, in 2017, 2018, we may have had an excessive amount of 2-year promos, and last year, we may have had a lot of 1-year promos. And so the culmination of not staying consistent on a time period on promos leads to some overlap of promotional roll-offs that are coming at the same time. So it's just purely mathematical in many respects that we can see the volumes of when people roll off. And as we fast forward to next year, because we know we've been pretty consistent with how long our promos have been going on in terms of the period of time before we did price for life, so before the summer, and so we can already see that we don't have that type of overlap situation in 2020 and 2021. So it's really just for whatever reason, during the back-to-school and going into Black Friday and Christmas over the last 2 years, we happen to have shortened or lengthened promos that are leading to an overlap of those roll-offs. And then in terms of your broadband revenue growth question, it is almost uniquely driven by price, and not price push, but price pull. So it's really about customers upgrading consistently here. And as people go for less bundled products, obviously, the stand-alone data product is a higher ARPU than the bundled data product. And so we are consistently going to see, and we've seen this quarter over quarter over quarter, double-digit growth in broadband revenue, which is really driven by some volume, but a high, high percentage of -- super majority percentage of it is driven by price, which are people continuing to drive up the product road map. And so given that we are opening up 1 gig in the Optimum footprint, and we're averaging about 200 megs in the Optimum footprint, we continue to have a long runway there. But more importantly, we're opening up a 10-gig capacity on our fiber-to-the-home already next year, which will allow us, obviously, to have even a longer product road map going forward on broadband revenue growth.

Can I ask a follow-up question on that? I mean your speeds are at 200 megs right now. But from an application perspective, there aren't really applications outside of gaming which really need that kind of speed in theory. But as you roll out 1-gig and 10-gig capacities, how are you thinking about the pace of these upgrades? More recently, as you have moved more towards 200, has the pace of upgrades slowed? Or does it remain consistent with what you guys saw maybe a couple of years ago when you guys were at 50 megs?

Yes, I think you can look at the chart, I don't know what page it's on, but it's a pretty straight 45-degree line that consistently sees the same revenue growth trends, both on data usage and on broadband speeds. I understand what you're saying relative to the usage necessities of broadband speeds relative to the applications available. But frankly speaking, that was the same argument we heard 4 years ago when we were at 50 megs on average, which is why do you need more? Our friends over at Netflix say all you need is 2 megs or 5 megs in order to download. But the -- there's 2 things that are happening. Obviously, device -- connected devices continue to increase exponentially. Two, everyone is wireless at home pretty much and watching video on a regular basis at home. And thirdly, applications continue to drive larger and larger usage. So you're talking about certain applications like gaming, but actually, security cameras, as an example. When you tether your security cameras to your broadband, it takes an inordinate amount of capacity -- of your broadband capacity. So there are going to continue to be more and more applications, and we're seeing that, where consumers either feel the necessity to upgrade because they want better speeds, right? Why do people buy Ferrari as opposed to Toyota? Because they want to go faster, not because it's a better car necessarily. But I think people have that perception that there is going to be a better performance if it's more expensive and they have more broadband speeds. And secondly, really, it is -- the reality of it is that there are more and more applications that are driving the need for more broadband speeds. And so I think as more and more products are available at more attractive prices -- because 1 gig, when it got launched, was probably in the multi-hundred dollars per month. Now you're seeing promotions on 1 gig at the $70 level, let's call it. That is very achievable, particularly when people cord shave from $150 down to $70 or $80 by getting rid of video. They're upgrading their broadband speeds pretty rapidly from there on. And we're not seeing that abate or slow down in any shape or form.

I think that's it. Thank you, everyone, for joining and let us know if you got any follow-up questions. We'll catch up with you in the next few weeks.

Thank you very much.

Thank you.

This concludes today's conference call. You may now disconnect.