Good day, ladies and gentlemen. Welcome to KPN's Third Quarter 2020 Earnings Conference Call . At this time, all participants are in listen-only mode. We will be facilitating a question and answer session towards the end of today's prepared remarks. If you would like to ask a question, you may do so by pressing star one on your telephone. Please note that this event is being recorded. I would now like to turn the call over to your host today, Mr. Reinout van Ierschot, Head of Investor Relations. You may begin, sir.
Thank you. Good afternoon, ladies and gentlemen. Thanks for joining us, and welcome to KPN's third quarter 2020 results webcast. With me on the call today are Joost Farwerck, our Chief Executive Officer, and Chris Figee, our Chief Financial Officer. Before turning to the Q3 presentation, I would like to remind you of the safe harbor statement on page two of the slides, that also applies to any statements made during this presentation. In particular, today's presentation may include forward-looking statements, including the company's expectations with respect to its outlook and ambitions, which were also included in the press release published this morning. All such statements are subject to the safe harbor statement. Let's now move to the core of the presentation. I'd like to hand over to our CEO, Joost Farwerck.
Yes. Thank you, Reinout, and welcome to our third quarter result presentation. Let me first take you through some of our highlights. First of all, in the quarter, we saw EBITDA growing at 1.3%, and we saw solid free cash flow growth. We also continue to deliver on cost savings, which is reflected in the results. Operationally, there were positive signs in the consumer segment. We saw the post base returning to growth and accelerating inflow of fiber customers fueling a stabilizing broadband base. COVID-19 continued to impact our revenues, but lower costs partly mitigate the impact on EBITDA level. I'm proud of our employees who continue to service our customers in these challenging times. We execute on our strategy, and with a solid set of results, we're now able to give a clear picture of the full year's outlook. Next, I would like to highlight a few key figures.
Corrected for a number of divestments, revenues declined by 3.7% year-on-year. Growth in wholesale was offset by lower revenues from consumer and business. Roughly 0.7% of this decline is related to COVID-19. EBITDA increased 1.3% year-on-year as the effect of lower revenues was more than offset by good cost control. Free cash flow increased more than 7% year-on-year to EUR 241 million. Chris will give you more details on our financials later in this presentation. Turning to our outlook. In the first nine months, we are on track with the execution of our strategy while maintaining a robust financial position. We are now able to give a clear picture of our full-year outlook. We expect adjusted EBITDA after leases of approximately EUR 2,320 million, CapEx at EUR 1.1 billion, free cash flow of approximately EUR 750 million.
To be clear, this free cash flow and CapEx outlook includes all acquisition CapEx as well. We reiterate our dividend commitment, and we intend to pay a regular dividend per share of EUR 0.13 over 2020. In the first nine months of the year, we have seen solid progress on each of our strategic pillars. As we generally do every other year, we will host a strategic update to the market in the afternoon of the 24th of November. On this day, we will provide you with an update of our strategic and financial ambitions for coming years. Fiber is the best and future-proof technology, and our organic fiber build increased further in the third quarter, even though this included the summer holidays during which construction capacity is lower. Fiber activations accelerated, providing us solid activation rate of approximately 50% over the 12 months past.
We continue to add new areas where we roll out fiber. We're currently active in 90 areas, and we are reaching completion in six of our 25 largest cities. Fiber clearly outperforms copper on all metrics and fuels a stabilizing broadband base. Fiber average revenue per user is higher than copper. The churn is much lower, and the Net Promoter Score is a number of points higher. Consumer fiber revenues grew nearly 8% in the third quarter, driven by a growing base and an uplift in ARPU. Fiber service revenues are bound to exceed copper service revenue declines as our rollout is ramping up further. All in all, we see a positive net present value from our fiber investments. Let's now move to the performance of the segments. Fixed consumer revenues were somewhat lower this quarter, fully driven by a decline of legacy services such as traditional voice and Digitenne.
The trend is starting to turn for mobile service revenues. Although still declining at 4.4% year-on-year, this is an improvement compared to the first half of this year. Q3 NPS stood at +12. Customer satisfaction is one of our main priorities, so this result is clearly disappointing and we have a plan in place to restore customer appreciation. This decline has to do with our customers having more questions and need for support now that the majority is working from home, which is putting pressure on our customer service center. A bit in industry trend, and a change in the interface of our Internet Protocol television product, leading to questions since some customers need to get used to the new layout and features.
Looking at our consumer key performance indicators across the third quarter, we can see that the converged base is relatively stable around 50% of broadband customers and more than 60% of postpaid customers. The total broadband net adds were in line with last quarter and moving towards a stabilizing retail base. The commercial success of our fiber rollout is reflected in 27,000 new fiber customers, and we're confident that this will contribute to an improving broadband base going forwards. After a long period of decline, the mobile postpaid base returned to growth, and at the same time, we kept the postpaid ARPU stable at EUR 17. Travel restrictions due to COVID-19 led to lower roaming revenues, and this was partly offset by increased national out-of-bundle usage. We're happy to see the underlying trend developing favorably, with improving inflow ARPU from our unlimited bundles and the effect of repricing wearing out.
We can move to business segment. Business revenues declined nearly 8% year-on-year, largely impacted by COVID-19 as we saw lower roaming revenues and delayed IT projects. Our strategic customer migrations continue to impact service revenues in the short term. After the migrations, the new propositions provide us significant opportunities for up and cross-sell of additional cloud, security, and workspace services. As soon as we obtained the new spectrum license, we switched on our 5G network and introduced distinctive 5G services for our B2B customers, as mentioned in this slide. We are currently the only operator in the Netherlands offering 5G value-added services and supporting businesses in developing new applications and optimizing their process with 5G. We are making solid progress on the migration of our business customers to the new portfolios.
However, the pace was somewhat impacted by COVID as we were often unable to enter the customer premises. We continue to innovate on our business product portfolio. For KPN One customers, we now provide a 4G backup facility, ensuring greater business continuity. We added several cloud communication services such as BroadSoft, Microsoft Teams to our Smart Combinations portfolio, and these are indispensable features for teamwork when working from home. Let's move to wholesale. Correcting for the sale of Next Level Data Centers, the data centers last year, revenues in wholesale increased by almost 5%, partly substituted by the sale of small assets. We added 22,000 broadband lines in our wholesale segment. Looking at our total portfolio of wholesale and retail, we see solid growth of broadband penetration on our network, leading to a total broadband network share of approximately 52% in the Netherlands.
We also renewed another long term mobile virtual network operator contract in the quarter. All in all, another good quarter for our wholesale division. We believe that sustainable business is better business. We have reached some important milestones in this respect, and we also have an ambitious agenda for the years to come. Commit ourselves to the three Sustainable Development Goals from the United Nations mentioned in the slides, and during the quarter, we made some nice contributions to these SDGs, as you can see over here. Our efforts in this respect do not go unnoticed, and KPN continues to be recognized by various ratings and benchmarks. Now I would like to hand over to Chris Figee to give you more details on our financials.
Thank you, Joost. Let me summarize our group financials for the first nine months by explaining like-for-like performance. The adjusted revenues of KPN declined by 2.9% year-on-year, partly attributable to COVID-19. The adjusted EBITDA after leases increased 1.5% year-on-year. Some lower revenues were more than offset by strong cost savings, net leading to a positive result. Our EBITDA margin improved 200 basis points year-over-year to about 45%. When we compare Q3 of this year to Q3 last year, we see a growing EBITDA. Year-to-date, free cash flow grew by 13% versus the same period last year. Our operating profit was EUR 150 million lower due to several incidentals like the book gain on the sale of NLDC in 2019. However, excluding these incidentals, operating profit would have increased by EUR 29 million year-on-year.
Similarly, net profit was EUR 144 million lower as it was boosted by the same incidentals last year. Excluding these, net profit would have increased by about EUR 31 million, mainly driven by lower financing costs. When it comes to cost and cost control, we are continuously digitalizing and simplifying our company, leading to improved services and more efficient operations. We are well on track with our cost savings program and realized an additional EUR 44 million savings in the third quarter. Also, due to COVID-19, we see lower costs in areas of travel, learning and development, and housing and facilities. In total this year, we've reached about EUR 115 million cost savings. In total, since the beginning of the program, we've saved EUR 256 million. We are confident that we will exceed our EUR 350 million target by the end of the year, 2021.
We're about 75% of our targets after seven out of 12 quarters have gone by. Operational free cash flow for the first nine months of the year stood at EUR 900 million, stable at 23% of revenues. This number was somewhat lower compared to last year, fully driven by different CapEx phasing through the year. In terms of free cash flow, we've seen a strong increase this year. Our free cash flow of nearly EUR 500 million year to date was 30% higher year-over-year and moved to also 13% of revenues, a high free cash flow margin. This is a result of lower cash restructuring, lower cash interests, some lower investment in working capital year to date, and lower cash taxes.
We ended the quarter with a strong cash position of about EUR 800 million and expect this to grow further whilst having paid EUR 381 million as a final spectrum payment and EUR 180 million interim dividend in the quarter. We refinanced a EUR 460 million bond in September. We have committed to being more open about our working capital position, as you can see on this page. Notable effects impacting our working capital position as to date are lower trade receivables, mainly driven by lower sales levels. Trade payables are lower due to specific initiatives. Generally speaking, we see a peak of incoming invoices in December and we see lower accruals driven by settlements, lower interest accruals, and bonus payments related to 2019. Our working capital program has yielded tangible benefits, being much less of a drag on free cash flow in the first nine months of last year.
Actually, in the third quarter, working capital contributed positively to the free cash flow. Year to date, working capital is still a drag on our cash. The prepayment related to spectrum auction is corrected at the bottom again. It is formally a prepayment, but not a working capital item. To be specific, this is a portion of the auction related to the 2,100 MHz frequency. KPN's total liquidity was strong at the end of the third quarter. It consisted of almost EUR 800 million of cash and EUR 1.25 billion undrawn revolving credit facility. During the quarter, we issued a EUR 600 million bond at a 12-year tenor with a very low coupon of 0.875%. We initiated the issue with EUR 500 million, but saw ample demand, and the order book was four and a half times oversubscribed.
As said, we also redeemed a 460 million EUR bond that was swapped to a fixed rate of just over 1%, lowering cash interest by five million EUR next year. In sum, these transactions lower our average cost of debt and increased the maturity profile of our debt book to about 6.4 years. All in all, KPN's liquidity is sufficient, abundant, and covers debt maturities for the next three years. In terms of balance sheets, our financial position remains solid. At the end of the quarter, our net debt to EBITDA ratio increased slightly to 2.4x , driven by the spectrum and interim dividend payments. We expect to drop marginally during the rest of the year. Interest cover ratio improved to 9.2x and the weighted average cost of senior debt was 65 basis points lower than last year.
Since we want to be open and transparent, we find it important to continuously improve our disclosure. We've consistently delivered more disclosure on critical items and are on track to fulfill and complete our disclosure agenda for the year. In the strategy update planned for the end of November, we'll give more insight into the open points. One, the fiber-related CapEx by splitting CapEx further into fiber and non-fiber and into households steered by consumer, specific on ARPU reporting, and thereby complete our disclosure agenda for the year. The summary to close off this short presentation. KPN had a very healthy financial quarter with EBITDA and free cash flow both up versus last year. We provided a more specific range of our full year outlook and are on track to reach it.
The encouraging signs in the consumer segment, the mobile market showing improvement dynamics, a strong fiber uptake, gradually outweighing a decline in copper base. In B2B, we continue to face revenue headwinds that are progressing with customer migrations and product innovations. In wholesale, we see ongoing success of our fixed and mobile portfolio, indicating a successful open network policy, we maintain a robust balance sheet and solid liquidity position. Finally, we look forward to informing you more on our visions for the coming years in about one month from now. Now back to Reinout and to your questions.
Yes. Thanks, Chris. We can now turn to your questions. As usual, please limit your questions to two each. In case there's still time left, you can always ask more questions later in this Q&A session. Operator, over to you.
Thank you, sir. Ladies and gentlemen, we will start the question and answer session now. If you would like to ask a question, you may do so by pressing star one on your telephone. Star one for your questions or remarks. Go ahead, please. Our first question is from Mr. Michael Bishop of Goldman Sachs. Go ahead please, sir. Your line is open.
Thanks. Just the two questions from me then, please. Firstly, on the Fiber-to-the-Home build, as we think about moving into next year and beyond, could you just give us an update in terms of the discussions with respect to contracting more build capacity in the Netherlands and how they're going and whether you're still thinking the same sort of five to 600,000 run rate is something to target in 2021 and beyond?
Is on the mobile service revenue. You mentioned that trends were now turning. How should we think about the mobile service revenue again going into next year as you potentially start to have some easier comps? Thanks very much.
Yeah, thank you. First on the fiber build. What we did is scaling up our fiber rollout production capacity to a level that we didn't do for a decade or something like that. A lot has to do with the lock-in of construction capacity, and that's what we did. Having said that, we're scaling up in a very efficient and solid way. Last year, we did like [ 120,000] fiber to the home connections. I think this year we will reach a level of 300. Five to six is a little bit stretched, to be honest, for next year. For me, it's important to scale up in a decent way to a level of 400 or above. Then we're on a scale that we've never done before at KPN.
Then we're on a scale that we do almost the same amount as the whole consolidated market this year in the Netherlands. For that, we have locked in the capacity. It's very important for us that we do it in such a way that it is, also from a pricing perspective, efficient. Mobile service revenues. First of all, maybe Chris, you can take over. I am, or we as a company are super focused on improving that trend, because for a long period of time we were under pressure. I think KPN is positioning itself much better in the market, even on the higher end of the propositions with unlimited. Also on more in the lower price segment, although most of the real low price propositions are eliminated in the market. I'm happy with that trend, and it's for us very important to continue.
Maybe you want to add something.
Yeah, Michael, on these mobile service revenues, if you look at that one single revenue line. On a sequential basis, they have been actually fairly flat during the year on mobile service revenues. If I look at the progress into current and future quarters, I break it down base times ARPU. Base, we witness growth for the first time in a long time. Actually, when you look back, the numbers start to grow end of June on a monthly basis. At this point in time, we still see continued positive base developments so far since June. When it comes to ARPU, our ARPU has been, of course, at a slight decline this year, all rounded to EUR 17, close around to EUR 17.
Where in the summer we expected a more negative impact from roaming. Of course, the negative impact was there, but countered by supported by out of bundle calling. Our ARPU over the summer kept up better than we actually had hoped for. The two other elements are the leakage from repricing of our back book is turning less and less, Net inflow values become better. Secondly, we see some ARPU support from more unlimited volumes. I don't think our ARPU will immediately go to 18, We see in general quite encouraging base developments and stable to possibly slightly growing ARPU going forward if we manage to hold on to an increasing share of unlimited. For mobile service revenues, we actually see quite a good outlook.
Thanks. Really helpful detail.
Our next question is from Mr. Luigi Minerva of HSBC. Go ahead please, sir. Your line is open.
Yes, good afternoon. Thanks for taking my questions. The first is on your outlook for CapEx and free cash flow. I think you are presenting a very convincing picture with regards to the fiber case, the fiber upgrades. Likewise, you signal in the presentation that you may exceed your cost savings target. My question is whether you can afford increasing CapEx going forward while preserving still the free cash flow generation that we will see this year. Whether there is enough room on the cost-saving side to just reinvest cost savings in CapEx while preserving free cash flow generation. The second question is on an update on the regulation side. Has the Electronic Communications Code becomes part of a national law? What is the outlook you are envisaging in terms of particularly wholesale fiber access? Thank you.
Okay. Luigi, when it comes to CapEx, you have to wait for more detail till 24th of November, I give you a few snippets. If we see highly effective returns on fiber, we are accelerating the rollout, going forward, CapEx allocated fiber will go up. On the other hand, we see non-fiber CapEx to come down. We'll save on non-fiber CapEx, gradually with the cost opportunity kicking in, that will also provide support to our free cash flow. The final numbers, bear with me, it's only four more weeks, we'll give you a more integral view.
The components will be around supporting EBITDA and cash generation with cost savings, slimming down non-fiber CapEx and increasing fiber CapEx, and that will lead to a final outcome. Also the point, of course, at the end of the day, after free cash flow, your dividends, the progression of our dividend, that is really sacred. I will definitely preserve that. The rest, Luigi, that is what I can tell you right now, and for the rest, bear with us in four more weeks, we'll give you the full picture.
Thank you.
Your question on regulation. BEREC guidelines to be implemented mainly with respect to symmetrical access, of course, is what we focus on. What we think is that we maintain our open wholesale policy. We allow access to passive infrastructure where possible. There is abundant investments in fiber in the Netherlands. There are at least two high-speed fixed connections in every home premises in the Netherlands. That is what we think why we don't need a Authority for Consumers and Markets regulation as we used to have in the Netherlands based on the European Electronic Communications Code. We understand that our regulator, of course, is working on that, and we're in close contact with them. We expect ACM to open up a consultation on their policy in the coming months.
There's some delay in the Netherlands, so they will implement, but the whole process of the consultation will take longer than planned for. That will move deeply into next year. Where we come from is that we appreciate the fact that we need a regulatory framework. Currently, we're not regulated. That's due to all the discussions we had with our regulator. We think that we should avoid further price regulation on our access obligation. The current framework is working quite well. The best proof point is the market, the Dutch market. If you look at what's happening now is that we lost 4,000 consumer broadband customers. We saw an inflow on the wholesale side of 22,000. Other service providers competing against KPN are doing good business on our network. That's the best proof point. We're in the middle of the process.
Will take some longer due to delay on the government side, but at the end, it's something we had to cover, of course.
Thank you.
Our next question is from Mr. Usman Ghazi of Berenberg. Go ahead, please sir.
Hello. Thank you for taking my question. My two questions are, there's one on free cash flow, and the other one is on network separation pieces. On free cash flow, I guess you put out a guidance for full year of EUR 750 million, which is implying that in Q4, the free cash flow could be down around EUR 30 million, roughly year-on-year. Given that the first nine months is up EUR 50 million year-on-year, why should we expect Q4 free cash flow to be down significantly, particularly given it's a big working capital inflow quarter. You're obviously doing well on that front. Any color there would be interesting.
The second question was, I was wondering if you could update us on just the board or the management thoughts on how you view the advantages or disadvantages of network separation at this point in the deployment cycle of fiber. Thank you.
Yep. Usman, on the first question, I'll take it. On our free cash flow. Indeed, your point mathematically is correct. A few points coming in. When you look at our EBITDA, not all our EBITDA is coming in cash in this period. This point in time, you see some delta provisions. Secondly, on the interest rate side, a small point is we've got a issued a bond with a short first coupon. There'll be some small interest payment in December. Finally, indeed, we do see working capital commitment in the fourth quarter that has to do with the phasing of CapEx through the year. Our CapEx is more evenly spread throughout the year than last year. We often had a spike in Q4. This year, Q4 will have the repercussions of CapEx increase in the year.
Of course, the fiber ramp-up would also require some prepayment. It's a combination mainly on CapEx timing on the fiber component that will have the drag on working capital in the fourth quarter. Some smaller effects being the short first coupon, so non-cash EBITDA or EBITDA that comes in in cash, but gradually over time, that ultimately to the fourth quarter free cash flow.
Yeah, on the topic of network separation, it's of course, always an interesting topic in our industry. First of all, I see the value of having a clear, visible network company on one side and a good service provider on the other side. That's exactly the outcome of our Simplification Program. We are simplifying the back end and the front end of the company at the same time. We're moving to a far more clear, visible NetCo, service company model, and it's very interesting to run your company like that. The current value we see is to keep it integrated as one company, and by that, create the most value. Of course, we looked a lot to what's happening in other countries. I think in some countries, network separation was a bit artificial. Splitting up a company with pricing agreements between the NetCo and the ServCo being artificial as well.
In the Netherlands, that part is moving to a more market pricing. Like I just said, on the wholesale side, it's working. We're simplifying the back end of our company. We're rolling out more and more fiber, and we're simplifying the ServCos. We have Solcon, Simyo, Telfort, KPN. We took out Telfort. We're migrating the broadband base as we speak. Probably in a couple of years from now, we have a logical moment where you suddenly see a ServCo and a NetCo being KPN. That's the outcome of our simplification program. I don't believe in an artificial split in the wrong moment of time.
Thank you.
Our next question is from Mr. Polo Tang of UBS. Go ahead, please, sir.
Yeah. Hi. I'd like to ask two questions. The first one is really just about the fiber cost per home passed. I think previously you talked about roughly EUR 700 per home passed. Are you seeing the costs come down as you scale up the fiber build and commit to more volumes going forward? The second question is really just around your business unit. We've obviously seen revenue declines there accelerate to -7.9% in Q3 after -5.7% in Q2. Can you clarify how much of this step down was because of absence of roaming, and how much was because of proactive measures and transformation that you were doing away from legacy revenues? Were there other factors?
Are there any data points that you can share that would give us comfort that business revenue declines can ease going forward, and that the underlying business trends in the unit are resilient despite COVID-19? Thanks.
Yeah. Thanks, Polo, for your questions. Your first question was on the fiber rollout costs. In the past, we worked above EUR 1,000 per house. We went down and down and down, and then we aimed for something like EUR 750, which was all related to the areas we picked. It really depends on the area one picks when it comes to rollout costs. On these areas, we do EUR 700 today. When we move to other areas, it could be like EUR 800 or EUR 900. It's really where we rollout. Others did roll out in rural areas, for instance. We never did that because that's like EUR 4,000 or EUR 5,000 per household. It depends on the business model. Maybe a one-off fee you can ask from your customers that you move in these kind of areas.
On the current rollout, I think, Chris, we do roughly for an activated line between EUR 800 to EUR 900. The activation is until the Fiber Termination Unit included. It's a fiber to the home, moving into a household, into the cabinet with an FTU on the wall. The only thing you need is a modem. If I look at the way we contract now, the contractor's costs are always getting cheaper because of the rollout, the numbers we give them, and the areas we select, and the time we give them to stay in an area like that.
Yeah, to me, it's a function of, I think, the naked price per home is gradually coming down. It's not a revolution, but gradually coming down. At the same time, if you move from the more dense cities to slightly more the outer suburban areas, they tend to have more digging distance. You have to dig more meters in the ground. That means that has an upward effect on the cost. The clear cost per house is coming down a bit, but you sometimes have to dig a bit further if you move away from the highly populated areas to the more suburban areas. To us, it's important you lock in your production capacity at the right numbers, at the right price, so that makes it predictable. Polo, your second question on the business segment.
You mean roaming or COVID has a significant impact on the business segment, mostly on the roaming. Interestingly, for example, in the consumer segment, we saw the roaming drop compensated by more out-of-bundle calling. You don't see it in business. I'd estimate that the EBITDA decline in business is about 60% driven by COVID-related revenues, 40% other. COVID is roaming but it's also networking, hotspots, and IT revenues. Of course, the roaming at some point, it's not going to decline any further. The year-on-year comps will start to look better in 2021. I would estimate 60% of the EBITDA decline is COVID, 40% is not. The other 40% has to do with some RGU pressure in mobile with underlying and the structural decline in fixed voice that is countered by increasing base numbers, but that still leads to a decline.
For a chunk of decline, year-over-year, I think the comps will start to look better next year as somewhat roaming cannot drop below zero. Also the uses of hotspots, Wi-Fi locations cannot drop below further next year.
Yeah. When it comes to business-to-business, we always talk about the top line pressure in B2B and when it's going to inflect. In the future, I hope to give you more insights in how this is built up between small and medium enterprise, large enterprise, and the real corporate customers. 50% of the EBITDA we make in B2B is SME. There we're much further than on the corporate customers. It is important that we give you more insights there. We're more positive than we were on B2B, or at least I am more positive than I was on the B2B outlook perhaps two years ago. It will not inflect soon, but part of it will, and that's the higher value part. That is important for us to focus on and to give you more insights in.
Thanks.
Our next question is from Mr. Keval Khiroya of Deutsche Bank. Go ahead, please, sir.
Thank you so much. Two questions, please. One on the FTE reductions and one on working capital. First, on headcount reduction, obviously, we have seen a slowdown in the rate of cuts this year, and also quite a modest cut in Q3. Would you mind giving us an update on how we should think about the pace of reductions going forward, and anything you can share about discussions with the unions as well? Second, on working capital, you've obviously done a very good job at improving the working capital outflow. You previously have discussed that it was to be negative as the fiber rollout ramps up. Is there anything you could share about how we should think about that going forward, whether it will still be the case as fiber continues to ramp up? Thank you.
Yeah. Thank you. On FTE reduction. First of all, when COVID-19 started, we took the decision somewhere in March to stop the reorganizations. The request of advices we were going to send to our works councils for the period Q2 and Q3. We stopped that for six months, and we gave it a go a couple of months ago. We delayed an FTE reduction. We didn't delay an FTE spend. We immediately took the action to send a lot of hired personnel home. We have a blend of roughly 10,000 KPN people and two and a half thousand hired people on a daily basis. We really did a tough action on the hired staff. Now we're scaling up the reorganizations again. On FTE, we will not meet our original plans.
Although things are moving up to a speed we are used to, a level we're used to. We're in good conversation with our works council and our unions. We will continue there. We're simplifying the organization, and we all understand, our works council as well, that it doesn't make any sense to stop laying off people when you simplify, because that's not good for the people either. You can't hostile them in KPN. We're good for our people. We increase salaries. We took all kind of measures to make them comfortable at home. We give them additional payout per month to support them at home. I think all in all, we did a good job, and that allows us now to continue our reorganizations as well, which is at the end, good for the company, and so good for the people working in KPN.
When it comes to your working capital question, the upward push on working capital from fiber definitely is there. That's inherent link to that piece of that business activity, where fiber you tend to pay more upfront for the non-fiber CapEx. That increasing, investing in fiber has an upward push of working capital. We've been able to mitigate some of it, both in our fiber contracting, so arranging payment terms with some of our construction companies and contracting partners, and improving the working capital on the non-fiber space by invoicing earlier, paying later, changing payment terms, thinking about rolling billing solutions for our customers, et cetera. Our working capital program has been all inclusive and broad, encompassing many features. That has been able to mitigate some of that fiber impact.
At the same time, I can't fully mitigate it because the inherent shift to fiber will create upward push on working capital. Again, we're trying to compensate in many, many other areas, and I'm proud of what we have achieved so far. The success of it has been pretty good. There still is an undercurrent of the working capital increase. You feel it when you see an increase in fiber. As we said also at the beginning of the year, the delta in fiber, the increase in fiber commitments, actually, that drives that working capital push, and that's what we experienced this year. Again, I'm pretty proud of what we've achieved and how we've been able to counter some of that. I can't take it fully away.
Our next question is from Mr. Paul Sidney, Credit Suisse. Go ahead, please.
Thank you. Good afternoon. Just a couple of questions, please, from me. The first one, just on fiber. Can I just clarify some comments you made on the fiber build in response to an earlier question? Did you say that you've done 120,000 by the end of 2019, around 300,000 likely this year? Reaching the 1 million by end of 2021 perhaps looks a bit of a stretch. Just so I've understood that correctly. Second question on the Dutch broadband market. Your retail broadband adds have stabilized, helped by fiber, and wholesale continues to do very well. Do you see this now as a much healthier balance between retail and wholesale, especially, post the move away from Telfort? Thank you.
I think to start with your second question, I think the balance is much healthier because in the past we saw us losing 20,000 customers, and we saw 20,000 customers coming in on the wholesale side. The current balance looks much better. Of course, on the retail side, we aim for crossing that zero line and grow. It's pretty good leverage on the wholesale side. We are successful on fiber. The fiber case works. We go above 60% penetration of a lot of areas, but that's done with the other service providers, our wholesale customers as well.
A long time ago, I was part of the team that built the fiber case, and we always included wholesale as an important part of the business case, because if you want to reach levels of 70% or 80% penetration, it's best for market to do it together with the other providers. This is a good balance. On fiber build, I didn't mention the 1 million, not to reach next year, but I think if you look at the line of 120,000 last year, this year, 300. We are going to reach that 1 million for sure. That I promise you. Probably it's not going to be end of next year, but a quarter later, two quarters later. For me, that's not that important.
That's just a number, we will continue to roll out fiber, we should do it in an efficient way, in a prudent way, keeping an eye on our financial but also operational excellence. We come from an era where fiber NPS was -14% because of everything we did in areas before we connected customers. The current thing is that we execute end-to-end on the fiber value chain, that's working quite good. It's a careful process. If we scale up above 400, reach a level of 450 next year, we do almost 100 more than we ever did. I like the idea of the 1 million in three years, doing like 600 per year is a bit too stretched for next year. I would aim more for 450.
Look, I think, we've done 73,000 homes passed in this quarter, in the summer. We'll continue to ramp up. You can see the weekly production scheme going higher and higher. We'll continue to increase our weekly production in these weeks. Expect the quarterly production in the third quarter of at least 75,000 to possibly a bit higher. That means if you think about that, we'll get very close to 1 million. We may miss it by a few months. If you look at the speed that we roll about a quarter, and I think you're going to get to 80,000-1 00,000 per quarter pretty quickly, the 1 million will reach. It may take a month or two longer. That's kind of the scale of the deviation that we're talking about.
As Joost said, I think I'd rather do it in a measured way, being effective and try to convert customers. That to me is the most important thing to be controlled. The million will reach, and it may take a month or two longer, that's kind of what we're talking about. Given the scale of the speed of what we do.
Just a quick follow-up. Once you hit the million, presumably the plan is still to just push on and keep on going.
That's why I say, I don't wake up in the morning with that 1 million in my head. Our Technology & Digital Office does, by the way. That's his target. We're not going to scale up and then suddenly stop. We've been there, and it's super inefficient. We need a machinery that is working consistently at against low cost, efficiently, first time right. We want to see these kind of numbers. We want to see that the connection is first time right, and then that we don't have to send two or three field engineers. I'll give you an example. We always talk about the rollout cost per homes passed. Behind that is a service field engineer cost that in the past, we used to send field engineers three times to a household before we connect a customer real-time.
Fiber rollout engineer, the one engineer into the home to install the FTU, and then at the end, an engineer to the households to activate the customer. We now designed a way of one engineer taking care of everything. Maybe these contract costs are not going down, but they do much more for the same amount of money, and they activate customers for us as well. That all is now in the machinery. We're improving there and scaling that up in a first time right way, and then ending above 400. I feel pretty good. We don't meet the 1 million next year, but we will meet it, like Chris said, a couple of months later.
That's perfect. Thank you very much.
Next question is from Steve Malcolm of Redburn. Go ahead please, sir.
Good afternoon, guys. Two questions, please. One just on B2B. I know you don't give us any EBITDA, but can you give us an idea of what the drag on group EBITDA is from the underlying declines in B2B is at the moment? I guess what the prize is from a group perspective on stabilizing that number? That'd be really helpful. Then just on the price rise, you take a price rise every June, July every year, but slightly different year this year. Just interested, any comments, you got any feedback from the customer behavior around that price rise? It seems like all the KPIs were pretty solid. Whether you were seeing any sort of particularly different behavior on TV, kind of spin down, fiber, those kinds of things.
If you're interested just to understand how the price rise landed this year relative to previous years. Thanks.
Yeah. Thanks. To be honest, it landed much better than last year. We have a new communication chief, by the way, and we really worked on the introduction of the price rise this summer. We made it less aggressive, and we explained that we do it on a level that we do on the Collective Labor Agreement to our own employees. We increased the CLA to our employees, and on the same level, we increased our tariffs, because we think that all companies should increase their CLAs, the salaries for the people, because of the consumer price index. A little bit on the positive side, we were there for us, I should say. At the end, we had a whole storyline build up on, listen, this is what we do on salaries increase to our people.
What we have to do, more or less the same to the increase of tariffs out in the market because of that. It landed quite well. It landed much better than last year. What we learned is that if you not only plan for the price increase and how to do it and introduce it technically, but also anticipate on the way you inform the media, you prepare the media for that, then it's very helpful. Not much of a negative impact there. We did it in the middle of COVID, by the way.
Yeah. Steve, to your first question. Sorry. Go ahead.
Yeah. I guess those are lessons that you will maintain in the business for the price rise next year. We should expect the communication to be similar, increased prices along with.
Yeah, no, absolutely. We aim for a price rise every year, as you know. Like I said, what I learned is that communication helps. We really hired one of the best people in the Netherlands. Although we are in communication, we always can do better on communication ourselves. That's what we did very well here. We learned from that, and we will continue to be better on that next year.
Okay, great. Thanks.
Steve, to your first question. If you think about the EBITDA today, and if you think of the delta EBITDA Q3 to Q3 2020 to Q3 2019, rounded numbers, think about B2C being about EUR 5 million off, the business segment being EUR 20 million off, and the rest is made up by wholesale and cost savings in the rest of the business. That gives you the EBITDA delta Q3 2019 to Q3 2020. That's kind of where it is order of magnitude. The EUR 20 million of business segment off, I think 50%-60% of that is COVID. The rest is underlying business pressure. That gives you a bit of a feel for the underlying dynamics of results. Now, you need to take it with a bit of a grain of salt because, of course, some of our TDO network business works for these segments.
These are not end-to-end numbers, but simply the EBITDA as reported internally, not with full cost allocation. To give you a feel for what the drag is of the business segment, between COVID and non-COVID part. Hope that helps you.
Okay. Just to sort of try and read between the lines, you're saying just over half the EUR 20 million quarterly in B2B is COVID. The rest is underlying. Would it be right to say that if you can stabilize that, there's a mid-tens of millions annual benefit to the EBITDA line? Is that a fair assessment?
If you stabilize the COVID portion, you mean?
Well, we assume the COVID-19 portion comes out or even reverses next year. Who knows what next year will look like? If we assume that genuinely is one-off, and just under half is underlying, and you can stabilize that bit, the underlying improvement is what I'm trying to say is, to get to stable would be a mid-tens of millions benefit to the EBITDA line. Is that fair?
Mathematically, you're correct. If you do that's right.
Yeah.
The hard work is stabilizing it. I'm not sure we can stabilize it over that easily. In an Excel world, you're correct, but that would be an afterthought after four quarters of COVID.
It's just that it's not happened yet. Yeah, I get the message. Great. Thank you.
Next question is from Mr. Joshua Mills, Exane. Go ahead, please.
Hi there. Thanks for taking the questions. Two from me. The first was just on the net promoter scores you showed us for B2C and B2B. Could you give us an idea of how the market NPS has developed over this period? I'm just trying to understand if this is a general frustration around COVID-19 issues, and that your position is still quite strong relative to peers, or if there's a more KPN specific issue here. The second question, just around your network strategy. We spent a lot of this call discussing your own rollouts, and in the past you've also done bolt-on acquisitions. Just be interested to hear your high level thoughts on network co-investment. Have you ever thought about doing a rollout alongside partners like DELTA Fiber?
If so, do you think that that would be allowed by the regulator or are there additional conditions to think about? Thanks very much.
Yeah. On the NPS, like I said in my introduction, it is a bit of an industry trend that the NPS is under pressure. Not surprisingly because suddenly from one day to the other, all people started to work from home. Asking for better support on whatever, TV, Wi-Fi, and also suddenly the full family, especially in the Netherlands, when the schools closed. We saw the traffic pattern changing after people started to work from home, and that was more or less the peak capacity on six o'clock in the evening. When the school closed, more or less, the peak went up, and we have a lot of capacity, so that wasn't a problem. That means that the whole family is depending on that one broadband connection. That's a lifeline of a household suddenly. That's when people start calling.
We have a product out in the market, very successful, twice sold out during COVID, called SuperWifi. It's mesh-based. You just plug it in, and you have much better Wi-Fi. Always people call for this kind of service, that's what we saw. We also see it as an industry trend. The thing is that we are on NPS, we were much higher than Ziggo, where we used to, years ago, be below zero, we increased to a super high level of 18, I think, beginning of this year. If you go high, you can fall deeper than the others as well. We see it coming down all service providers in the Netherlands and in the industry. Since we were far out the highest in the market, we are falling more points than the others. Yeah, it is an important target, a holy grail.
The bonus payment for all employees in KPN depends on the NPS score. We are obliged to really focus on that. Our network strategy, we did do every once in a while a co-invest deal, smaller ones. Usually, we move on a network, or we share rollout plans, and after a while we consolidate. That's done on a small scale. We did one last quarter. I forgot the name of that.
Friesland.
Friesland footprint, that's all small. We're always interested in those kind of opportunities when it works out in a positive way for KPN. If there's anything new on the horizon on that matter, we will of course mention it.
Yeah. I mean, Joshua, I think the regulator would not necessarily be against it as long as you continue to apply an open network policy on whoever you partner with. We've got an open network policy, so as long as whoever you partner with is open to that. I don't see any major regulatory objections.
Interesting point is that every now and then we have complaints about us pushing others aside when they try to roll out fiber. It always leads to questions from municipalities. If you share forces with another, that one is solved as well. It worked in the past, maybe it can work in the future as well.
Great. Just one very small follow-up. On the Net Promoter Scores, are you still on the consumer side ahead of [audio distortion] or have you dipped below them as a result of the NPS drop this quarter?
Yeah. We are on + 12 by head. I think they are around eight.
Okay.
Yeah, I always focus more on my own performance than Ziggo's. That's my problem probably, but I think we're above them still.
Very clear. Thank you.
Next question is from Mr. Ulrich Rathe of Jefferies. Go ahead, please.
Thank you. I have two questions, please. The first one would be the personnel cost reduction, excluding any divestment, was -15% in the third quarter, and it was -9% last quarter. much faster. I think during your prepared remarks, you talked about slowing down the hired work quite sharply. I was wondering, of that -15%, is it possible to get a sense of how much of that is sort of this step down in hired work due to the COVID-19 situation, and how much is a more sort of underlying sustainable rate? The second question is, at what point under Dutch rules would KPN have to disclose the approach of an interested party or the party that might want to take over KPN? What are the rules there?
At what point the situation has escalated to a point where you have to sort of announce it or talk about it? Thank you.
Yeah. Okay. Thank you. Yeah. On personal spend, 15% is not what we do on an annual level and not what we do year after year. That is pretty tough. A part has to do with COVID in the first place. When we realized that we had to close all shops and send all our service centers home, we stopped field engineers for a couple of weeks to work in customer premises. We realized that we had to do something. We kept on paying also all the hired hands, I think, for a month. That was much better than other companies did in the Netherlands. Then we took the action since we stopped the reorganizations to lay them off. By that, keep a very good eye on the cost side.
Costs related to people working, like travel expenses, we really stopped, and we did cut a lot there. Reduction we were planning to do anyway. By the way, what we also learned is that inefficiencies in our company are suddenly very visible due to COVID-19. After this pandemic crisis is over, we will benefit from that because we will never go back to the same cost level of hired people, travel expenses, office spend, et cetera. Part of the COVID reduction, we will probably keep for the years to come. Still, it was a good quarter on personnel spend, I would say. A little bit helped by COVID, in some kind of a funny way.
Yeah, I think if you look at on KPN, if you look at the staff developments, it was before the restructuring reorganizations have been delayed somewhat and are being picked up right now. Since then, we've installed a pretty drastic hiring freeze. Have a look at the number of new hires in our real new hires in the group, we're looking about five to eight a month. Which is quite little when you look at a 10,000 people staff level. Effectively installed a hiring freeze. With that, we can still shrink our full-time equivalent base somewhat because there's natural attrition that is ongoing. People are still retiring. Secondly, we've saved a lot on, as Joost said, on hired staff, external staff. We're restructuring our shops and the way we operate our shops, finding ways to cut back on cost.
I think we're going to continue to be low and reducing on FTE spend, but again, maybe not at the same magnitude and pace as this quarter. Your second question was, when we are obliged to publish information in case of approach by private equity. We are obliged to immediately publish price sensitive information in general, provided it's sufficiently concrete. The necessity to publish anything in case of an approach by a party will highly depend on the concreteness of any proposal. Of course, on our consideration on such a proposal. Whenever there is price sensitive information, we have to inform you then. There's no reason to publish anything on the rumor, I mean.
Makes sense. Thank you very much. Thank you.
Our next question is from Siyi He of Citi. Go ahead, please.
Hi. Good afternoon. Thank you for taking my questions. I have two, please. The first one is on your B2B top line. I understand that the business migration has been one of the biggest drags to your top line development over the past few years. Now, given that the majority of the heavy lifting of customer migration almost done, I wonder if you can help us to think about what's the next step for your B2B plan. Kind of how should we think about the trajectory going forward. My second question is on Huawei. I understand that you are in the process of replacing Ericsson with Huawei in your antenna network. Do you consider there could be a potential risk to your current mobile network strategy given what's happened in Europe in general towards Huawei equipment?
I wonder if you can just give us idea hypothetically, if Netherlands were to ban Huawei in both core and antenna network, how do you think of potential costs incurred and whether there is contingency plan built into the current CapEx budget? Thank you.
Well, on the first question on the B2B top line, some of the revenue decline has been heavily self-inflicted due to migrations. We've made quite some progress there, especially on the SME side. I think in the large corporate segment, some migrations are still to be done, and possibly a few may have revenue consequences. We're carefully weighing the timing of that. We've taken into account the lessons learned from the previous migration and seeing how to manage it and how to limit the revenue implications. We're not done there yet, but we've made quite some strong progress. What does it mean going forward? Well, if you look at our SME, our B2B business, SME is about one-third of revenues, but over 60%-70% of the EBITDA in that business. There most of the migrations are behind us. Our focus will be on cross-sell and up-sell.
Already we're seeing gradually clients taking on more than two products, and taking up our integral product proposition. Our focus will be on the SME segment, growing there through higher ARPU mobile, getting unlimited in the SME segment out, increasing our cross-sell. In the corporate segment, you'll see maybe some remaining migrations. There's some revenue implications, but in the larger corporate segment, we really focus on value, not on volume. Volume there immediately goes at the expense of your margins. That's kind of how we think about the business segment. We have some migrations to be done, and then focus on SME, on cross-sell, up-sell, increasing. It's like a more for more strategy. With regards to unlimited in our mobile base. In the larger corporate segment, continue to focus on value first and volume later.
Yeah, on Huawei. Let me first say we have a multi-vendor policy. We use, in our networks, Ericsson, Nokia, and also Huawei in some areas. We're in constant dialogue with our government on the topic of Chinese technology. We also look a lot at what's happening in other West European countries. We had a lot of contact with Deutsche Telekom, BT, et cetera, and what they are doing. We already announced our policy before our government did anything on this topic, when we said we're going to replace our critical systems for Western technology over time. In the life cycle of assets. That is a very important one. You're right. We are moving Huawei in our Radio Access Network. Last week we announced that we will build a new 5G core network with Ericsson.
We were w orking on a tender for a long time, because in the life cycle, we need to upgrade our core network in the coming two years. Huawei is in that domain as well presented, and we're going to phase them out and move Ericsson in. When it comes to critical domains, we think it's best to move in other suppliers. When we are in the process of upgrading or rebuilding an asset. On radio, yeah, we are constant dialogue with our government, and we think we move according the guidelines we get from our government.
That's very clear. Thank you.
Our next question is from Mr. Frederic Boulan of Bank of America. Go ahead please, sir.
Hi. Good afternoon, gentlemen. Two questions from my side. First of all, on the free cash flow outlook. If I understand, your 1 million fiber ambition for 2021 is unchanged. If anything, you might be a few quarters late. Beyond that, you're saying that the incremental rollout could be to a degree absorbed by reductions elsewhere. I'm trying to understand a bit what it means for your free cash flow. You've not reiterated the 2021 outlook today, which implies a step up in cash flow in 2021. Can you detail whether the broader framework is still valid or if there are other moving parts we should be aware of, working cap or elsewhere?
Second, just to follow up on the previous question on potential PE interest, whether you can just clarify for us whether you've had discussions with any potential partners and what you think would be key guarantees that a potential buyer would have to put on the table for an offer to be acceptable for KPN, the board, the foundation, et cetera? Thank you.
To start on the PE topic, I think, like I said, there were some rumors in the market, and I have nothing to add to that. We don't react on rumors. If there's anything we have to mention to the market, we will do that in a prudent way, but we don't react on rumors. That's for the PE topic. On free cash flow next year. We gave more than clear guidance for 2020. We're planning to provide you a regular strategy update on the 24th of November. Since we're entering 2021, and it's the final year of the plan we announced in 2018, and we have new management in place, and we think it's a good moment in time to inform you further on the 24th.
Like usual, that will include our strategic ambitions, not for 2021 only, but for the period of 2021-2023. We will give then a full integrated picture of our strategy. Without going into further details today, maybe Chris, you can give some background to our thinking.
Yeah, I think I mentioned, we will say whatever we do, we protect our progressive dividends. You may see some more allocation to fiber in different forms. See how we can do more, can we do faster? I mean, of course, we talked about the 1 million homes. There are other opportunities as well. For example, in business parks where you can invest in. There's a number of fiber initiatives that we'll undertake. Fiber CapEx will go up. We'll be scaling non-fiber CapEx down. Some of that will lead to more guidance for next year. You have to bear with us until the 24th of November, if you don't mind.
Thank you.
Okay. That's it. Thanks for joining us today on the Q3 call. If you have any further questions, please contact the IR team at KPN. Thank you.
Thank you.
Thank you.
Ladies and gentlemen, this concludes today's presentation. Thank you for your participation. I wish you all a very good day.