Thanks for the introduction. Hello, everybody, for our today's call on the preliminary results of the year 2020. I think we've all experienced an extraordinary year in many types of things. During the presentation, we will also relate a little bit to some of the learnings and some of the actions we had to take. Overall, for us, most important is that our omni-channel strategy worked out really well. As you can see on page number four of our today's presentation, we were able to grow the subscriber base across all subscriptions with a significant increase of +243,000 net adds. We're very happy about that. That also shows the sustainability and the robustness of our business model and of our fragmented distribution structure. The EBITDA ended close to EUR 426 million, which is very well in the guidance. Ingo Arnold will explain the major drivers.
There are certainly a couple of savings from the COVID crisis, but we believe that the savings that we've had there or the subsidies that we have gathered there will be replaced by a recurring cost-effectiveness in the future. On the cash flow side, there we ended up with EUR 237 million as well within our guided range. What were the key drivers and the key achievements during this past year? We've split it, as always, in three chapters. One is on mobile communications, and I think you've learned all that before. I think if I look back, there are three major things that have been driving our abilities here and our performance.
One is that we have not only launched app-based portfolio, but we have also based on the learnings and based on the system architecture that the colleagues have built up for that portfolio, we were able to optimize a couple of our legacy processes within the company. The effect of these tariff plans is not only the number of subscribers, but also the impact on our digitization of legacy systems. The second one is certainly that we've been able to start a couple of new collaborations, such as Netflix bundles with our product portfolio. I think the success in numbers is a little less than we thought, because when we started it, which was late in 2020, was obvious that in the first lockdown, Netflix has already gathered everybody, all the couch potatoes, into their subscriber list.
Still, I think it is one more innovation and one more good signal that we've been able to join forces with such a global brand. Certainly, the finalization of the LTE migration was a substantial effort. Having seen a couple of ARPU deteriorations the year before based on this migration and transformation. We have seen a more stable ARPU in the [Non-English language] from the Corona effect in the customer base, and we're very happy that this is now finished. On the TV and media side, once again, VOD bundles, a couple of new channels joined. A full set of Turkish channels for the Turkish community. A successful price increase with freenet TV, bearing the fact that we have lost a couple of customers. The goal was to increase gross margin and EBITDA, which worked perfectly well.
We have founded and started our new activities with five owned channels on the DAB+ multiplex. Those have been launched in October, and from April 2021, we will not only run our own five channels, but the full multiplex is now up and running, and the external partners are also paying the technical fees to Media Broadcast. On a group level, I think in general, as I said in my introductory sentence. I'm happy that the omni-channel strategy works out well, and we see also that the transformation that goes along with it within overhead and back office, is doing pretty well. We will engage more into that subject matter during the first six months of 2021, leading to leaner processes, smarter treatment of the customer, and improvement of crossover customer experience. We've also launched a number of initiatives on the key modern topics such as ESG diversity.
We had also learned to adapt the, what I call the new normal, the way everybody like you are yourself now working more from home and finding the right way to do so, not losing efficiency and effectiveness. Keeping things lean and smart. I think there is a lot of small things that we have not anticipated, and the learning curve was extreme, but it was implemented pretty well, and I'm very happy about the entire organization. On the more corporate side, Ingo will go into a couple of details, but certainly, the sale of the Sunrise shares and the follow-up impact on refinancing, et cetera, was a great success. Having spoken about Corona, I hate the word, and I think we're all fed up with it, but still, we have to live with it for the time being. I think I won't go through all the topics.
We obviously had to adopt all the rules. We have implemented it. It's a side information maybe, we had in total 68 cases so far across the 4,000 people. We have had no real damaging case so far, and we know that we have not had any infection coming from within the company, so we're very proud of that, and people are very happy. We had to close the shops. We have also partially closed all shop-related functions, and put them into short-term work. The strategic focus of the company, and I think that goes along with a trend in the entire industry. Certainly, gross adds lost even more in their importance. Renewals customer base development is in an accelerated mode. Even more important, we have seen churn intake or termination intake slowly going down, but continuously going down over the full year.
For your information, none of our employees had to accept any pay cuts. I think that is important. I think that keeps morale at the very high level. Our shop staff is still in the shops. They're still working from the back office, and they're talking to the customers. They do video conferencing. They do all kinds of things. We do click and collect. I think the learning curve in retail was exceptional, and seeing that not only with the mobile shops but also with Gravis, I think that was a real step forward. We think that the sustainability of our retail chain has shown proof of concept during this past year. On the dividend, a quick one. You know that we have suspended it back then for the EGM. I think that was a very wise decision.
We have had then room and freedom to do the refinancing. If we include the extra dividend, which we're going to pay out this year, plus the two share buybacks. We have shown to our shareholders that we keep our promise to make them the owner of 80% of the free cash flow. We have had a lot of positive feedback on how we did it and how we treated it. Not only from institutionals, but also from retail. I think that was the right decision at the right point. We've also shown that we need to adapt to new information and to latest updates. If we're doing really well, we should then keep the promise as we made. Ingo will also explain the new financial policy and guidance in a minute, to share with you what our overall thoughts are.
On page seven, I think it's a very nice illustration of where we've ended up, right in the middle of what we've planned in EBITDA or free cash flow, little lower. Ingo will talk about working capital impact in a minute. Certainly, the second shutdown in December gave us a bit of trouble over the Christmas period. He'll explain it. Page eight, the details on the net adds. As you can see, postpaid as we guided and as we continue to guide, a slight increase. FUNK and FLEX almost doubled, certainly on a reasonably small level. waipu.tv with a strong year and a net add of almost 170,000. I think that is really great. Still, we are waiting for the hockey stick adoption rate in Germany of IPTV.
We know that the legacy and the tradition of the well-learned and well-known remote control for cable and satellite is a hurdle here. I think the numbers speak for themselves that we're getting better and better, and the entire, let's say, segment of IPTV is gaining even more trust within the German community. On freenet TV, we have to accept 10%. I wouldn't even call it churn. I think it's people that have then decided that given the offer or given the quality of the offer, combined with the maybe different pricing, they stopped the service for a period of time. If you increase prices by 25% and lose 10% of the customers, then obviously, I think it was the right decision and the right move, and we will certainly continue to seek for a smart price increase every now and then.
I'm not saying it's going to happen in 2021, but given the learnings from the last year, not only on freenet TV, but also on waipu.tv. We feel comfortable that price elasticity need to be used in order to maintain or improve gross margin. On page number nine, I think that is some of the statistics that we have shown with Q3. If you compare 2019- 2020, it speaks for itself that non-brick-and-mortar channels gained in importance, not only on gross adds with now a share of 55%, but even more on renewals. This equals to a challenge in our shops. Our retail chain must generate even more sales in hardware, accessories and other services, and in all those categories. During the open times, if we compare them on an index, they have done a record year on accessories, on et cetera.
I think that is a very positive message. I think that also, by the way, goes for our partners, the free retailers, but also MediaMarkt and Saturn. Even though, as you can see on the right-hand side, the channels that we have full control of have now a substantial volume of 70% of all transactions, which also means that over the past number of years, we have increased this share. The part of the value chain that remains within our control and within our pocket is constantly increasing. There is, on the next page, even more detailed look at the postpaid development and both new tariff plans. I think they're both doing really well. We have had some improvement on performance with the product itself, but also with the campaigning. These days we focus our campaigns.
We are under the review of how we spend TV money in the future. If we do it more on the mobile, on, let's say, the tariff brands, whereas in the past, we've put the money more on the retail brand. I think that is one of the key strategic decisions that we have to take during the course of the first six, seven months of this year. At this very moment, we do not spend any money on TV. We have pushed that into the second quarter because of the current lockdown situation. Page 11, I think that the curve clearly shows, and we have also explained that already 12 months ago, I consider freenet TV becoming a long-tail business. We have significantly reduced the team there.
We have taken the opportunity to replace them from Cologne to Hamburg, where in Hamburg, all the competence in online and distribution and sales is located, whereas in Cologne, it's the pure TV technology. I think that has given us significant benefits. We are able to use synergies across brands and across products. As I said, we have reduced the headcount in the pure team by about 60%, and the numbers on a EBITDA and gross margin level proved that this was not only the right decision, but also that the small and very flexible team can do much better in these long-tail aspects. On the next page, kind of the last one from my side on the past year, you can see the overall curve on the subscribers of waipu.tv. We are now positive with the monthly EBITDA contribution since May.
We will turn into a positive annual EBITDA in 2021. This is kind of a breakthrough not only for us, but also for the team. We will also most likely buy some of the old shareholders out of the company so that we increase our share, which is, I think, a very logical step, if we turn into positive and want to consolidate as much as possible. Overall, we can clearly see that we are a market leader, and we have extended our market leadership during the course of 2020. Our growth is significantly higher than the one of Zattoo or the pure IPTV MagentaTV of Deutsche Telekom. Not including the legacy EntertainTV, obviously, but the pure IPTV. For me, the 2020 is obviously the real path. What are the things that we're working on in 2021?
Churn reduction and optimizing of the customer experience in the customer base. Certainly one of the goals, the increase of captive channels, the extension of service levels and types of services in the shops, is one of the key targets. We're still in the lockdown. Only 1/3 of our shops is open, not public open, but open for click and collect and repair. All the new technologies such as meeting or scheduling of meetings with our shop staff, video and telephony consultancy from our sales reps with the end consumer, all these things have been implemented and become now operational normal, and are subject to even more optimizations. We are about to launch our first 5G contract. We're not expecting huge uptake, in terms of branding, in terms of proposition and perception, it is a very important step.
My personal view is that if we reopen the shops, that we will see a flat stream in any of our retail. I think people are desperate to go back to the shops. We are preparing for that, not only our staff, but also in terms of availability of SKUs and product in the shops. On the TV and media side, freenet TV obviously going into a long tail, which means renewals, no significant innovations, no significant investments into new customer acquisitions. In parallel, as I've said, we have started the five radio channels. They are now not only available on DAB+, but also on podcast and on streamings. We have, with our partner together, even founded a media agent to sell the inventory.
We expect this to not contribute significant profits and revenues during this year, but a positive contribution moving into a significant business in the course of the next couple of years. On waipu.tv, I've already mentioned that we most likely will extend our shareholding to take more of the benefits and the profits into freenet and also making it available for our shareholders. Having said that, I'd like to hand over to Ingo for a more detailed look at the financials of the past year.
Thank you, Christoph. Good morning, everybody, from my side. I would like to start on page 14 with the financials. I'm very happy with what we reached here during 2020, and even with the fourth quarter. I read in the comments this morning that some of you were disappointed by the figures. I was a little bit surprised by this. I try to clarify the things and hopefully get some color into it. On the revenue side, in the fourth quarter, we reached EUR 670 million. Yes, definitely this is lower than what we reached in the fourth quarter of 2019, but this is based on the shutdown here, what we saw since the 16th of December.
For our Gravis stores, the period from 16th of December to the end of the year or even to the beginning of January, this is the most relevant period in terms of revenues. It is not so relevant for our groups in terms of profit, but in terms of revenue, it's very relevant. Therefore, yes, we lost some revenues in the fourth quarter here, but without any big effect on the profit side. All in, what we see is that it is a relatively stable revenue. On the gross profit side, if we look into the adjusted figures here without the regular topics, yes, it was down from EUR 897 million-EUR 882 million.
I think you have to put into consideration here that in the fourth quarter, yes, maybe on the first view it looks a little bit disappointing, because if you compare 2019 with 2020, you see a loss of something like EUR 7 million here. EUR 5 million out of this is freenet Digital. This is the company what we sold at the end of September. The biggest effect is resulting from a business which was not core before, but all in which was not generating any EBITDA. On the EBITDA side, we all in-generated an adjusted EBITDA of nearly EUR 446 million, without the differences from Motion TM and from the international call effect. If you look into this adjusted figure, then you see that we increased EBITDA on a year-to-year basis by something like EUR 9 million. Moving to page 15, to the mobile performance.
Yes, on the revenue side, you see what I already mentioned, that the close of the Gravis stores in the mid of December had a negative effect here. This is what you can see in the fourth quarter. All in, if you see the whole year figure, this is nearly the only effect what you have, the difference from the fourth quarter. In the gross profit, I think this is very important to mention here because it is for a long period the first time, and therefore, I'm very happy about this. It was possible for us to increase the gross profit on a quarterly basis between the fourth quarter of 2019 and the fourth quarter of 2020. It was possible to increase it from EUR 174.8 million- EUR 175.4 million.
I think this is the best proof of concept what we could give, that it is really a very resilient business what we have here in mobile. On an EBITDA side, yes, correct, not all of these positive effects from gross profit are transmitted into EBITDA. There is one reason for this. What we did here was, we built some provisions for bad debt. You could call it conservative, but what we do here is we do, on a regular basis, we do a lot of analytics about the macroeconomic trends, what we do expect for the future. If we look into it at the moment, we see some signs here that the economical situation out here could go worse during 2021 after the pandemic based on unemployment rates and so on. Therefore, I read it in one comment this morning.
Yes, maybe it is perfect to call it preventative. This is something what we did here. I think during the year, we will see if it was really necessary to build this provision or if there will be a chance to release it again. I would not do so today because we do not know all the effects of COVID-19, but I think we will see during the year. Moving to the KPIs of the postpaid business on page 16. The quick intake of postpaid customers make me very optimistic for the future, because even with a quarter where we lost nearly two weeks at the end, it was possible for us to increase the number of postpaid customers by 74,000. It was a very strong quarter, without the lockdown, it would be even better. Very good intake on this side.
A sign that the mobile business is worth a lot, that it is resilient, and that it is very strong. On the ARPU side, yes, this is something which looks a little bit negative because the ARPU went down in the fourth quarter by EUR 0.50. Still you have some roaming impacts here, as business is not running as it was before. Without roaming, it is nearly stable what we see here in the ARPU. On the Digital Lifestyle side, we see a decrease in the fourth quarter. In the whole year, even with all this pandemic effects and with the closures of shops what we saw, we see that the revenue out of the Digital Lifestyle is nearly stable with EUR 189 million.
The reason for this, and it's also a positive message from my side, what we reached during the last years was that we moved the revenues from a reselling base to a subscription base. Much more of our Digital Lifestyle revenues today is in subscription models, and therefore, it was getting more and more sustainable by the time. On page 17, I move to the TV and Media segment here. What we see in the revenue here is, yes, it looks very stable. We see increases from the growing subscriber base from waipu.tv. It's a very stable revenue situation. On the gross profit side, here again, you see if you compare the quarters, what you see here in the gross profit. If you look into 2019, there was in the second quarter something like EUR 42 million, in the third quarter, EUR 42 million.
In Q4 2019, there was an extraordinary effect of EUR 5 million. The gross profit in Q4 2019 was EUR 47 million. Without these extraordinary, it would only have been EUR 42 million. Average level in Q2- Q4 2019 was something like EUR 42 million. What do we see in 2020? We see a level of EUR 45 million. The figure of EUR 44.6 miilion in the fourth quarter, in my eyes, it's not disappointing, but it's a confirmation of the gross profit, what we saw in the last quarter. Moving to the EBITDA, here again, we saw a EBITDA of EUR 86 million in 2020, which is nearly EUR 6 million higher than in 2019. All in, it is something 20% of the EBITDA of the whole group. It is a very relevant part of our business now. Some details on page 18.
I only would focus on the middle boxes. There is the Media Broadcast B2C business, which is freenet TV. Here you see there was in the delta to last year was an increase of EUR 1.4 million in the EBITDA. On the other side, if you look into the Media Broadcast B2B business, here you see the effect, what I was already talking about in 2019. In 2019, we had a very positive effect of EUR 5 million, which were shown that time in the B2B business, and therefore now it looks much weaker than in the other quarters. But it's only this extraordinary effect. In the Exaring business, we see that, and Christoph was already talking about the positive monthly EBITDA figures. All in the EBITDA of 2020 is EUR 7 million higher than the EBITDA of 2019, and it will be definitely positive all in in 2021.
Moving to page 19, maybe here, on the first view, another disappointing figure. Also here, I think this needs some clarification because on an adjusted level or on a normalized level, I would say the free cash flow was EUR 249 million. With the closure of the Gravis stores, there was a negative phasing effect from working capital. This negative phasing effect in the last weeks of the year, brought us this decrease by EUR 12 million- EUR 237 million. There will be a positive effect in 2021, definitely out of this, because it is only a phasing. Therefore, I think on the first view negative. If you look into it, we are much higher in the range of the free cash flow, and we are also here very successful during 2020. Another success is shown on page 20.
I do not want to discuss all the figures. I think the most important one is below the tables. On the one hand, at the end of 2019, we had a net debt of EUR 1.55 billion, now we do have a net debt of EUR 269 million, bank net debt. We optimized our balance sheet. It was very important to optimize it, now we have a very healthy situation. This is something what you can see on page 21 in a result, because what we see here is that now we have an equity ratio of 40.4%. We have a leverage of 1.7x. It's a very healthy structure and totally different to the figures, what we showed at the end of 2019. Moving to page 22. Christoph was already talking about the suspended dividend.
I agree totally that it was a correct decision, what we took last year on the information, what we had. It made a lot of things on the capital markets possible for us to refinance, and it was a one-time suspension. This is what we already announced some weeks ago. This year, we will pay again a very high dividend with a good dividend yield, what we grant. It will be perfect, but it is important what we show on page 22, that for the shareholders, the payments, what we promised to give to the shareholders, they really happened in a different way, but we stand to what we promised. On page 23, we show the new guidance. First of all, the subscriber guidance in postpaid. We still see the possibility to increase the customer base further with a moderate increase.
On the freenet TV RGU. We expect a moderate decrease, a further decrease. What is important for us here is we report this RGU, but our priority and our focus is the gross profit of this business, and this is what we did with the price increase in 2020. Yes, if you only look into the RGU, maybe this looks disappointing, but if you look into the profitability of the business, this could be increased dramatically, and therefore I think this is also the idea for the future. We do not know when it will be possible to do another price increase, but we definitely will focus on the profitability of the business. It's only a second priority to focus on the RGU. waipu.tv, here we expect a solid further growth. It's growing month by month, quarter by quarter.
There is no sign that something could change here in 2021. The financial guidance in revenue, we expect a stable revenue here. In the EBITDA, we expect an EBITDA between EUR 415 million and EUR 435 million. Why this range? We did, as everybody do at the moment, we also did some worst case scenario calculations on COVID-19. We calculated how long the shops could be closed and so on. In all the worst case scenarios, what we calculated at the moment, we did not end below EUR 450 million. Therefore, EUR 450 million is the lower end of the guidance range here for the EBITDA. If you ask me today, yes, I would expect to end up in the higher end of the guidance, but we are at the beginning of the year now, and I think we have to wait what would happen during the year.
Something similar with the free cash flow. It is lower than in 2020, and it has to be lower because we will not receive the dividend from Sunrise, which was EUR 46 million in 2020. We also paid some more interest. All in, there will be something like EUR 35 million what we will lose at the end of the day without the Sunrise stake, and therefore now we have the range here between EUR 200 million-EUR 220 million. With the phasing effect, what I explained before from Gravis, I would also say here what I do expect today with all these uncertainties, because now we are only in February, but with the uncertainties what we have at the moment, I would say yes, I do expect to reach the upper end of this guidance. This is what I would say today.
Moving to page 24 for a detailed free cash flow bridge. What we see here is that on the net working capital, we see something like -EUR 25 million here, what we do expect for 2021. In 2020, it was something like -EUR 35 million. Here we see a lower value also driven by the phasing effect, what I explained before. In the tax payments, EUR 40 million. There are still some postponement of payments here. If the authorities ask us to pay or request the payment, EUR 40 million could be possible. If not, could also be possible only to have EUR 30 million, but in a normal way of doing, I would expect EUR 40 million. CapEx, EUR 45 million. We still do have some investments into digital radio. It's something similar to what we did in 2020.
Christoph already described that in the first half of the year, the second multiplex will work 100%. There is still some work to do and some investment to do, and therefore, we will see these investments during 2021. On the leasing side, here we see -EUR 70 million. It is similar to what we saw in 2020. It was slightly higher than in 2019, because there were the investments into digital radio, and there were some new rent necessities out of this. The interest payments, EUR 35 million, EUR 10 million lower than in 2020 because of the reduction of the bank debt.
The quarterly breakdown on the right-hand side, what I would say today is, I think in 2020, if you look into it, the third quarter was something like too good, because there were some phasing effects because we got some money from the network at the end of the third quarter, which I already mentioned in the last call. It was something that was too good, therefore the fourth quarter was too low. What I do expect for the next year, without any big phasing effect, I would expect something between EUR 45 million and EUR 60 million per quarter. Page 25. Our financial policy for the next years. On the leverage side, yes, our leverage now is below two, but we would like to have some room for maneuver, so we do the share buyback during 2021.
This could lead to something like 2.0x or 1.9x during the year in the leverage. Therefore, we thought we do need some room and a leverage of three in the industry is something which is really normal. Today, I would not expect us to see something above 2.0x during 2021, but we would like to have the room for maneuver. Today I do not see anything which could increase it above 2.0x. In the equity ratio, we left it on a level of 25% here. Today, there is a lot of headroom because actually we do have an equity ratio of 40.4%. What we do expect during the next month is even to increase it. On the dividend policy, we stick to our policy. It is linked to the operational performance, which is reasonable for us, which makes a lot of sense in our view.
Therefore, we left it on a level of 80% of the free cash flow. Also in the future, we will be a company which will deliver and grant a very high dividend yield. I think this will be still part of the story to invest into our share. Saying this, I would like to hand over back to the Operator and ask you to start the Q&A.
Thank you. We begin our question and answer session now. We already got a few questions coming in. The first one coming from Christian Fangmann from HSBC. Your line is now open.
Yeah, thank you. Good morning. I have three questions. The first one is on your statement regarding Exaring and that you may extend your shareholding. Just trying to understand what magnitude, how much would you be buying, or what is up for sale? What could it cost in terms of investment of cash outflow for that? The second question is on the mobile business. You had a very strong net add performance, but the ARPU was kind of weak. Can you maybe explain, first of all, how you generated this big amount of subscribers, and then are these really low ARPU customers coming in? Trying to understand the momentum here. Lastly, on the TV side, the Q4 EBITDA was down year-over-year. You mentioned a bit the one-off effect in the previous year quarter, waipu.tv is growing, is now positive.
freenet TV should at least generate a little bit of incremental EBITDA year-over-year. Why are we not seeing a bigger uptake in EBITDA year-over-year? Is maybe the legacy B2B, the Media Broadcast business, the issue here? Maybe you can explain a bit more the moving pieces within the TV EBITDA segment. Thank you.
Yeah, thank you for your questions. First one on Exaring. There is still, as you know, more than a third of the shares with founding members. There is one of the, let's say, maybe not founders, but early participants that let us know that this institution would be happy to sell their shares because they have some other investment opportunities. It's not a family office, but it's kind of a family investment, or has been a family investment. They have told us already a year or two ago that they would be ready to sell a couple of their shares. We have now, only last week, agreed that we will have a deeper look into it.
I think the total will be a maximum investment of EUR 10 million, and the share price that we would buy to them is below the one then that we have invested. It's taking an opportunity because one of the original founding shareholders wants to step out. I think it's a smart move because we're getting them rather cheap based on a specific situation in the environment. I said, it's about EUR 10 million. Not yet agreed finally, but highly likely. We have not agreed yet on the payment terms. I'm not sure when it will hit the cash flow. It will be, I say, somewhere around 10% of the company. We will increase close to 70%, which I think is, at this point of time, a very intelligent investment. On the ARPU side, well, I could not remember that question for many years.
I think it's a very good question. What we see there is the entire way we do customer acquisition is not based on ARPU, but on customer lifetime value. That leads every now and then to things that look a bit weird. Yes, the ARPU overall, the ARPU of new customers only recently was a bit lower than the customer base overall. The customer lifetime value that we have per gross add last year was better than the year before. I'm not talking about euros, but like cents and over lifetime, we talk about euros. What I'm trying to let you know that we are not, when we think about is there an opportunity, how we would fit it, what is a good campaign, which campaign do we extend because we're successful? It's not so much the ARPU or it's not at all the ARPU, it's the customer lifetime value.
We have two curves that not always correlate into the right direction or not go in parallel, it's ARPU and margin. I can assure you that the margin of those customers is even better than the one on the customer base. We have accepted it's a lower ARPU effect. To give you a flavor of it, the share of SIM-only that we have sold in the second half of the year in our own shops was increasing. SIM-only tariff plans typically have a lower ARPU, but the equal absolute, and obviously the relative better margin than the subsidized contracts that include hardware. That is the driving force. Also on SIM-onlys, the entire credit scoring, et cetera, is much more relaxed because you do not hand out a hardware to the individual customers.
The potential damage if one of those customers goes into bad debt is much lower than with the subsidized hardware-based contract. That is basically the driving force into that. For sure, in a year or in a phase where we have a higher online share, ARPUs tend to be a bit lower because of that specific fact that the proportion of SIM-only is higher.
Hello, Christian. Concerning your question about the profitability of the media business, I already tried it before. If you look into the quarterly figures, then you see in 2019, from Q2- Q4, there was something like an average of EUR 42 million without the extraordinary effect. If you look into Q2 2020- Q4 2020, you see something like EUR 45 million. It is something like EUR 3 million a quarter. It would be EUR 12 million a year, and it's something like an increase of 8%. You could do the same math also on the level of EBITDA. I think the only thing what is disturbing at the moment is the one-off effect from Q4 2019. This is disturbing the view. If you leave this out, then you definitely see an increase, and there will be a further increase in 2021.
Okay. Thank you. Maybe just one follow-up on Christoph's points. Obviously, we're still in a lockdown scenario right now in Germany. Can you say a bit on kind of the momentum you are seeing during the first part of Q1?
Yeah. No surprise. It remains with a similar structure. We have, obviously, the online and direct channel is the one and only right now. The SIM-only share is even higher than it was in Q4. That's the driving force. Once again, the good piece is that we save a bit on fixed costs in the shops. The good thing is that we have lower hardware sales and subsidies. The life cycle results of the SIM-only equal that equation out. The damage is coming from lower attach rates and up-sellings, and also from lower digital lifestyle activities and accessories. I think that is it. I would say right now from a bottom line, I think the negative effects that I've just described and the positive ones from cost savings equal out. We are not expecting a real opening before mid or even end of March.
I think the first quarter we will show some wounds, but As Ingo said, all our simulations show that overall, we will, even under a very negative assumption, overall, we would still remain within the guidance that Ingo has just explained.
Okay. Thank you for the update.
Thanks, Christian.
The next question is coming from Jonas Blum from Warburg Research. The line is now open.
Yeah. Good morning. Thanks for taking questions. I got three, please. Firstly, just following up on Christian's with regards to your guidance for ARPU. I mean, your guidance for flat ARPU. We shouldn't expect roaming to come back in 2021, or is there some other parts included in this guidance? That's first. Second, around a comment that was mentioned yesterday in Telefónica Deutschland's conference call, since they expect Drillisch to migrate data traffic away from their network, and consequently, try to leverage also partner business. Is that something you also expect to benefit from, significantly, perhaps? Are you more looking into offering a balanced portfolio in terms of networks for your customers going forward? Finally, just on your waipu.tv growth trajectory. Just wondering, Mr. Vilanek, you once mentioned that you were eyeing for five million customers by 2023.
I guess this is not up to date at the moment, but what do you think is a fair market share for a product in the medium term? And what's a fair assumption for the customer base by 2023? Thanks a lot.
Yeah. Jonas, thank you. I think on the ARPU, we expect flat. I think there will be some roaming coming backward, but we do not expect it to return to the original level of 2019 yet. I think that the German news in the morning was that most likely they're going to allow international traveling in summer, but nobody's booking yet. I think we did not incorporate any of this potential upside, and we see how it's going to happen. That is assuming a slight uptake, but not a return to the old status. On the Telefónica statement. At this point of time, 1&1 and Drillisch do not run their own networks yet. We will certainly benefit from a competitive environment with four networks.
Be it either that Drillisch will give us very favorable conditions in order to enlarge their market share, or the other three to defend their market share. I think, under any assumption, the whole development is a positive one to us. We have also done analysis on what we have learned about the national roaming agreement. So far, we still believe that our conditions on data, with Telefónica, remain competitive and will allow us even in the future to compete any offer of Drillisch. In that sense, I think from a competitive mechanics, we expect a benefit, and we do not foresee somebody being in a position to undercut our pricing significantly on the network. I think the third one, I can't remember whether I said 2023. I think I said 2023- 2025 to be on the safe side.
You're still putting the finger to the right point. I said that in my introduction, in my statements on 2020. I think we're really happy with the growth rate in waipu.tv. If we compare it to Zattoo. If you compare it to others, we have seen TV Spielfilm going away. We have seen Magine going away. We see that the high investments and, with full respect, the great progress that Deutsche Telekom does with their own IPTV. If we do a one-on-one comparison, we still grow much faster. I think our market share on the IP should remain at a high level, and it's a matter of definition, but I think I always said 20% of the market should be the right number. I was assuming that whether it's 2023 or 2025, approximately 30 million will, in a way, consume IPTV.
If we assume that there is the majority of them paying in any of these methods, it is important for us that we include VOD services and we bundle VOD services with waipu.tv, in order to make the people also take our linear offering and then pay for it. I think it's got to be slower than today. I think there's maybe one part of my statement was, how would I say that, not precise enough. I don't know whether all these 30% that move into IPTV, whether they will really cord-cut, and will use it exclusively or is it an add-on. I think that is the big uncertainty.
When we do customer research, we realize that those people, they switch off. It takes a real while for people say, well, now I cancel my cable TV, or I take away my satellite dish. This is still uncommon. In that sense, the uptake is slower. If you ask me today, I think to have more than one million subscribers is definitely reasonable for going by the end of 2022, early 2023 for us. The five million is not going to happen in 2023. That's for sure. I think there the market is just slower. The ultimate question remains to be answered, whether there is a turning point where suddenly people fully understand the convenience. Suddenly the uptake might go super steep. I may be, in terms of timing, less optimistic today than I was before. I'm happy that we are highly profitable.
If I assume one million subscribers, whenever we know that this is a very significant contributor to our total EBITDA in a two-digit million number. I'm happy about that, but we have to accept that this is going less aggressive than we thought.
It's very helpful. Thanks.
The next question is coming from [Yemi Falana]. Please go ahead, Sir.
Thanks for taking my question. Just a couple from me. Firstly, on cost control, it seems like you took 8% of the fixed costs out of the business this year. Is that something that you think you can continue to do? Do you expect to still be able to cut costs in this kind of way and of this magnitude, or do you think some of those costs will come back as the COVID recovery begins? Secondly, just focusing in on ARPU again, could you potentially elaborate on some of the specific headwinds that you see, given the fact that you're guiding for stable? I presume that there are some headwinds that you see offsetting the potential for a COVID recovery and roaming recovery in the back end of 2021.
Just kind of framing that with the fact that Telefónica have also said that they expect a stable pricing outlook in both the value and high-end segments. Finally, if I could just squeeze another one in. Your freenet FUNK product is starting to show real customer traction. Is it still the case that that's a similar ARPU level to your postpaid mobile base? Thanks very much.
Yes. Thanks for your questions. Maybe I start with the cost question. In a way, it is still not 100% clear if all or what part of the costs which were reduced in 2020 will be recurring in 2021. Therefore, definitely part of the cost savings will be recurring. There are also parts like the money what we get from the government because the time of work was reduced. This is definitely not recurring, and this was EUR 3.5 million last year. Now we have, again, a quarter where we will get this money, so therefore in 2021 it will be recurring. If you had asked me three months ago, I would have said not.
Yes, make the long story short, I think biggest part of the cost savings will be recurring, but there will be some costs which will come back. For example, marketing will get higher again because we will invest again. We had a lot of learnings, and then there is one part of the cost, which is not in our hands, and this is why we built the provision of something like EUR 6 million at the end of last year. We do not know how the payment behavior of the customers will develop. It was very fine during the crisis, but we do not know what happens after the crisis, when maybe the unemployment rate, for example, will be higher. All in, there will be recurring parts, but also parts will be come back.
What I would say is from best guess from today is something like 50%-75% of the cost savings will be recurring.
Yeah. I think the other one, on the ARPU, I think my answer is the same as before. We have to see the acquisition mix. You said the average on FUNK is lower than the average of the base, but the purchasing model behind it allows us to create a very attractive margin. Having said that, you might also ask the question, why don't you do more FUNK? Well, the FUNK thing is not pushing the shops because it's app only. Whenever we decide on where we put the money on, we have a, I would call it, a sophisticated model how to optimize.
It was clear to us after the lockdowns in the first/second quarter last year, that we need to push retail a bit and give it more potential to win customers and do the renewals, obviously, because you could consider our brick-and-mortar business a fixed cost business. Logically, we have said, well, don't do a lot on freenet FUNK these days because you can win on the others. Now vice versa, if the shops are definitely locked down and we have no possibility, we move into pure online, and pure online is then freenet FUNK/FLEX and SIM-only. We balance this out, and that is part of the job that we do internally with our category management and our sales planning. It's a multidimensional optimization with a lot of drivers. Maybe once again, I can only do an illustration and not give you all the numbers.
Let's assume, or we have had that situation in Q4, we were saying, okay, in order to reach some of our volume bonuses with the network operators, we're still missing, and just for the sake of argument, we're missing 100 units of new customers. We sit down and say, which is the right and the best channel to generate those within the next four to six weeks? You have one dimension, which is the life cycle. We have one dimension which might be any kind of fixed cost or variable bonuses with our channels, and the third one is ARPU, and the fourth one is the type of contract. The fifth one might be even investment into marketing. At the same time, we'll talk to hardware manufacturers that say, well, we have an overstock here and there. We would subsidize the handset.
We would give you an extra marketing fund. They're really jogging with five or six variables all the time. This is when we later on, Ingo and myself, need to answer questions on what has happened, then we certainly remember the decisions we've made and also the reasoning, it is something which we do on a monthly basis. The side effect of this, if this was not the case, I guess we could run the company with 40 people instead of 1,500. Yeah. I hope I can at least give you a flavor of what really the sophistication is than to say, okay, we can do it with MSH, with Media Zattoo. That's good because they will be happy, the short-term margin on Media Zattoo is lower, the long-term is good, that we have another partner.
Our captive channels, we cover some part of the fixed cost and so on and so forth. It's kind of a consultant answer. It all depends.
Thank you. Much appreciated.
The next question is coming from Titus Krahn from Barclays. Please go ahead, Sir.
Good morning, everyone, thank you for taking my question. Just two topics, please. The first one would be on waipu.tv, could you maybe elaborate a bit more on what have been the main drivers of the strong net adds in Q4? To what extent have customers taken up the new tariffs, including Netflix? Just trying to find out what the subscriber mix has been over the last year and the last quarter, what impact do you think that should have on ARPUs for the segment? The second question, just a very quick one on the free cash flows. Given that the stores remain closed so far in Q1, what impact do you expect on working cash flows after we have seen the impact in Q4? Is the reversal probably likely to happen at a later point in the year?
Yeah, Titus, thank you for the question. I love the first one because now I'm opening Pandora's box. The driving force was the price increase. The price increase that we have announced for the big product, for the Perfect product, was becoming real in January. We knew that any customers that would hook up in November or December would basically be automatically part of the price increase. The marketing team created a campaign, which included a significant number of bonus months, which was refinanced by the price increase, which was not in place at the time of the customer seeing the offer. That campaign was the most successful that we've ever had before. We could offer the customers a price reduction over a period of six months, which paid back because implicitly, they have already accepted the price increase.
Basically, it was a very smart idea on a campaign, and that was done with and without Netflix and became even more effective. This, let's say, concept and mechanics was discovered anywhere early November. We've tested it. Early indicators showed that also the customer, the attrition patterns of the loyalty remains on the same level as the normal, and this is what they've pushed. Smart campaigning is the answer.
Concerning your question about the free cash flow, I think the situation in the first quarter of 2021 is little bit different than in the fourth quarter of 2020, because we had to order all the hardware and all the stuff already in November last year, where it was not clear if there would be a lockdown at the end of the year or not. It was not a surprise, but we had to prepare ourselves for the business in November, therefore we already ordered the stuff. Now we have a much longer period in the lockdown, we optimized our ordering and so on. What I would expect that a counter effect of what we saw in the fourth quarter of 2020 should be possible in the first quarter of 2021. This is what I do expect at the moment.
Thanks so much. That's very helpful.
The next question is coming from Francesca Schild from BNP Paribas. Please go ahead.
Hello, good morning. Thank you for taking my question. It's just one for me, please. Regarding what you were thinking about employees being made redundant maybe at some point. Sorry, not employees. Sorry, I'm getting confused here. Not related to the company, but to do with people in the wider sector. What was the amount of bad debt provision which you took in 4Q 2020? What was the rationale behind this in terms of perhaps people in the economy losing jobs? Have you seen any change in customer behavior or invoicing which suggests you might not be able to collect the cash? Is this simply based on a high-level macro view? Will you need to take any more provisions next year as well? Sorry for the jumbled question. I hope you understood that.
Thanks a lot for the question. I think first of all, I would like to start in 2020 because what you normally would expect is if you have a crisis, you would expect that the payment behavior would get worse. This was the first very positive surprise what we did have during 2020, because the payment behavior was as good as it has never been before. Now it is the question, what will happen after the crisis?
We are currently having technical problems with the speaker line. We are going to a quick pause and answer your question after that. Thank you for holding.
Okay, sorry. Now we are back. What I heard, it was a small technical problem. I do not know what you get of my answer, the payment behavior in 2020 was very fine, surprisingly fine. In 2021, I think we have to see what happens after the crisis if the unemployment rate will be reduced. I think there are a lot of patterns from the past. We use these patterns to calculate or to pre-calculate what could happen. On these pre-calculations, we build this provision, what we have done at the end of last year. We think that from today's point of view and what we do expect is that this provision will be high enough even if the payment behavior would be worse. If it would not be worse, which could also happen, then there could even be something like a reserve in our results.
Thank you. Very clear.
Okay.
Looks like we have no further question available. For closing remarks, I'll give back to the speakers.
Thanks, everybody, for joining today's session. We look forward to the next one early May in Q1. As always, very happy to take even more questions and more detailed discussions with our investor relations team. Have a good day. Stay healthy. Goodbye.
Goodbye.