© 2024 Alpha Spread Limited. All Rights Reserved.
www.alphaspread.com is not operated by a broker, a dealer, or a registered investment adviser. Under no circumstances does any information posted on www.alphaspread.com represent a recommendation to buy or sell a security. The information on this site, and in its related newsletters, is not intended to be, nor does it constitute, investment advice or recommendations. The information on this site is in no way guaranteed for completeness, accuracy or in any other way. In no event shall Alpha Spread Limited be liable to any member, guest or third party for any damages of any kind arising out of the use of any content or other material published or available on www.alphaspread.com, or relating to the use of, or inability to use, www.alphaspread.com or any content, including, without limitation, any investment losses, lost profits, lost opportunity, special, incidental, indirect, consequential or punitive damages. Past performance is a poor indicator of future performance.
You don't have any saved screeners yet
Intrinsic Value is all-important and is the only logical way to evaluate the relative attractiveness of investments and businesses.
Good morning, ladies and gentlemen, and welcome to the freenet AG Analyst and Investor Conference Call, Q1 2018 Results. [Operator Instructions]Let me now turn the floor over to your host, Mr. Christoph Vilanek.
Thank you very much for this opening, hello to everybody, good morning and thanks for joining today's update on Q1 2018.I'm jumping straight into the presentation on Page 3 with the summary of our key performance indicators. The revenue on an annual basis and we will always split information with and without IFRS, but on a year-on-year basis, the revenue was growing by 4.3%, mainly through a bit more of hardware. The EBITDA is up 5.6%, and if you take that trajectory, then we would end up on the upper end of our guidance that we have published a few weeks ago. Gross profit overall stable underlining the strength of the core business in mobile, I think, as well as TV and media, I do not agree on some of the comments that I have read this morning. We have still growth in digital lifestyle revenues, plus 17% year-on-year. Going a bit more exclusive to mobile communications, plus 1% or 60,000 postpaid customers on stable ARPU, also slightly growing ARPUs on prepaid and no-frills which is certainly also result of the downward trend in customer base of prepaid. Both agreements with -- annual agreements with Vodafone and T-Mobile have been signed, which is very much in our favor, some other years, it took way longer than this year.On TV and Media, we have 5% adds on freenet TV plus 33,000 -- 31,000 in waipu.tv, and overall registration, waipu.tv is expanding across the 600,000 users number and we have a number of changes, which I will relate to later on this specific page. On the details on the mobile customer base, you will immediately see that postpaid, as already mentioned, was growing about 60,000 overall -- overall the customer structure improved, postpaid plus 3.6% year-on-year. We have a slight decrease in the no frills customer base, as I already mentioned. That is mainly a result of the downturn starting in Q3 2017 with the new ID regulations. The customer ownership as such is stable and been a solid sustainable basis of our business.On Page 5, here is the details on the individual ARPU sections per segment. Postpaid very, very stable, EUR 21.4, above the 2017 number, bit below 2016s, I think stable is really the perfect description of that development. No frills going up, same as prepaid, as I said certainty determined by a clean-out of inactive customers and no inactive joins anymore.Digital lifestyle, number for Q1 2018 total revenues EUR 42.6 million, which is a significant growth from previous year, more than EUR 6.3 million and overall the traction remains stable. It is the -- the growth comes certainly from the number of subscribers, but also of the value per subscription. We are about to launch our freenet video, which we will replace the acquisition of Stromberg customers in -- which also will lead to increase in the revenue base because here we do the full charging ourselves, the launch of the product is couple of days ahead of us.On freenet TV, our DVB-T or terrestrial TV -- a total number of paying users by the end of first quarter on the level of [ 1,022,000 ] customers. This development certainly proves that either it's still a selling of STBs and C+ modules, but also the fact that existing customers finally decided to watch private channels. So there is a constant income or increase of the customer base, which I guess some of you and some of us have not thought that would happen after the turnover from the old system to the new systems. That is still going on. It also proves that there is in the GFK reports where we relate to on the right-hand side, we have seen about 60,000 units sold during the first quarter, the hardware sales and those will lead on a, let's say, 50% conversion rate will lead to another 30,000 customers at least in the second quarter of this year. There is one important element I'd like to mention. These hardware sales are very, very comparably low to the previous year obviously and the income from the license fee and from the hardware was already -- and it already happened last year. So this is one of the explanations why EBITDA on TV is a bit lower. The hardware licenses were accounted already in -- during the year of 2017. Our Sat product was launched by the end -- early in Q2, will only softly come in, because we did not replace the old C+ modules that was still out on the shelf space and this will only happen now continuously with the sell-out of the old ones. So most likely we will only see Sat numbers during Q3. But still we do believe in the 1.2 million subscriber numbers by the end of 2018, I think the [ past ] is the right one with the [ adding ] that we have seen during the first quarter. On waipu.tv, we also see an increase, which is totally in -- within our projections. We have by the end of the first quarter the paying users amounted 133,000 and the registered total users crossed 600,000 user number. We are definitely short at -- we are now ahead of any other competitor in that field. We have done a lot of testing another round. So we are, for example, so far all customers were acquired with a negative option. Now, we have changed into a positive option end of March and we'll see the conversion during this quarter and certainly we saw that the intake works better, because we set the final from opening an app down to the real registration is better in a positive option. We have also HD private channels added mainly from Pro7Sat1s and we are testing also the uptake recharge EUR 1 extra on the smaller packages. We are right now testing a lot of peak flows in order to figure out what's the best way of upselling is. We have launched silently again a beta version of a browser -- version of waipu, which means that you can use waipu also on any desktop or laptop. We can have our new EPG with a much improved recommendation engine and we are adding more or less every other week new VoD channels with a very interesting result, interesting meaning that we can see some of them, we will create big uptake. We do not charge those extra channels right now. We are more in the mode of figuring out how we target individual customer groups, how this works, how conversion works and obviously we have added a couple of -- more than 10 new linear channels during this quarter. So all in all, I would say it was the typical kind of Q1. The [ executives ] have continued with the normal business, did the homework and we are well on track if we project the performance from the first quarter to the full year. And having said that, I'd like to hand over to Joachim for the financial overview.
Thank you, Christoph. Good morning, ladies and gentlemen, welcome also from my side. I'll be talk you shortly through the financial statements. I think we delivered a solid set of financials for Q1 starting with the revenues which grow to EUR 873.7 million without IFRS15 impact. This is mainly attributable, as Christoph already mentioned, to higher hardware sales. Nevertheless, of course, we have to report EUR 689.6 million according to IFRS 15, which is mainly due to the reclassification of commissions. So EBITDA amounts to EUR 106.5 million where Sunrise participates with EUR 9.9 million. Nevertheless, EBITDA is more than EUR 5 million higher compared to Q1 2017. Depreciation and amortization is in line with our expectations. Interest results came down to EUR 10.9 million because of the favorable refinancing we did in 2017. Taxes on income increased from EUR 6.7 million to EUR 7.5 million and [ disposition ] includes all the time this deferred tax, income items. However, the cash-relevant tax expenses increased from EUR 5.3 million in Q1 17 to EUR 9.2 million in Q1 18, mainly driven by the higher taxable income, and as a very solid group result, we show EUR 46.7 million. Let's all [ take note ] the next page of segments, mobile and TV again just to make here the key messages. Mobile communications, of course, revenue stayed where it is because of IFRS 15, but I think much more important is really the gross profit, which is a very stable one, as well as the EBITDA, which really demonstrates what we have predicted before that we will deliver very solid mobile business. TV and Media, the revenue came down, main impact for that is really that we had this license fees charged for the sold hardware in 2017. Now, this piece is missing, which is reflected then in the revenues, which is also reflecting a little bit on the level of gross profit, that's why we showed there an amount of EUR 37.4 million. Nevertheless, EBITDA is growing -- now growing because of course we have media broadcast with less marketing expenses and EBITDA participation of media broadcast amounts to EUR 14 million and we have an EXARING with minus EUR 6 million EBITDA because there we spent more marketing, we have done a lot of TV campaigns in the first quarter. Obviously, the impact of the increase of subscriber numbers we will see then in the second and third quarter. So I think we really fit that exactly with our subscriber numbers what we have anticipated so far. This brings me then to the balance sheet. There are no major changes in the area of ordinary business. The increase of the balance sheet by a roundabout EUR 350 million is the result of IFRS 15 and 9. The following positions are affected, other assets rose by EUR 580 million and this is essentially driven by the capitalization of subscriber acquisition cost. Trade accounts receivable decreased by EUR 215 million, mainly caused by netting with liabilities, which are recorded for the first time due to IFRS 15. Trade accounts payable decreased by EUR 61 million, also because of these netting effects and other liabilities rose by EUR 470 million round amount and this is fundamentally due to the linearization of our commissions due from mobile network operators. So there are really mainly only technical impacts because of IFRS for the balance sheet. Cash flow information. Cash flow from operating activities amounts to EUR 65.8 million, this is lower compared to Q1 2017 and the reason is we also have a predicted and forecast that our position for the net working capital is much -- is higher in 2018 compared to 2017 and this is -- that's an increase of roundabout EUR 15.7 million because we have to do some payments to Media-Saturn.Cash flow from investing activities, we see that CapEx position EUR 11 million was much higher in Q -- 2017. Obviously, we had to do all the investments in the DVB-T technology to be ready for the switch of this technology in Q1 2017. So I think it's a fair number for Q1 2018. And cash flow from financing activities amounts to EUR 16.5 million and includes the leasing for [ DFMD ] and the interest payments, so that we also deliver again solid free cash flow of the EUR 54.8 million and you remember that once we have the guidance for 2018 in the last call we predicted an amount of roundabout EUR 50 millions on the level of free cash flow.On the next page, then we have again the detailed bridge between EBITDA to free cash flow. I think we see that change in net working capital, I think we will still be able to deliver the roundabout EUR 25 million we guided for the full year, you see the tax payments, CapEx we spent so far, we have to eliminate obviously Sunrise EBITDA because there is no cash impact. And just for the full picture, it's not Q1 relevant, but in the green bar for the full year we added already the Sunrise dividends which flowed in Q2 with the EUR 37 million just to give you the full picture because we have it on board right now.And then the last 2 pages, there is not much to say, then really we do confirm our guidance for 2018 with all the key parameters financial wise and operational wise.Thank you from my side, and I think we open the floor now for Q&A.
[Operator Instructions] And the first question comes from Wolfgang Specht.
Question from my side on the TV and media segment. Can you give us an idea of the split between the B2B business and the B2C business here?
The B2B business remains on the level that we saw on the previous years, I mean if you look at DVB-T and you count around [ 1 ] million subscribers by the end of period, then the average might be around 1 million which would create about 40 million in net revenues, but only under the assumption that all the vouchers have been used right at that moment. And according to that, the gross margin contribution, as we always said, is anywhere between EUR 2 and EUR 2.50. So I think then you can do the math. We will not give you a full breakdown on the EBITDA level, but again stable B2B and simplified matrix on B2C. Again, what we explained, I think that was not clear last year, and the charges of the license fees on each and every hardware is not on itself through hardware but on the produced hardware. And I think we have not explained that back in the last year in full detail and that might have led to -- or it certainly leads to a disproportionate distribution of EBITDA last year compared to this year.
And then maybe on the customer growth side, you definitely need to add some traction in order to meet your full year targets. Can you give us an idea what could be additional measures aside of what you started, for example, the idea to launch the waipu product on a white-label basis?
Well, I mean I think on DVB-T, we are very much on track and I think we will also see some net adds from the Sat business in the second half of the year. On waipu, as you awfully mentioned we have started the TV campaign, but it was on the -- if I remember correctly, around the 15th of March, so there is not -- no straight impact. I think the main drivers for an increased growth is -- will be -- well an announcement or a publication on the browser version, which we did not to -- didn't do so far and the fact that we'll not need intermediate devices such as Amazon Fire or Google Chromecast for some TV sets from June. We will also come out with a real, a pretty version in the course of the year, and referring to your question, well, there is a number of players within the market not only in Germany, did have currently request for proposals on what you would call a white-label version, that's correct. In some of those pictures, we participate, but we have not anticipated growth from those areas for the coming 3 quarters.
Maybe the last one on the MSP side, can you give us at least a rough idea how has the service revenues developed in the first quarter?
Well, the service revenues are -- I think Joachim can explain the impacts on the mobile revenues from IFRS in a minute. Other than that, service revenues were overall if you multiply ARPUs times average customer base, they are slightly increasing, but I think that is a simple multiplication. The main impact on the revenue side is that the commission payments from the network operators do not run into the revenue line anymore. But I think if you want to have more details, Joachim can elaborate on that.
I mean just again the IFRS 15 impact, I would even say I think we are on the stable level. The net classification of the commissions on the network operator brings down the revenue level, that's all about IFRS. For the mobile revenues, certainly we are in a more -- very solid stable base for the service revenues.
The next question here is Robin Brass from Hauck & Aufhäuser.
I have 2 questions, first concern on the postpaid segment and I would say everything is very stable, was it fair to assume that you don't really see any changes in the market from competitor side, maybe any pricing lower or higher here, so everything is stable and on track as well? And secondly also question concerning EXARING, so here what is also your plan, maybe, going forward to, let's say, having more of a critical math this year, because at least it looks a little bit and it goes back to the question before, but it looks a little bit challenging to reach a critical math here. So do you assume this year you could also do an M&A activity or you are in track to grow organically here, what is your view?
On the postpaid, I think overall what we see is that there is a bigger number of tariff offerings with higher data numbers included across the 3 network operators in Germany. We also see that the data usage on average is still going up, I mean this was all reported from our colleagues on the network operators to take the share in our ARPUs, 34%. I think these are the trends that we have seen for the past few quarters and they remain the same. The things that did not happen and some analysts at least anticipated was that after the merger of United Internet and Drillisch that prices also on the Telefonica network on the lower level would increase, and this is definitely not the case, if you would open today, check free advantage of our [ Veriovx ], so the price comparison [ portals ] you will still pump into offerings of EUR 299, EUR 399, EUR 499 per month with including 1 gigabyte. So I think on the very low discount end of the market, we did not see a relief, but there was also no real additional drop or more aggressive offerings on the typically area where MobiCom TVTEL is acting, so tariff plans anywhere EUR 15 to EUR 45. So I would say there is -- I don't see a change in any trend on the competition except for the fact that Telefonica is doing job with those 10 gigabyte tariff plans, even though there is still some skepticism on the network. I think they are doing a great job there, but other than that no changes. There was also a question that we have had in the last couple of days on the split between the networks. Well, I think we always did managed network split based on value generation on our side, so we have still Telefonica more on the somewhat lower tariff plans, but that will change now in Q3, because we have redefined some of their tariff plans on our portfolio in which this will be visible in the shops in June, July. But again this will not change in a significant manner over market situation. On waipu, well ,there are still 2 players out there in Germany or 3 to be precise. There is -- imagine there is Zattoo and there is [indiscernible] film which is a sub-label of Zattoo from all what we learned and imagine is that they have almost stopped their activities at least on the marketing side. They are upgrading their customer base in terms of -- they have been free of charge and now they're charging up to EUR 10. We have learned from our research that there is a certain proportion of customers going into the pay mode, but not significant. So they certainly are more on the dying end of their customer base, whereas Zattoo is doing good. And we are -- I'd say I'm happy with any market participants these days because the uptake in general of IPTV is still lower than in many other countries. I think that is [indiscernible] and I have to say I also underestimated the stickiness of German TV users with their cable or other [indiscernible] and we took the benefit from that in DVB-T2, but we see also that the uptake of IPTV is a bit lower. Nevertheless, we do believe that the summer quarters are the better ones for us, because people are more outside when the weather is good, there is the world championship coming up, so all these drivers will definitely, let's say, improve brand awareness and awareness in general for the entire segment or for the exist pass via IP. I would not rule out the potential and organic growth, but there is no concrete meetings, plans for business or negotiations around -- at this very moment.
The next question comes from Ulrich Rathe from Jefferies. Please go ahead.
First of all just to clarify in your prepared comments, you were making a statement on reaching an agreement Raj.com earlier or more quickly than expected [indiscernible] sure they're just referring to the LTE, MVNO or something else? That was my first question. And the second one is also clarification, I mean you said you don't want to break down the EBITDA in the TV media segment, but there was a comment on losses reaching I think EUR 6 million again, I didn't quite catch it, because it was very fast speaking, so I didn't quite catch it if you could clarify, And then more substantially maybe in the freenet TV base, you are sort of reporting the installed base and then the paying users, I'm just wondering what do you expect to happen over time in terms of the sort of non-paying base that stays non-paying. I mean do you have a sense now that it's been going on for some time that there is a certain base of use that's just happy with the free channels and they will never convert. What was your expectation for how this unfolds and because I also assume that the penetration of paying user keeps going up? And my last question is, on Slide 7, the installed base information essentially for end year - end 2017 and for end of March is sort of indicating larger than EUR 2.5 million for both, but the bar is sort of rising. Is that because you don't have much visibility on the installed base overall or do you have a sense of how fast that is still rising at this point in time? Thank you.
My comment on the so-called annual agreement, which Vodafone and [ DT ], let me refer to my colleagues, we have [ really ] always said it would never give a guidance unless they have that agreement, that was the reason, but we were always a bit later in the year with the detailed guidance. I was mentioning the fact that we have these annual agreement, these annual agreement determine whatever contribution we get from new tariff plan, whatever bonuses, marketing funds, et cetera, et cetera. The fact that we have agreed quite early this year, I think, just underlines the very good relationship that we have and it also pins down that we have -- we agree with DT and Vodafone also on aligning their activities. So I think I put it into the presentation to support the fact that our planning is at this stage even more solid than it was before and the fact that the relationship is a very good one. So maybe it was a qualitative statement which might have led to a different perceptions and thanks for that question on that one. On the EBITDA of the TV segment, we have had the discussion when we did the annual results. So what are the -- there is forces that obviously drives the EBITDA upon DVB-T2, that is mainly the number of subscribers. So that is -- certainly that adds gross profit and gross margin and consequently you would expect all to be ending in the EBITDA. But there is also some downward trend to it. There is still marketing spending now for the Sat, so you could not say last year they spent whatever [ EUR 15 million ] on TV and now this year has no marketing, that was one comment. The other comment is that last year the total of the installed base -- of today's installed base and I'm coming back to your question on that, it's not only the installment base, it's anybody who did produce hardware for DVB-T2. When their hardware was manufactured and shipped to Europe, because typically came from China, we were charging a license fee for each and every piece of hardware. And this was the -- this license fee is not related to a sell-through to the end consumer, but it's only related to the fact that the piece of hardware was produced and brought into the market, so we had an effect last year that even if there were -- I think there were in total about 3 million units. We charged license fees for 3 million units already last year and those license fees, well, they are not coming again, they are our one-time payments that it happened last year and this leads to an overall and these license fees is basically straight -- going straight to the EBITDA, because there's no effort on our side and this is the piece that is missing this year and so we have an upward trend from subscription and we have a downward or a one-off downward compared to last year from those license fees. As you say, the -- I think you're right on the bar, but the heading on the bar is the lower end was the paying users and not the upper one. We do not have very detailed numbers on the sell-through. On C+ moduled, we can monitor it, because we are the only producer and seller of it. On the STB or set-top box hardwares, we do not have reports on sell-through. The only figures we get is the ones from GFK and they are indicative, but not superb precise in that area and this is where we took the information. From the -- during the first quarter, we had around 60,000 units sold including our C+ modules. So if you reverse engineer it, we would say we are now on a conversion rate of approximately anywhere around 40%. We do believe that this will continue, but we also monitor in some of the hardware that we have sold long time ago are only now coming into the pay mode. And that might be the fact that people have put it into their summer houses, and whatever in their first round of the year, they didn't care about private channels and now there's some program on which they like and then they added. So it's hard for us to really monitor. We do question us on the individual customer base when they come in, and then we ask them what the motivation was, well, the answer is always, well, finally I came into the -- well, I wanted to watch private channels because there was specific program and so on and so forth. But projections are in this field extremely difficult, but from what we know, from what the sell-through reports of GFKs, we are optimistic for the [ 1.2 million ].
And let me add one additional comment, Christoph, I think you also asked about rate split up between EXARING and MEDIA BROADCAST for the level of EBITDA and definitely that's what we are -- the information we give you. You got it right, it's a minus EUR 6 million EBITDA for the EXARING and it's a positive EUR 14 million for MEDIA BROADCAST in Q1 17 -- 18.
The next question comes from Lizzie Dove from Goldman Sachs.
Just a couple from me. Firstly, I mean how much are you going to spend on advertising for waipu.tv and would you be willing to maybe kind of go over and above and maybe miss your guidance for this year? Secondly, you mentioned that you may be going to start changing your tariff with Telefonica Deutschland, is that because -- is that your choice or they're shifting their focus to their own sales channel? And then just one data point, do you have the number of gross adds for this quarter, please? Thank you.
We will not disclose the full detail on the waipu marketing spendings, the budget that we have created led to the numbers that we have published in the guidance. And we have said that the current number within the guidance also on waipu was more than 250,000 paying subscribers. If this would be reached, then we would be more on the upper [ decade ] of our guidance. If marketing spending would be as efficient as we expect, then we would be ready to spend more money and then we would most likely have -- well, we definitely have more subscribers and then we would be more on the lower end of our guidance anywhere between [ EUR 410 million to EUR 420 million ] whereas in the first one we will be more on the [ EUR 420 million to EUR 430 million ]. So, yes, we could imagine to do so and use the spend that we have communicated, but we are -- we do not expect or we do not plan or we do not project, I mean, any of our internal scenario planning going lower than the EUR 410 million. The other question was on the Telefonica. Well, we have -- the way we plan our network split within our customer base is one that is driven by lifecycle contribution. And I think it's fair to say that with the Telefonica tariff plans that we have had or the agreements that we've had for a long time, tariff plans with more into gigabyte for us were more attractive to be acquired within the Vodafone or the T network and we have agreed with Telefonica now on the new set of purchase metrics for bigger data volume and this is why we will do more than we did in the past. There is a natural difference in our customer base between Telefonica and the other 2 is the fact that in Media-Saturn, Telefonica represents themselves. So this is why any changes Telefonica will not hit our customer base in the same dimension as it would be with the other 2.
And the number of gross adds in the first quarter amounts to 585,000 customers. It's less compared to the last year because we had it in legislation against terrorism.
The next question here is Christian Fangmann from HSBC.
I have 2, if I may. And the first one is actually on waipu again. Could you kind of give us a feel of what the churn level is? I mean, these guys -- the new customers have monthly rolling contract. So clearly you're trying to convert the new free subscribers to be paying the subscribers, but you chose to see how the whole base is developing going forward, it would be good to have a feel of how the base that is paying [ develops ] in terms of churn. And then secondly on cash tax. So, the EUR 8 million, it looks like a low number, I mean you're guiding for EUR 50 million for the full year, and my understanding was, there was EUR 10 million of, let's say, rollover from '17 to '18, and so I was expecting that to be paid more or less in Q1. So maybe if you could give us a bit more color around the phasing of the cash tax for the year, that would be helpful. Thanks.
Yes, I will take the waipu question. Yes, we do measure churn. But we do not have a, let's say, extrapolatable metric so far, and the net adds that we saw in the first quarter of 31,000 there was -- I think that was about 8,000 to 9,000 customers that you would call churn. So they went out of the scheme, but part of it were because it was negative option, and I have to say that we have -- we are not yet in a good [ modus ] or we have not enough data to extrapolate it, because we've tested all kinds of channels, all kinds of different subscription methods or all kinds of promotion. Overall, we -- so far there is about 10% to 15%, but that's within the first 12 months, is that a reasonable figure which we can take for the future? I really don't know, I really don't know yet. And we will definitely come back to all of you with a, well, more reasonable and plannable number, but it's too early yet because there are too many dimensions, very different from Google Chromecast to Amazon. It's also driven by the way you can do the payment, is it via Amazon account, is it via our mobile, invoice part of the customers are straight billed to MobiCom [ TVTEL ]. So we are really in the middle of testing and it's too early to say. We have to be monitoring the net promoter score and still on the level of 37 with the paying users, so that is a very positive indicator. We do have a number of people telling us that they really now finally substitute their original TV access, but it's too many data to give a solid forecast, and this is why I ask you for a bit more patience, but I can assure you that I'm impatient as well, so I'm knocking the door of our BI team every day, but they still say that it's not solid enough to give projections.
And you're fully right, and that's what is communicated. There is a rollover in the cash tax item of roundabout EUR 10 million from '17 to '18 unfortunately and that's what I have to repeat again. Now, it's very hard to predict how this request will come in in the office and will be due for payment. So we think the EUR 10 million will occur in 2018, that's why we also keep the guidance on cash tax up with EUR 50 million. But so far, not the full amount is already here in our company and so there is no reason to pay.
Okay, that's helpful. And maybe one follow-up if I may. Actually, I was just interested in the development, yes, now I had the first month of basically offering the freenet TV over DTH, so satellite. So I mean I know that you can disclose anything, but how did the first month go in terms of reception in the market?
Yes, the first months was while we did not even have 1000 new customers, but that is I was trying to explain. It's only working with the C+ module and we have still C+ modules with DVB-T only on the shelf, and we decided not to replace it. So only slowly we are getting -- going into the market with the product. The first promotions with special offers and with more visibility will only happen during the Q2. This is why I said I'm not expecting a significant numbers not in Q2, coming in in Q3, but the real one in Q4. We do have run a TV campaign right now, we do expect that the World Championship will help a little bit, but still it's not so visible, it's kind of a soft launch for the sake of existing hardware on the shelf space. So it's a very low these days, but not as a surprise to us.
The next question comes from Simon Coles from Barclays.
It's Simon from Barclays. I just had a very quick one, could you just tell us how much of your sales are now coming from online channels and how that's changed versus the year ago?
Well, I think overall, reaccounting transaction, so that's including the renewable and the gross adds, and on the total, the retail branch is doing still about 60%, it's about 40% online and that is a growing number from last year, but we do not measure it on a percentage, what we do measure, because, well, we have click and collect, we have [ Oppo ] and all these other stuff, but it's a growing number and it's a -- transaction is more than 40% that's online.
The next question comes from Usman Ghazi from Berenberg.
I have 2, please. Firstly, I was just wondering if you could give us some insight into what the headwind is on EBITDA this year from the absence of these license fees, so we can have a better year-on-year comparison. And then the next question was on the mobile gross profit. I mean it's flat which is fine, as it's as per your guidance was stable, but I'm just wondering why it's not growing when your mobile service revenues are obviously doing quite well with the growing customer base with growing ARPU, so any insight there would be helpful? Thank you very much.
Well, on the license fees, I guess you referred to the DVB-T2, the license fees -- the difference from, let's say, income from license fees is around EUR 5 million. So it's EUR 5 million less than last year.
That's for Q1 or--?
Well, that's Q1 -- well, I think it's Q1 and Q2. So it's first half year and it should be different. On the mobile gross profit, well, I think you're right, on the one hand we have stable ARPUs and the growing customer base on postpaid, but on the other side, we also have customer acquisition cost, retention cost and -- well, if you have 60,000 more customers on postpaid and you look at the average growth margin, then the growth is not that significant, it's only 1% on the total customer base by -- on a quarterly basis and parts of it are certainly eaten up by acquisition cost, parts of it are eaten up, because the data revenues are -- they have lower margin over the past than voice has, even though the ARPU overall is higher still, SMS is kind of a dying media which was on many cases still charged, roaming revenues are changing again. So there is not a single, but a dozen of impacts that lead to the fact that the gross profit is stable and you don't see kind of a relative or a proportional increase.
Can I ask maybe a follow-up, I just wanted to confirm that the EBITDA is not impacted by the IFRS 15 adjustments. So, it's only the revenues which were getting impacted. Is that right?
That's 100% right.
One follow-up question from Ulrich Rathe from Jefferies as well.
Can I just follow up on the comment you just made on data revenues versus the voice revenues, is this sort of more general statement on sort of particular tariffs, because I thought that a lot of your margins are essentially determined by the tariffs, your selling principle. And I think sort of the split between voice and data revenues within the tariff, is this a bit arbitrate. So I'm just wondering how exactly did you mean that, is it that simply that you are taking more data, heavy tariffs, and that in more data, heavy tariffs margins are lower or is there sort of a formula there, I'm not entirely sure? Thank you.
Well, there is nothing that we have always talked about, at least I remember that I said it that on a data tariff plan or on the data component of the tariff plan, our relative margin is lower. One driver is that on voice, we can still do arbitrage gains, we still have a proportion of metered revenues, at least on the purchase side. Whereas on data, we do not meter, so we source with any of the networks, 1 gig and it doesn't matter whether the customer uses only one megabyte or the full 1 gig. So these kind of arbitrage gains that have happened before are minimized on the data, and this is one driver. The other one is that on data upgrades for example, if the customers run into the -- I'll say that into the -- if they cross the limit and then they meet a speed upgrade in order to continue while the speed upgrade is a specific option and our margin on speed upgrades is lower than in general on service revenues. So then if we do big data tariff plans, we still try to be cheaper than the network operators. Well, we have to take it from the margin out of the data there. So data component is typically an extra in our purchase model and that's the reason why. So it is systematic difference from purchasing voice-only, then we have combined tariff plans where the margin is kind of an average of data and voice and there is also pure data component and pure data components have a lower margin.
Thank you very much. There are no further questions.
Okay. Ladies and gentlemen, thank you very much for joining, it was great pleasure. Thanks for your questions. We look forward to our next meetings and our next rounds, and I wish you all the best, and thanks again. Goodbye.
The conference is no longer being recorded.