Thank you for standing by. This is the conference operator. Welcome to the Rogers Communications Inc. Second Quarter 2020 Results Conference Call. As a reminder, all participants are in listen-only mode, and the conference is being recorded. Following the presentation, we will conduct a question-and-answer session. To join the question queue, you may press star, then one on your telephone keypad. Should you need assistance during the conference call, you may signal an operator by pressing star and zero. I would now like to turn the conference over to Paul Carpino, Vice President of Investor Relations with Rogers Communications. Please go ahead, Mr. Carpino.
Thank you, Ariel. Good morning, everyone, and thank you for joining us. Today, I'm here with President and Chief Executive Officer Joe Natale and our Chief Financial Officer, Tony Staffieri. Our Chief Technology and Information Officer, Jorge Fernandes , will also be available during the Q&A session after the presentation. Today's discussion will include estimates and other forward-looking information from which our actual results could differ. Please review the cautionary language in today's earnings report and in our 2019 annual report regarding the various factors, assumptions, and risks that could cause our actual results to differ. With that, let me turn it over to Joe to begin.
Thanks, Paul, and good morning, everyone. I'd like to cover three topics in my remarks. One, I'll start by talking briefly about our second quarter results, which Tony will expand upon and provide additional detail. Two, I will also make a few remarks on our priorities during the quarter as we adjusted our business operations during this anomalous period. And three, finally, I'll share how we're thinking about the business as the economy starts to open up and the critical role our industry and our world-class networks play in Canada's recovery. Firstly, as you fully expected, our second quarter results reflected three full months of the COVID-19 economy. We saw notable impacts across all of our businesses as sales and new business activity essentially ground to a halt.
But as we said last quarter, these metrics are COVID-19 specific and do not reflect our underlying fundamentals, nor do they diminish our long-term growth prospects. Importantly, as you would expect, we took full advantage of the short-term extreme environment to re-examine each key aspect of how we run our business. We wanted to make sure the decisions we're making would set us up to power out of this difficult period. COVID-19 did not change our plans nor the course we were on. Instead, it greatly accelerated the pace of change. We are doing things today that we thought would take many months or quarters to accomplish, and the business will be stronger as these changes become permanent modes of operating. All business units were impacted in Q2. In wireless, all metrics reflect the impacts of the economic shutdown as customers isolated and stores remained closed.
We estimate that industry sales volumes were down by 80%-90% in the quarter. Customers shifted from wireless usage to home internet usage. While metrics like churn were down to a record 0.77% and phone subsidies were down about 45% on a year-over-year basis, most other impacts put short-term pressure on our results. With roaming, for example, travel simply stopped, and roaming revenues were down approximately 95% from a year ago. As we have discussed during the past year, we knew overage fees were coming down as we proactively transitioned to unlimited plans. We saw additional overage declines during this lower usage period. The lack of new activations as many of our stores remained closed further impacted service revenue.
While we anticipate most of the COVID-19-related impacts will recover as the economy opens up, there were several positives in the quarter that point to the underlying strength of our business. First, subscription revenue is holding up very well and is flat year-over-year. While there were some customers affected by the economic impact of COVID-19, the number of customers moving to smaller plans has been in line with our expectations. Secondly, and supporting this view, the shift to our unlimited plans continues to be strong. We are now at over 1.9 million unlimited subscribers and have the most customers who are not paying overage fees of any carrier in Canada. This is an important accomplishment as Canadians look for value in the current environment as we head into a 5G world. Finally, we started to see some volume slowly come back as stores began to open up.
At the beginning of the quarter, 90% of our stores were closed. Today, nearly 90% of our stores have reopened with modifications to protect the health of our employees and customers. In fact, there were a couple of days in June and late June reminiscent of some of the stronger promotional periods we typically see in the back half of the year. It is still early days, and we'll see how customer confidence responds to the economy opening up, but these are encouraging early signs. In cable, our business was stable but felt some impact as we continued to provide free content and additional supports to help our customers through the period. Additionally, our strong presence in the new condo, new home, and Airbnb markets, which slowed during the second quarter, impacted our business. On the positive side, our cable subscription business remains healthy.
We're also seeing reduced promotions and discounting in our connected home business. Overall, we expect gradual improvements in these markets in the second half of the year. While representing less than 15% of our revenue and less than 3% of our total Adjusted EBITDA, our sports and media business saw the most pressure in Q2. The material loss of advertising revenue with the suspension of live sports affected the entire industry, including Sportsnet. The lack of game day revenue and in-stadium promotions from delayed Blue Jays baseball also contributed to a tough quarter. Similar to wireless and cable, we're seeing some positive signs. With live sports scheduled to come back, advertisers are calling, eager to participate in the return of live sports.
Our sports and broadcasting resources are an incredibly valuable set of assets, and their contributions to our business will recover gradually with this pent-up demand for sports entertainment. Live sports, above all other types of content, drives a loyal and permanent appetite by fans and audiences. This quarter, in particular during the global pandemic, our focus was on three things: one, keeping our employees safe, two, keeping our customers connected, and three, driving the right priorities and investments for the recovery and the future. To say that COVID-19 has permanently changed how we operate is an understatement. We've pivoted in the moment to ensure Canadians can continue to rely on our services, fast-tracking service offerings that we planned and launched in record time. We did this while most of our workforce worked from home.
22,000 of our 25,000 people successfully shifted to a work-from-home model in the second quarter, including all of our customer care agents. It was an important quarter for our customer care agents. We moved to a permanent work-from-home model for our agents in Ottawa after seeing positive customer and employee feedback. We'll be applying these lessons learned to other customer care sites. We also crossed a very proud moment this quarter with the transition of the remaining customer service positions to Canada. Today, all of our customer service teams across our brands are based in Canada. Our Canadian-based team members are experts in our products and services, and as members of their communities, they can relate to the needs of our customers, do a great job of serving them, and better support our lifetime value metrics.
Just as our team stretches across Canada, so too does our extensive physical distribution advantage. And while more of our over 2,500 store locations in Canada are now open, we are advancing our digital-first strategy, an important factor in our long-term growth. Digital sales adoption is up over 15% year-over-year. Over 90% of our five most common service transactions at Rogers are now conducted by customers online. Virtual assistants are helping more customers with routine requests. These conversations have grown by 130% year-over-year to over a million as AI technology continues to get smarter and better understand customer intent. This digital enablement and our continued customer improvements are why we continue to see fewer calls into customer care, down 20% year-over-year. We've also adapted and expanded a contactless, Pro On-the-Go service, a key market differentiator for us.
The service is now available to over 10 million Canadians in the Greater Toronto, Greater Vancouver, parts of Southwestern Ontario, Ottawa, Calgary, and Edmonton areas. We will expand to more markets this year. This personalized phone delivery and setup support service brings the store to a customer's front door at no extra cost. In the early days of the pandemic, we also enhanced our TV and internet self-installation service. This change represents a clear competitive advantage in our market. We already provide a 1 gig capability across our entire cable footprint to drive greater penetration, and this has significantly removed customer friction, including eliminating the need to schedule an installation appointment. Now, we can drive greater efficiency through enhanced self-install capabilities. This quarter, we also introduced a new virtual assistance tool for our tech support teams.
With that app, they can now solve many issues right away without needing to schedule a service appointment. We're on track to save our customers an estimated 400,000 hours of their time and save us approximately 100,000 service truck rolls this year. These changes have been helpful to serve our customers during the pandemic, but they will continue as we move through and eventually out of it. They offer new service advantages to our customers and offer significant cost efficiencies across our businesses. We're proud of these advancements, and our team members are feeling it too. Our recent employee pulse survey shows employee pride is at 93%, an all-time high, an important marker for the strength and resilience of our 25,000 team members across the country.
Even during the most disruptive business environment we have seen in our lifetimes, I want to highlight how proud I am of our company and our team members and how they've responded to supporting the needs of our communities. During something as life-altering as COVID-19, our teams felt it was our responsibility to help the most vulnerable in society. We launched Step Up to the Plate with the Jays Care Foundation to help fill 390,000 hampers of food at the Rogers Centre to get as many as 8 million meals in the homes of Canadian families in support of Food Banks Canada. We raised over CAD 1 million through the Hearts and Smiles campaign, selling T-shirts and masks with all proceeds going in support of the Frontline Fund to help Canadian frontline healthcare workers.
We connected vulnerable Canadians, including providing devices and free wireless plans in partnership with Women's Shelters of Canada, Big Brothers Big Sisters of Canada, and PFLAG to maintain vital social connections when people needed them most. We also recently launched the 60 Project. It's been 60 years since Ted bought his first radio station, CHFI, with a CAD 85,000 loan. To mark this milestone, we evolved our 60th anniversary to focus on ways we are giving back to Canada and investing in others. Volunteerism is more important than ever, and a key pillar of the 60 Project is the 60,000 hours volunteer challenge. Rogers employees and their families will donate 60,000 volunteer hours across Canada to have a meaningful impact in our local communities. Looking ahead, we're optimistic about the future and the underlying strength of our business and asset mix.
If I can recap, we are the largest wireless franchise in Canada, the biggest cable operator in the country. We own and operate our own national wireless network. We were the first to launch 5G in Canada and have the largest spectrum portfolio amongst our peers. We deliver the best network experience in Canada. Just last week, umlaut, a global leader in mobile network testing and benchmarking, awarded Rogers the best wireless network in Canada. This follows J.D. Power in April, ranking Rogers number one in the West and Ontario in its Canada wireless network quality study. Our media assets are focused on sports, and demand for the return of live sports is high. Pride is at an all-time high with our team members, and we have CAD 5.4 billion in available liquidity and a strong balance sheet.
Overall, we have a formidable set of assets and an incredible team activating them. We are very confident in the long-term prospects of our company and for Canada as we work to power out of the COVID-19 period. Just as our resilient networks provide the digital scaffolding during this health crisis, our country's technology infrastructure will underpin Canada's recovery. If connectivity was a lifeline during COVID-19, it will be the bottom line to Canada's recovery. Today, the digital economy is the economy, and our country's tech-driven recovery will require the right investment-oriented regulatory environment. This is one of the most important lessons we can take from this moment. We're part of an industry that has never been more critical to society and to our economy.
From powering new stages of innovation on Canadian soil to ensuring more small and medium-sized businesses have a fighting chance with an online presence to receive and fulfill orders, strong networks are essential to Canada's economic recovery. Thank you, and let me now turn the call to Tony, who will provide more detail on the quarter. Over to you. Thank you, Joe, and good morning, everyone. Q2 was, without doubt, the most volatile quarter we have seen in our business as we went from a hard-stop, standstill economy that required us to focus on the safety of our employees, customers, and communities to the start of what we hope is a sustainable recovery for our country. Canadians are doing the right thing to help each other get through this, and we're glad to see that hard work paying off.
I'll provide some of the COVID-specific impacts we have seen in each of our businesses, which were significant in Q2. We have not adjusted our reporting numbers for these impacts, but we wanted to give you the transparency on some of the dynamics during this quarter. Let me start with our wireless business. In wireless, service revenue declined 13% year-on-year, driven by approximately 90% lower roaming revenue due to global travel restrictions, the waiving of roaming fees, as well as a decline in new activations for both postpaid and prepaid services during the COVID-19 pandemic. On a year-over-year basis, these reduced volumes and the resulting reduction in various fees we typically earn, combined with the roaming decline, contributed to 7% of our year-over-year revenue decline. These declines are COVID-specific items, which we anticipate will recover as we move out of the pandemic environment.
Additionally, we saw a CAD 60 million decrease in overage fees. Over CAD 50 million of the decrease was a result of strong customer adoption of our Rogers Infinite unlimited data plans, and the remaining reduction was related to the COVID-19 pandemic as people stayed home and relied on Wi-Fi for data usage. Unlimited plans have done well and continue to have strong underlying fundamentals. COVID-specific items noted had a direct flow-through to our service revenue which was down 13% on a year-over-year basis. Loading was essentially flat as we maintained our discipline by avoiding aggressive price reactions to some of our peer promotions. We matched where needed but felt there was no need to drive aggressive promotion when our employees and customers were still concerned with the safety of their families and communities, and in particular when total market volumes were down substantially.
With the lower activity, churn dropped dramatically to 0.77% for the quarter. While this is lower than normal, we don't view any subscriber metric during this period as being meaningful to any long-term franchise value of our wireless business. Wireless EBITDA was down 19% for the reasons noted, as well as a significant majority of the total CAD 90 million incremental provision for potential bad debt exposure is reflected in this segment. We will continue to evaluate the economic environment and performance of our customers in the second half of the year to assess the need for any future bad debt provisions. Based on our current assumptions, we feel this quarter's provision will capture the vast majority of the impact based on what we see at this time, but we will provide any future updates if required.
On the handset subsidy front, total net handset costs on a cash basis were down CAD 80 million compared to last year, or about 44% year-over-year. Let me now turn to cable. Revenue was down 3% due to lower ARPA associated with some bundling packages, as well as the continuation of providing select free video and internet overage services to customers during the pandemic environment. Additionally, we have delayed some price increases. Homes passed, and customer relationships grew 2% and 1%, respectively. We remained focused on our connected home roadmap driven by our Ignite TV and internet product. Despite low activity levels in our markets, internet net additions were 5,000, and Ignite TV grew by another 18,000. With the flow-through items noted above, cable-adjusted EBITDA declined 5%.
We estimate that excluding the COVID-specific impact items, as well as the additional incremental provision in cable bad debt, Adjusted EBITDA would have been approximately flat year-on-year. Our media results continue to be significantly affected by the COVID-19 pandemic environment. Revenue was down 50%, associated with lower advertising revenue due to the economic shutdown. We also have significantly lower sports revenue, including at the Toronto Blue Jays. Adjusted EBITDA was down approximately CAD 100 million, reflecting the flow-through of lower revenue and some lower costs. Moving to consolidated results, total service revenue was down 16%, and Adjusted EBITDA was down 21%. Adjusted EBITDA includes CAD 90 million of incremental bad debt provision related to COVID-19. This provision represents approximately 2% of our receivables and is at the lower end of the CAD 50 million-CAD 250 million range we referenced last quarter.
We estimate the total COVID-related impacts in the quarter on revenue were approximately CAD 725 million, or 19%. For adjusted EBITDA, we estimate COVID impacts in the CAD 300 million range, or about 18%. To be clear, we have not adjusted our numbers for these impacts. We're just providing some transparency that may be helpful going forward. We invested CAD 559 million in CapEx for the quarter, which was a year-over-year decrease of 25% and reflected a CI ratio of 17.7%. The decrease in capital expenditures was driven by delayed expenditures and permitting associated with access due to the pandemic. We also continue to see improvements in cable CapEx efficiency associated with self-installed internet and Ignite TV. We generated free cash flow of CAD 468 million this quarter, a decrease of 23% year-on-year.
The notable decrease in free cash flow is associated with the lower EBITDA flow-through and some incremental interest payments this year. Our cash tax rate as a percentage of adjusted EBITDA was 5.8% in the quarter and should be in that same range for the full year 2020. Despite the short-term economic impact of COVID, the company's liquidity is very healthy at CAD 5.4 billion available. Additionally, our balance sheet is well-structured with long-term maturities and low interest rates on our outstanding debt. Our weighted average interest rate at quarter end was 4.23%, with an average term to maturity of 13.6 years. We recently strengthened our position with an additional CAD 1 billion of two-year funds at an effective interest rate of under 1%.
We ended the quarter with a debt leverage ratio at a comfortable 2.9 times, and we see our leverage position continuing in the range of 2.5 to 3 times for the next few years. We believe this is sound and reasonable, given the spectrum options on the horizon and the continuing downward pressure on interest rates. While Q2 was unique and had several short-term challenges, we have responded and emerged in a very strong position. Our business execution was disciplined and very responsive to the needs of our customers in this complex environment. We continue to have exceptional network reliability at a time when demand has never been higher, and we pivoted our operating models to adapt to the new and evolving environment.
As the economy moves forward, we are well-prepared and highly engaged to assist our customers and the nation as we gradually and positively move out of this environment. In terms of an outlook, let me provide you the same level of transparency we used in our Q1 call and provide a snapshot of how we are trending on some key forecast variables. We won't provide specific guidance because it's still too difficult in the short term to predict the various combination of factors that could impact our financials, but this color should be helpful as to how we are thinking about Q3. In general, we anticipate modest sequential financial and operating improvements in Q3 for each of our businesses as the economy starts to open up and live sports slowly resume.
Additionally, we expect to see some gradual cost efficiencies materialize in the third quarter from efficiency initiatives learned through this period, as well as benefiting from continued year-over-year reductions in CapEx and handset subsidy savings. More specifically by business, in wireless, June saw a notable recovery in loading as most stores were starting to open, and July is trending a little bit better as well. We do not know what back-to-school will look like as customers are only now slowly getting back to shopping, but the economy is opening up, and that should help in our and the industry's recovery. We believe ARPU in Q3 will be in the same dollar range as we saw in Q2. We were down almost CAD 100 million year-over-year in roaming, which significantly impacted ARPU, and we expect the same year-over-year dollar decline in Q3.
We do not anticipate roaming to recover in the near term, but as travel opens up, our roaming will benefit from the recovery. In terms of overage revenue, Q3 will be our first full quarter of year-over-year comparison since launching Unlimited, and we anticipate overage will be down CAD 60 million on a year-over-year basis, as we have previously highlighted. Even in a COVID environment, we have seen no material impact on the underlying fundamentals of our unlimited plans. Impressively, we are very close to the 2 million customer mark in unlimited plans. These plans continue to have higher ARPU, lower churn, and higher customer satisfaction. In less than a year, we have become Canada's largest provider of unlimited plans, with customers enjoying access to our premium national network. Canadians love these plans, and we anticipate ongoing leadership in this area as customers continue to resume their mobile activities.
Offsetting some of the roaming and overage pressure on ARPU, we expect to see some benefit in Q3 as economic activity and activation revenue picks up. However, it is still difficult to predict how active customers will be in the back-to-school period. In both our cable and wireless businesses, we continue to work with customers to manage bill payment terms if needed. We will continue to monitor the environment to see if additional provisions are required, but do not expect any additional provisions in Q3 to be substantial. We continue to see positive demand for our internet and Ignite TV offerings in this work-from-home environment. Loading should remain positive in Q3 as new condo and housing builds start to recover, and self-install in both internet and Ignite TV continue to grow. OpEx and CapEx-related installation and upgrade costs should also improve.
Capital intensity for our cable business should continue its steady downward trajectory as reduced volumes, self-installation, and construction delays continue, although to a lesser extent than Q2. In our sports and media business, we will likely incur losses in Q3, but the restart of the MLB and other leagues will translate into the resumption of some advertising revenue at Sportsnet. However, we expect Adjusted EBITDA to remain negative for sports and media until fans can return to watch the Jays live and drive game day revenue and advertising. Overall, cash flow and liquidity remain strong, and maintaining this financial strength will remain our priority for the rest of the year. Based on the current run rate for the first six months, CapEx for the year will likely be down approximately CAD 500 million.
However, I want to be clear that this reduction is primarily based on projects that have been delayed in the current environment. Our network and 5G development spend are full steam ahead. As the economy resumes its gradual recovery, we are positioned very well to drive growth with the best assets: a very strong balance sheet and a highly passionate and engaged workforce. As we have demonstrated in the past, we will use our leading market position, largest wireless company, largest cable company, and largest sports assets to create long-term value for shareholders. Let me now turn the call back to the operator to commence with our Q&A.
Thank you. We will now begin the question and answer session. To join the question queue, you may press star, then one on your telephone keypad. You will hear a tone acknowledging your request.
If you are using a speakerphone, please pick up your handset before pressing any keys. To withdraw your question, please press star, then two. Once again, to join the question queue, please press star, then one now. Our first question comes from Vince Valentini of TD Securities. Please go ahead.
Thanks very much. Let me ask you a couple of questions on ARPU, if I can. First off, if you're down another CAD 60 million year-over-year in the third quarter on overage revenue, correct me if I'm wrong, but you should be basically at zero by then, should you not? So that this will stop being such a big headwind as the crisis and then the migration to unlimited gotten you down to sub-1% of service revenue coming from overage now?
The second question, I'll just throw it out to you first so you can think about both of them. Thanks for those numbers. If I do the adjustment on the 7% impact you mentioned from roaming and activation fees and sort of COVID-direct things, plus the CAD 60 million impact from overage in Q2, I still come up with almost a 3% decline in wireless service revenue on a year-over-year basis, even backing up both of those items, which would suggest the underlying trend is still not great. I thought we'd been talking about higher MRR for people moving to the unlimited plans, plus maybe some benefits from lower equipment subsidies gradually flowing through ARPU numbers. Of course, you've had positive trade backs on a full year or trailing 12-month basis.
So, if you can flesh out a bit more why that underlying service revenue growth is still -3%, even with the two adjustments, that'd be helpful for everybody, I'm sure, because I'm sure everybody's a little bit surprised at that ARPU and service revenue growth number this quarter. Thank you. I'm interested. Tony, thank you for the questions. A couple of things. I'll start with the overage. We had anticipated that, based on our projections of number of customers on unlimited, if you were to compare numbers to pre-COVID that we were quoting, we would have been slightly higher in terms of number of customers on unlimited, albeit we're very close. We had expected about 2 million. We said at about 1.9 million customers on unlimited. And so some of the overage drag will spill over into Q3 as a result.
But the bigger reason is. Keep in mind the seasonality. And while we have the overage melt of customers moving to unlimited, there are also customers that would have continued to incur overage based on their existing plans and usage patterns during the summer months. We're not going to get that. And so, much like in Q2, I talked about CAD 50 million related to unlimited, CAD 10 million of that secondary factor. As we move to Q3, that CAD 10 million will be more in terms of we expect lost revenue, and the overage melt will be less than the CAD 50 million. So that's sort of the unpacking of that. And then secondarily, try to articulate the ARPU declines. And as you get to a net number of just under 3% or slightly less, there are a few other COVID-related impacts in terms of freebies that we provided.
And so that negative 3% moves to about minus 1%. And the minus 1% is in line generally with what we expected. We had said, based on pre-COVID, we were going to get into the second half of the year with positive growth in ARPU. And so Q2 was always expected in terms of underlying ARPU to be slightly negative. And that's where we stand. I would say it's less than 1% short of what we expected. We are seeing some what we would call ARPU growth pressure as customers just didn't get the volume coming in on higher plans. That always helps. And then I would say very subtly, customers looking to optimize their plans had a very small impact as well.
So all that to say, the underlying subscription revenue is solid and flat year-on-year on a dollar volume, and the underlying subscription ARPU is just slightly negative, but nothing that we're concerned about.
Thank you.
Thanks, Vince. Next question, Ariel?
Our next question comes from Dave Barden of Bank of America Merrill Lynch. Please go ahead.
Hey, guys. Thanks so much for taking the questions. I guess the first one for you, Joe, if we go back to the decision you guys made about the dividend policy and choosing not to have a growth policy but rather have more of a tactical approach to the dividend, is there anything about the degree of uncertainty that you're facing this year that makes you want to potentially reconsider the amount of capital you're allocating to dividends in the short term? And then a second question for Tony.
As you kind of gave us those numbers, thank you so much for the color as we kind of headed out of June into July. What kind of visibility do you have on kind of the mix of EIP versus subsidy customers, and if that's been affected at all by maybe the economy or the aftermath of COVID? Just some kind of color on where we stand in that transition.
Thank you. Sure. Dave, thanks very much for the question. First of all, let's start from the beginning. Our cash flow expectations for the year are intact, and we expect to land the year at roughly CAD 2 billion of cash flow in that zone as we have anticipated overall. And then so there's really no change from that perspective.
It's coming from the fact that we're doing better from a CapEx efficiency point of view in terms of self-install activities that we've quoted. It's doing better from some of the better unit prices that we're getting in the marketplace. There is a bit of a slowdown in terms of shovels in the ground around housing starts, and therefore some of that work has been delayed. Generally speaking, Tony quoted a circa $500 million lower CapEx. It really holds the cash flow position intact as a whole. Then in terms of our capital allocation policy, nothing's changed from that perspective. Our number one priority is investing in the business, investing in 5G, investing in the future of the business because the underlying fundamentals are still there and still very important to us.
If we have surplus cash, then we do have an NCIB that's out there and available to us. Should we choose to take advantage of it? And that would sort of be the second priority. Third priority, we'll look at dividends from a time-to-time basis. But so really no change from a dividend policy point of view. The second part of your question, David, I'm not sure I understood. Let me highlight a couple of things, and then I'll allow you to elaborate on it. We've moved to 100% EIP from the subsidy model. And so as a result, what you'll see on our balance sheet is a split out between contract asset and, in some of the details, device financing receivable. In total quantum, that has come down slightly as a result of the lower volumes. But I'm not sure if that was your question.
If you could please just clarify, Dave. Yeah, I apologize. I think what we saw kind of creep into the competitive landscape was that the subsidies were creeping into the EIP plan, and the phones were not being sold at full price but rather at a subsidized price, and I was wondering if you could kind of comment on how that's evolved or appears to be evolving as we get into the third quarter.
All right. Gotcha. So a couple of things on that. When we introduced, if you rewind the clock way back in July when we introduced EIP plans, the intent was to be on a roadmap to substantially reduce the amount of promotional discounting on handsets. We knew it wasn't going to be a quick turnaround, and as we fast forward to Q2, what we saw is promotional discounting down circa 20%-25%.
So keep in mind, some of the numbers you see in terms of discounting are relative to MSRP. There have been some changes to MSRP for some devices. And secondarily, there are incentives provided by OEMs, as you're aware. And so that provides or pays for some of that discounting that you might see in the marketplace. So overall, we're pleased with the way that's trending overall for where we're at.
Okay. Great. Thanks, guys.
Thanks, Dave. Next question, Ariel? Our next question comes from Tim Casey of BMO. Please go ahead.
Thanks. A couple for me. Tony, could you talk a little bit about the seasonality of roaming and how we should think about that going forward, given, as you mentioned, travel and certainly business travel doesn't look like it's going to come back anytime soon?
Just wondering how we should think about that both from a service revenue and even debt impact over the next four quarters, both maybe if you could from a relative perspective and also seasonality. Then second question, probably more for you, Joe, how are you thinking about your cost structure going forward? I guess where roaming is, one side would just be keep it as is and wait it out and wait for things to return to normal. But I'm just wondering if you think the new normal is going to be significantly different that you are going to have to make changes to your distribution base or other big cost items. How is your thinking evolving on that? Thanks.
With respect to the first question on roaming and the profiling, think about if we were to, you said before roaming was a roughly CAD 400 million annual business for us. Think about Q1 as being the lowest, followed by Q4. Then Q2 and Q3 are really the high points, Q2 having the most business-related roaming typically. Then in the summer months, flips over to consumer roaming at its peak. In Q2, I talked about roaming revenue being down CAD 100 million. It is about the same, maybe slightly more in Q3 in terms of year-on-year impact. So it'll be in the circa CAD 100 million-CAD 110 million range we expect for Q3. Tim, on the second question, we have a very active cost management program underway. In fact, it's been added all the way through COVID in the last number of months. Think of it this way.
There are a number of activities that I would have called business-as-usual, margin improvement, cost reduction activities driven by a number of things already being rationalized in our business, whether it's call volumes coming down 20% or digital adoption going up or some of the other sort of efficiency measures. We largely took a pause on those during Q2 and during the COVID period. We focused certainly on managing what I would call discretionary spend. Some of it happened for us as travel stopped and things of those nature stopped. But we managed discretionary spend. But some of the structural cost programs that we had underway were paused as we focused on, as I said earlier, focused on safety and well-being of our people, keeping our customers connected, and then thinking about the future and how we leverage these ideas and opportunities for the future.
The second category of cost improvement ideas that have been accelerated through COVID, and I think are substantial and material in nature. We went to a system of supporting field service with tools and apps from video chat to monitoring and analytical tools. We think that's going to save 100,000 truck rolls, as an example. 100,000 truck rolls is a pretty significant expenditure on an annualized basis. You add to that the self-install. We went from really almost no self-install in January, a very small pilot project, and now we're at 100% self-install. Our goal is to stay in that zone in terms of self-install and continue to leverage the benefit not just from a cost point of view but also from a customer flexibility point of view.
Think about self-install as a situation where we can make sure that the connection is working outside the home, either remotely or whenever it's convenient for our team to do so, and then send the customer their CPE, their equipment, for inside the home, and they can schedule a time if they need support. Otherwise, they could do it themselves. We've launched a whole bunch of apps to walk them through the installation process. If they get into trouble, they can, in the moment, hook up a video chat or schedule a video chat. And that's worked really, really well. We're never going back from that place overall. Another area is around channel mix. Out of necessity, we have moved more of, albeit limited, sales volume. Gross adds and wireless were down about 38%, but there was still reasonable volume to try some of these ideas and push them.
That is indexed more to digital sales and direct sales, where the COA profile is vastly different from some of the third-party retail and other channels that we have as a result. We moved the call center to work from home. Service levels and productivity have never been better from that perspective. In the case of Ottawa, for example, we decided to close the building that was coming up on end-of-lease. Now the customer service call center operations in the Ottawa area are on permanent work from home. We'll make that decision in a few other geographies as we kind of roll through. We have a series of these that are material in nature, born out of COVID or accelerated through COVID, that we're going to take action on through the piece.
I think it's fair to say that through Q2, we didn't really go after any structural costs. It just didn't seem like the right thing to do, and secondly, we were looking to see what the recovery might look like. To your point, Tim, if this was going to be a Sharp V recovery, it would have seemed like a bit of a crazy thing to kind of go student body left, student body right, but seeing now that the recovery may be a little more extended, ramping in the right direction, but we're not sure what's in store for us in terms of a second wave or what else might be out there, and therefore, there's much more focus on taking action on some of these structural costs and ideas that I just itemized, and you'll see those coming to fruition.
And they're just not going to help the back half of the year. They will help the structural profile and the margin of this business through the next long while. And then as the economy does recover around things like roaming revenue and the like of things that we've discussed, we'll get the full EBITDA flow-through of that recovery into a much better, more efficient cost structure. So there's a team working very hard exactly on this point. We've got a series of actions that we'll unveil to you as we make them happen.
Thank you.
Thanks, Tim. Next question, Ariel?
Our next question comes from Maher Yaghi of Desjardins. Please go ahead.
Thank you for taking my question. And thank you for all the information that you gave on the COVID impact.
I wanted to ask you on the cable side. I think you took longer than usual to implement some price increases on the internet. How much has that contributed to your -2% revenue growth here? You talked about the lower home phone pricing environment. What is driving these pressures at this point in time? Is it you're seeing economic pressures at your customer level that you're trying to alleviate by these new pricing promotions that you're doing? And my second question is on government funding. We've seen other companies in the media segment, for example, benefit from some support on the government side. Have you been able to access any of these fundings for your media business or any other segments that you're operating?
Thank you. Thanks, Maher. First of all, let me start on the cable question. We had a price increase in motion, and we halted it.
We halted it and didn't do it whatsoever. And we are now looking at the timing of that price increase, which will happen later on this year. So there's no question that contributed to the downward pressure on the revenue front overall. And there were some other pressures in the cable business that are temporary in nature. One is we gave some concessions to our customers during the COVID period. We removed overage caps on internet and Wi-Fi usage. We gave people a lot of free content. A number of things we thought were the right things to do. They're now gone from the equation, and they will kind of help to ameliorate those results overall. Maybe Tony, you can give a bit of a breakdown of that, and I'll come back and talk about the funding situation.
Sure. All right , just add to top up Joe's comments.
So the price increases across all our products, not just internet. And as we implement those in the fall, what you'll see is not only ARPA improved, but had we done it in Q or as originally planned, then what you would have seen is a much better profile in Q2. As you look forward for our cable service revenue business, you should expect sequential improvements in Q3. And as we approach the end of the year in Q4, the fuller impact of the price increase and the taking away of some of the freebies that you saw in Q2 come out. And so you'll see a healthier growth profile, albeit modest, in Q4, but it's the beginning of resumption of the strength of cable revenues.
On the question of the funding, Maher, we did qualify for and receive some funding for our media business that essentially ground to a halt. We availed ourselves of that funding for the reasons you would expect, the choice to support either furlough employees and have them go on individual subsidy or to keep employees here, take advantage of that support mechanism, trying to figure out in anticipation of when the games might resume. I think that trade-off worked out well for us because the games now are coming, and they're all coming at once, fast and furiously. Having people on standby and ready versus having to call them back from furlough was an advantage to us as a whole. It was really kind of, think of it as a flow-through.
It either would have happened directly for individuals or as a flow-through by keeping them ready for the games to come.
Okay. Thank you. Can you quantify that amount that you received?
Yeah. We're not disclosing that number for a few different reasons. But in many respects, it's irrelevant, as Joe said. The subsidy either would have been at the individual level or at the company level. And so, as I said, we're not disclosing it.
Okay. Thank you.
Thanks, Maher. Next question, Ariel?
Our next question comes from Drew McReynolds of RBC. Please go ahead. Yeah.
Thanks very much. Good morning. First, a clarification, I guess, for you, Tony, on the bad debt expense in the quarter, the CAD 90 million. Did I hear that correctly, that the bulk of that is in wireless?
That's right, Drew. The bulk of it is in wireless.
Just to clarify, it's the incremental provision that we booked in the second quarter.
Just for modeling purposes, are you able to just give us the numbers maybe offline or now by segment on the incremental CAD 90 million?
Yeah. Roughly CAD 80 million for wireless and just slightly above CAD 10 million for cable.
Okay. Perfect. Two others for me. First, on the back-to-school dynamics, not to get too much into the weeds here, but just can you just remind us what the normal biggest deltas are on the seasonal aspects of back-to-school just so we can kind of better understand what could or may not happen this year. Then secondly, probably you, Joe, on the 5G roadmap with the 3,500 MHz auction delay.
From your perspective, what does that do for your kind of goals here in the next year or so on 5G, both to the positive and negative, again, just related to that 3,500 delay? Thank you.
I'll start with the back-to-school. Hopefully, I got where you're going with the question. In terms of if volumes continue to be lower, then we'll continue to have the savings related to handsets. But similarly, some of the fees, like activation fees that we typically get and a few others along those lines that are not insignificant, they certainly impacted us in Q2, but those would then continue to impact us on a year-on-year basis in Q3. Based on the volumes we are seeing, we expect it to be slightly better than we saw in Q2 as volumes continue to rise.
But to the extent they're down, then that will continue to have an impact on service revenue and ARPA. I don't know if that answered your question.
Yeah. And sorry, Tony, just on the cable side, is there any dynamics there?
Typically, on internet, what you would see on back-to-school is internet volumes pick up. Again, with much of it looking like it's back-to-school online, we already have much of that captured in our base today. And so the incremental would be much more muted this year, we expect, than prior year.
Thank you. Drew, on the 5G, I'll make a few comments. And Jorge Fernandes , our CTIO, is here with us, so I'm going to ask him to help support with some commentary. So on 5G, it's full steam ahead in terms of our focus and development of 5G. As you know, we're the first to launch.
We also have a very advantageous position with 600 megahertz spectrum across the country. Jorge will talk about the deployment plans on that spectrum. Yes, the 3,500 auction regrettably has been delayed. But bear in mind that we have other spectrum available to us in the mid-range frequency. Jorge will kind of cover our plans on that front, including the tranche of 3,500 we already own.
Hi, Drew. Good morning. Thanks for your question. As you would have heard me talk about this before, 600 megahertz is indeed our foundational spectrum that we're using for rollout. Given that this spectrum has now mostly been cleared across the geographies for usage, we're rolling that out as we speak. We expect to have very good coverage using the 600 spectrum that we acquired.
As you know, in Southern Ontario, we have a particularly good advantage in that sense as well. As Joe mentioned, the fact that we have Ericsson as a single vendor, that allows us to use the dynamic spectrum sharing that I've talked about before as well. This is a great advantage for us because, one, some of the important flagship devices will support both 600 and spectrum sharing, which allows us to use some of our existing 4G spectrum to provide coverage and capacity for 5G without having to do major work on our network. Over time, as Joe mentioned, when 3.5 becomes available for wireless usage, we will then add that on to our strategy. This doesn't really change any of the plans that we've already communicated in the past. We feel very good about the plan that we have in place and we're executing.
Thank you.
Thanks, Drew. Next question.
Just one more quick comment on that. 5G is about the network. 5G is also about the commercial construct. I mean, part of our decision to go to unlimited, as we said earlier, we're close to 2 million customers on unlimited, is that we need a consumer construct that complements the capability and availability of 5G. It would be a shame to have a 5G network and have a 3 or 4G pricing construct with the overage considerations around it. So the two go hand in hand. In 5G developed markets across the world, that's an essential pairing of capability, network, and customer construct.
Thanks, Drew. Next question, Ariel?
Our next question comes from Simon Flannery of Morgan Stanley. Please go ahead.
Hi. Hello. This is Diego Barajas, filling in for Simon. Thank you for taking my question.
Going back to the incremental bad debt provision range you provided, can you just speak to some of the trends you're seeing both on the consumer side and the SMB side recently?
Sure. In terms of early on, we had anticipated that we would see, I don't know what word I'd use, but material amounts of either deferrals or inability to pay and would otherwise be disconnects. I would say what we're seeing in reality over the last, not only in Q2, but in the last few weeks of July as well, is much lower volume in terms of delayed payments or suspended accounts. So I would say that's trending much more favorably than we expected. And so in putting out our provision, we try to be prudent and conservative and try to capture, as I said in my notes, the vast majority of the risk.
So we see the incremental exposure as being very minimal as we look to Q2 and Q3. We'll have to see how some of the specifics unfold, but overall, trending better than we expected.
Thanks. That's very helpful. And then secondly, on the media side, can you maybe give some color on what the financial impacts may be with timing changes to the leagues as well as related to media, what you're seeing in the advertising market and outlook for the rest of the year?
The games have just recently been announced in terms of scheduling, etc. And so as Joe mentioned in his opening remarks, bookings have been solid in terms of advertising revenue. On the media side, subscription revenue continues or affiliate revenues continue to be solid. And so with the resumption of advertising, we're quite optimistic about the revenue profile in Q3.
And we'll see what Q4 brings. But to be clear, we still expect, when you think about the broadcast fee costs and the loss of game day revenues at the Jays, just overall, we still continue to expect a loss overall in media in Q3 and probably Q4 as well.
Great. Thanks a lot.
Thanks, Diego. Next question, please.
Our next question comes from Jeff Fan of Scotiabank. Please go ahead.
Thanks. Good morning. Hope everyone is doing well. My first question is just a clarification on the free cash flow being intact for this year. Regarding CAD 500 million of CapEx being down and the free cash flow is intact, it's sort of assumed that your consolidated EBITDA decline would materially improve as we get into the second half of the year. I'm wondering if that's correct, especially compared to the -20% in the quarter.
And then my second question is probably for Joe regarding the wireless retail environment. I'm wondering, given what we've seen so far, whether you've seen retail traffic actually pick up as you reopen stores through June and maybe in the early part of July. And if physical retail traffic doesn't pick up and yet there's some pent-up demand regarding phone upgrades and so forth, how do you feel about your digital platform to be able to enable activations? And I'm not talking about just queries or customer support, but actual customer activation going from start to finish and being able to address that potential pent-up demand. Thanks.
I'll tackle the first part of your question. I think a couple of things. One, in terms of specific profiling, we do anticipate EBITDA to be stronger in the second half for a few reasons.
Some of the revenue we talked about profiling will be slightly better in all three of our businesses. And two, once again, the bad debt provision that you saw in Q2 and weighing on EBITDA, we don't expect that magnitude to repeat itself in Q3 and Q4. And so that'll be a benefit. When you combine that with our CapEx outlook, the $500 million is an estimate. And so if we go back to the macro picture, we had originally anticipated to deliver free cash flow at just above $2 billion. We see a stable path to that. And whether some of the dynamics will vibrate within EBITDA or within CapEx, maybe slightly more than 500, slightly less. I don't know that we want to put that fine number on something that is a bit fluid during this environment.
I think the key message to take away is we're on track for continued solid cash flow delivery.
Thanks, Bill.
Jeff, on the wireless retail environment, as I said, in the last part of June, early part of July, we've seen a lot of good retail traffic. And we've had some volume days that are reminiscent of sale periods in the back half of the year as a whole. Mall traffic has been good given the circumstances. I would say to you that our factory capacity is down a bit just because of the conditions we're employing to keep customers safe and hygiene intact, etc. Think of it this way. Depending on the size of the store footprint, we limit the number of people in the store. So in some of the mall locations, we actually have lineups and people kind of come in as one person goes out.
Just the nature of COVID right now, we have to work through that sort of trade-off. So at the end of the day, our goal is to continue to execute well on the physical bricks and mortar side. And as you recall, we have a strong advantage on that front with 2,500 locations across the country. At the same time, to continue driving and investing on the digital and direct side of things. And I think on that front, we've got a good capability and a growing capability. The one issue that we've been managing well is what I would call customer authentication eligibility. And so I'm not talking about, as you said, service transactions. I'm talking about buying transactions online. The challenge in our business has always been authenticating the customer and driving eligibility understanding.
The teams have been working on that from well before COVID, from last year and driving hard on that front. When you think about a transaction where you're going to put a very expensive phone in someone's hands, it's really important you get the authentication and eligibility right. The teams have done a good job of that and managing the fallout on a daily basis, and frankly, it's going in a very good direction. You couple that with our other direct capability besides the web, and we think we've got a position of strength. Pro On-the-Go is a material advantage, we believe. Pro On-the-Go now is available in almost one-third of the country, and we're growing region by region as we speak the rest of the year.
That offers an ability for a customer not to even think about going into a store, but they can transact with us either online or transact over the phone, and then the store experience shows up at their door. That's unique to Rogers' capability, and it's going very well. The satisfaction scores are off the charts, and the economics are very compelling from a COA point of view and attachment rate on other auxiliary products, etc. We think it's a good capability to have as a whole. We've gone from, as an industry, we've gone from a very small percentage of sales happening online. The industry is typically operated in North America, not just Canada, at maybe 5%-10% of transactions happening online. You can think about that world right now. The non-store transactions are closer to 50/50 at this point.
The key, we believe, is to be good at both. Be good at both. We already had a physical bricks and mortar capability, and we're going to continue to invest in and drive capability on the direct side. That will serve us coming out of COVID as well and allow us to be more efficient, not just from a store footprint point of view and retail distribution cost point of view, but the COA is a remarkable impact on our business. In some third-party channels, we will spend CAD 200-CAD 300 dollars sometimes per, outside of any sort of subsidy, just truly on commission and fees, etc., to the retailer. During a promotional period, if you add a gift card to it, you can go north of that number. Whereas in the direct channels, it's sub CAD 100.
And therefore, there is a structural cost advantage available to us with the capabilities that we have in our building. So that's sort of the game plan as a whole, and we feel well equipped for wherever this goes. If this goes to a place where there's less mall traffic because people don't feel comfortable shopping, we see evidence of some back and forth in different markets in the US and California and places like that where it kind of opened up and closing or curtailing again, we're ready for either eventuality. And we've built our plans and our approaches to swing in either direction from a mixed point of view, and we're prepared for it.
Okay. Thanks, Joe.
Thanks. Thanks, Jeff. Ariel, next question.
Our next question comes from Aravinda Galappatthige of Canaccord. Please go ahead.
Good morning. Thanks for taking my questions. Two for me.
I think in the past, you've talked a little bit about sort of the inbound calling coming in from those segments of your subscriber base, getting some financial stress and looking to reprice. I know you're seeing that originally in cable and perhaps also a little bit in wireless. I was wondering if you can give us an update there. How material is that sort of component? And was that a meaningful element of the decline in Q2? And then secondly, maybe just touch on the promotional intensity as sort of the volumes start to come back. I've seen some sporadic promotions that have been, at least on the face of it, looked a little bit aggressive, particularly 20-gig plans around the CAD 65 level, maybe lower than that. Anything material to touch on there? Thank you.
Aravinda, I'll start with the first one in terms of call volumes and re-rating.
As I referenced earlier, we had expected the worst and were surprised by much lower volumes coming in in terms of reprice. And that's in both cable and wireless as customers look to optimize their plans and/or try to get into some of the promotional pricing that they may have previously seen in market. But I would say those were very minimal. They did impact our pool in both wireless and cable, but a very small, to a very small extent, much less than 1%. In fact, even less than half a percentage point in each of those. So the activity was there, but very small. Aravinda, in terms of the promotional intensity, our general stance during Q2 was this is not a sales quarter.
When you're facing the fact that 90% of your stores are closed and you've got thousands of people with nothing to do because they work in those stores, it seems counterintuitive to be aggressive around promotion. There were some promotional skirmishes in the quarter led by our competitors, and of course, we matched and we were right there every step of the way, but our stance was very much, let's focus on the basics. Let's work on our operating model, cost efficiencies, and managing cash flow and liquidity, managing the health and safety of our people, all the things we've talked about, and we're ready for the competitive environment to whatever degree it evolves. I would say that on the equipment subsidy EIP front, it has been a disciplined mindset from the get-go in January when the industry moved to EIP.
I imagine there will be promotional aggression points throughout the various periods as people try to understand what volume is out there and how to swing the volume their way. I would say to you that between our bricks and mortar channel and our digital channel, we're well equipped for that competitive intensity, always have been, and we'll continue to be that way.
Thank you.
Thanks, Aravinda. Next question, Ariel?
Our next question comes from Richard Choe of JPMorgan. Please go ahead.
Hi. I wanted to ask about the cost structure in the wireless business. The drop-off in service revenue has been an impact, but it seems like the overall cost structure has stayed the same. Is this something that could change over the next few quarters, or how long would it take, and how do you view it?
Sure. If we look at our operating expenses outside of the equipment subsidy piece of it, which we've talked to, you'll see if you back out the bad debt piece that I spoke to of roughly CAD 80 million, you'll see year-on-year declines. Some of that decline relates to variable commissions that we saw less of in terms of volumes. But the rest of it does relate to real reductions year-on-year in absolute dollar cost structure. Having said that, as Joe referenced earlier, we are looking at moving forward with some of the efficiencies that we see not only in wireless but across all our businesses. And we think those will drive material year-on-year advantages. And you'll see those picking up in the back half of the year, some in Q3, but primarily into Q4.
And then you have about two million on the unlimited data plans of your 9.4 million postpaid. What's the kind of target there? And are we through the worst of the ARPU pressure just from the transition to unlimited, or should we expect more pressure from the transition to unlimited? In terms of the transition, a couple of things.
If you were to ARPU profile, we've talked about the COVID impacts, and I would call those sort of variable revenue pieces of it. And then, Richard, I think what you're getting at is the underlying sustainable subscription ARPU and what that profile looks like. I talked about us continuing to be impacted by overage in the third quarter, but we are fairly confident we'll continue to see good improvement in underlying subscription ARPU, especially as we head into the second half. So I think we've covered those pieces of it.
And maybe I'll leave it there. Yeah. Just, Richard, when we did the move to unlimited, we did it on the basis of a number of key opportunities. One, to reduce churn. Two, to have a net ARPU positive impact when you look at downgraders versus upgraders. Three, to drive the likelihood to recommend or advocacy from customers. And four, to diminish the calling patterns and the impact on our operations because they call less overall. All four of those items are exactly intact in terms of where we expected them to be around the move to unlimited. On top of that, yes, there's been more pressure on overage. People have stayed home. But the behaviors that they're building inside the house around video conferencing, Zoom calls, online shopping, all these things, we've already seen a trend in the last few weeks of June and July.
Those behaviors have moved outside the house into the wireless and mobile world. So we think it actually plays very well into both the strategic decision to go to unlimited, but also in terms of the value economics of going to unlimited. And that's all completely intact. Where we go north of 2 million, we're going to keep pressing on the point. I mean, it's Rogers only. It's not available on our flanker brand, Fido. So it'll just naturally progress over time as people want to rise up to the $75 price point and as they have a greater appetite for use. But we believe that we are moving through the thickest part of the overage melts through Q3, and then we'll see that really ameliorate in the quarters that follow.
So it's fair to say outside of COVID, after Q3, you're on the other side of this?
It's fair to say that outside of COVID, it was going to happen in the Q3 period. COVID has made that a little more complicated because of the additional pressure on the overage that we talked about. So all things being equal, all things being equal, yes. The question, if your hesitation is, I don't know what's going to happen with COVID in the second half of this year. I'm looking at what's happening in different parts of the U.S. economy with, I said, the seesaw around, "Let's go out. Let's come back in," etc. And I don't know how that's going to play into this dynamic. But if you're going to hold the current conditions constant, absolutely, we see coming out of this somewhere in the Q3, Q4 time period, as we've said before.
Great. Thank you.
Thanks, Richard. Ariel, we have time for two more questions. Certainly. Our next question comes from Batya Levi of UBS. Please go ahead.
Great. Thank you. Two questions. First, on the wireless side, with some pickup in activity now, can you talk about how we should think about churn in the second half of the year? And on CapEx, can you provide more color on future CapEx plans? As in, would the roughly CAD 500 million lower CapEx this year on delayed projects be added to a normal run rate next year? And maybe how you think about the normal run rate intensity for cable and wireless going forward. Thank you.
Hi, Batya. I'll take the churn question, and I'll ask Tony to cover the CapEx question. On the churn front, no question, we've seen a massive improvement in churn this quarter.
Going from 0.99 postpaid churn and wireless to 0.77, I mean, that's not sustainable in an open market as a whole. Yes, we'll continue to improve churn because we've been on that path the last few years. We'll continue to see the right sort of march to better churn over time, which is great. But also bear in mind that gross additions were down, as I said earlier, for us, about 38% or 135,000 year-over-year. So as gross additions come back, we'll see more froth in the marketplace, and therefore churn will get back to that normal improving trajectory that we've seen the last few years. But it's just not sustainable at 0.77 in that range outside of COVID.
Batya, on your question related to CapEx, the CAD 500 million less this year, some of it, much of it, will flow through to future periods.
How much of it ends up being in 2021? Difficult to predict, again, sort of how the whole pandemic plays out. The second part of your question on CapEx maybe is more helpful. We continue to see throughout the period and probably into next year, wireless capital intensity in the 12-14% range. We may see it 13% or below this year, but resuming back up to 14% and maybe even slightly above next year. Again, it's really around how much work we can get done. Those are kind of the ranges. Then on cable, we had a previously stated goal of getting to 20-22% capital intensity by Q4 of 2021. We are tracking ahead of that.
As we push forward some of the investments into next year, that's probably still the right goal to think about for us for the exit rate in 2021.
Got it. Thank you.
Great. Thanks, Batya. And last question, Ariel?
Our final question comes from David McFadgen of Cormark Securities. Please go ahead.
Oh, great. Thanks for squeezing in. Just a couple of questions. So when I look at the 7% impact to the wireless service revenue, you said about 100 was lower roaming, and then I guess the balance was lower activation fees. That should come back as customer activity picks up, right?
That's correct, David.
Okay. And then just on the roaming revenue, can you give us an idea? I don't know if you have this or you're willing to share it, but can you give us an idea how much of that would be business versus personal?
Just trying to understand as the world returns to normal how that could come back.
We don't split that out, David. And in some ways, it's a bit of an arbitrary. It just relates. I think the better barometer that we'll look to is just travel in general. Because whether it's on business accounts or the personal accounts that eventually flow through that expense, it's the total travel or the consumer absorbs it. We look to probably the more relevant one is total business, total travel, I should say, as a better barometer. But we don't see that moving for some time.
Yeah. Okay. And then just lastly, just a clarification maybe, just on the Jays, if the Blue Jays are not playing live games with people in attendance and decent attendance, does that mean that that would keep the EBITDA for media from going positive?
It would just be such a drag. It would make it very difficult for media to be net positive without the games. I mean, it's possible, depending on the amount of advertising revenues, and so we don't want to stretch ourselves in terms of trying to forecast that too far out, but the Jays' loss of revenue, game day revenues, is a material amount for the media business.
Yeah. Okay. All right.
Thank you. Thanks, David. I'll turn it to Joe.
Just a few comments before we go. I think it's important to bear in mind, and this is the way we look at the business. There is a solid business with a strong and resilient base of recurring revenue. I mean, in wireless, 90% of the revenue is in that strong recurring base.
We spent a lot of time today talking about the 10%, the vast majority of which is impacted by COVID and will recover on various timelines, roaming being one and activation fees being another, but it will recover over time, and as I said earlier, we're making our way through the curve on unlimited and the overage melt roughly to the same schedule that we discussed overall, so it's a solid business, strong base of recurring revenue in wireless. The same can be said of cable. The cable there, we're looking at the macro trajectory of the business as we've been driving cash margins, as we've been driving resiliency. We did forgo the price increases I described for all the right reasons.
As we kind of continue driving on the cost efficiency side and get the revenue metrics where they need to be, that business has all the potential we've described before the COVID period. And on the media front, as I said, small piece of our valuation, but at the same time, these are things that will recover given the importance of live sports as a whole. And the last thing I'll say is you can count on us to adjust and adapt the cost structure, the operating model to the new realities. That's the nature of our responsibility. And we're looking at every aspect of the business to understand what the new operating environment might look like in the short term, in the medium term, and the long term, and to pivot appropriately to make those adjustments as necessary.
But we've got strength in view more than ever in terms of the growth prospects of this business and the industry as a whole, given the importance of our services and offerings to both individuals and businesses into the long term. So thank you for your time. Thank you for the questions. And we'll talk to you next time.
This concludes today's conference call. You may disconnect your lines. Thank you for participating and have a pleasant day.