Update: Since this Q&A was done, Sprint has now launched three-channel carrier aggregation using 60MHz of 2.5GHz spectrum so it is now accurate to say they have 20MHz, 40MHz and 60MHz in use.

Sprint has been under a lot of pressure lately.

In addition to waging its own internal battles to get its finances and network under control, the carrier has also had to woo investors and analysts, asking them to believe in its vision of a network turnaround.

Now, after two years of work, Sprint finally has some results under its belt. The carrier has steadily improved its ratings from third-party sources like Nielsen and Root Metrics, and in July posted solid results for a second quarter in which it gained 173,000 postpaid phone net additions and cut churn to 1.49 percent.

To get the story behind the transformation, Wireless Week sat down with Sprint COO Gunther Ottendorfer. He told us all about Sprint’s vision around performance, network and spectrum, and why he’s never been more excited to be Team Yellow.

This interview has been edited for length and clarity.

Wireless Week: In the June quarter Sprint pulled in 173,000 postpaid phone net additions and posted record-low churn. Sprint CEO Marcelo Claure attributed this to Sprint’s improved network. Can you dive a bit into the improvements Sprint has made in the past year? What has Sprint done to improve its network performance? Which of these measures has had the most impact? How is Sprint planning to continue these performance improvements?

Gunther Ottendorfer: So, I think what’s important is we are really very happy with the results of the work of the last two years in improving our network. As a consequence of those, our network is currently performing at the best level ever. That’s not only seen by our own measurements, but it is seen by independent benchmarks. So we have significantly decreased the gap to the competition. And your question was how did we do that.

I think what we have done is we have done quite a lot of work on the basics of the network. So a lot of optimization, a lot of daily monitoring, improving the parameters of the network. That is one piece of it, a lot of groundwork fixing the basics.

Beside that, we have rolled out our 2.5 spectrum band, and with the roll out of the 2.5 spectrum band we gained a lot of capacity and reliability for our network. All of that started from November of last year under the brand of LTE Plus. It’s an LTE Network but it has a Plus against conventional LTE networks. Our Plus is that we introduced LTE Advanced features, and the two features we are using on our LTE Plus networks are carrier aggregation and antenna beamforming. Both help our customers because carrier aggregation brings us higher throughput speeds but also better reliability, and the antenna beamforming is improving the performance on the cell edge, is improving individual user performance by the beamforming features.

So, LTE Plus is the main driver for those improvements, besides fixing the basics that our CTO has done very successfully over the last two years.

Looking into the future, we will continue to do that. We will roll out more LTE Plus markets, we will also continue with our densification and optimization strategy. We will densify our network and by that bring more capacity and more reliability to our customers.

WW: On T-Mobile’s second quarter earnings call, of course John Legere had something to say about your densification plan. He was kind of saying that you’re rushing through it, but maybe don’t have the financials. Do you have any concerns around the financials of being able to execute on this strategy?

Ottendorfer: Not at all. As [CFO Tarek Robbiati] has elaborated in our earnings call, with the financing steps he’s taking we have the liberty to execute on our plan and that’s exactly what we are doing.

WW: You guys are sitting out the spectrum auction. You’ve said about a quarter of your 2.5 GHz spectrum has been deployed. Can you talk a bit more about your spectrum position? How and why is 2.5 GHz important for Sprint going forward, and what kind of pacing are you looking at for lighting that up?

Ottendorfer: Here again, the performance you see on our network, a big part of it is driven by our use of the 2.5 GHz spectrum, which means in turn we could achieve those improvements because of that spectrum. If we can in the future use even more of that spectrum, we will improve further.

The spectrum itself is the lifeblood of wireless. In wireless the more spectrum you have, the easier and more efficient you can build your network. We have more than 160 MHz of 2.5 GHz spectrum in the top 100 U.S. markets, and we can grow into that spectrum. We use currently 20 or 40 MHz of it. So as Tarek rightfully said, we have lots of room to grow. That is the big difference to the other carriers. We can on the same site activate more spectrum and by that significantly improve performance and capacity for our customers.

The other thing that is important to keep in mind is 2.5 GHz is the biggest LTE ecosystem because it’s not only in use by Sprint, it’s also used by other big carriers around the world. It has the biggest potential as high band spectrum for high capacity networks.

Historically, networks have been built in medium and low band spectrum because they were built for voice services. When you want to build a high throughput data network, it needs to be built around high band spectrum. And that is why 2.5 GHz is so important, because it helps us to build a network structure that is already future proof for 5G.

As I said, with the 2.5 GHz currently we have 20 or 40 MHz in use and we can grow and activate more of that spectrum without spending a lot of money.

WW: And what sort of pacing are you looking at for rolling that out? I know you’re trying to get as much out as possible as quickly as possible but is there any sort of solid or fluid timeline for that?

Ottendorfer: We update the market regularly when we achieve something, but what we don’t do is hand out the playbook because we also adapt to what we see, how customers use your network. We will look at where they are using it, how they are using it, is the demand in the places we forecasted it or is it in different places, and then we can adjust our roll out.

WW: During the earnings call, Tarek mentioned something called “small sites.” Can you speak more to those, what they’re made up of, what kind of range they have, or how they stack up against either macro sites or small cells?

Ottendorfer: We have a small cell strategy, the densification is based on small cells. But we have other tools, like building cell sites that are not traditional high-cost macro structures, but are like pole attachments. So using infrastructure that is already there and activating that in an efficient way.

We have extended the toolbox of our network to include solutions like that so that we can roll out more efficient but also faster, because it’s faster to go on existing infrastructure than to build infrastructure from scratch.

WW: How big of a roll does that play in your densification plan or in your overarching network vision?

Ottendorfer: Well, when we see that customers have growing demand in an area, we look at what is the best solution to implement with two criteria: speed and capital efficiency. We decide based on that should we go for a small cell, do we need several small cells, can you build a poll attachment on existing infrastructure, can you go for a rooftop solution or do you go to an existing tower or build a new tower. And that is how we look at it for every single case.

WW: Sprint has caught a lot of criticism both other industry players and concerned analysts around CapEx. Obviously, these non-traditional infrastructure setups are costing you a lot less, but $300 million still seems low by comparison. So how are you guys looking at CapEx?

Ottendorfer: Look, first of all, we also got a lot of credit for delivering good performance improvements with the strategy we have used so far. I think many people have come around and said ‘now we can see that your strategy is working’ and that the activation of the 2.5 GHz and using the 2.5 GHz ecosystem is going much faster and much better than we ever anticipated. So we have a growing understanding in the market that what we are doing makes a lot of sense.

As part of the densification and optimization strategy, we are using that infrastructure and we are using our spectrum assets. That gives us the ability to use less CapEx than the other because it’s a lot cheaper to activate another spectrum carrier on a tower than to build a new tower. So in many cases we don’t even need a new card, it’s just a software change.

WW: Does Sprint have any plans alter its CapEx going forward? Are there maybe plans to ramp in ‘17 or ‘18, or are you going to play it by ear?

Ottendorfer: Tarek answered a little bit, but basically the quote was that we will spend more in the next quarters.

WW:  During second quarter earnings, Sprint CTO John Saw was talking about the difference between two- and three-carrier aggregation. Can you speak more to that? How much is already on two-carrier aggregation, how quickly are you ramping up to three and how is that impacting both network performance and network perception?

Ottendorfer: As I said before, LTE Plus really helps us with customer perception because it’s great for customers that have carrier aggregation-capable phones. In the last quarter, three out of four phones sold were carrier aggregation capable and I think more than 20 models that we are selling have carrier aggregation built in. So we have a big base of customers that can use it. We have rolled it out to these 237 markets currently and we will continue to roll it out further.

Our plan is to make our network ready for three-carrier aggregation and to come out with that in the next year. We believe that will bring another push for people that are looking for really good and reliable data throughput and also need high peak data speeds. We also know that we are the only carrier that is capable of doing that because the others do not have that much contiguous spectrum to do it.

In the top 100 markets we have 160 MHz of 2.5 GHz spectrum. In the beginning we used 20 MHz of it, 20 MHz is kind of the basic unit in network roll outs. Then we went to LTE Plus and we’re using 40 MHz – that’s why Tarek said we have activated a quarter of our spectrum in the 2.5 GHz range. When we activate three carrier aggregation we will activate a third carrier on 2.5 GHz, so it will be 60 MHz of 2.5 GHz spectrum. And that’s why we can be so capital efficient, because we don’t need to build another site, we don’t need to build another spectrum frequency or technology into it, we just activate another 20 MHz of the spectrum we already have on that site.

WW: AT&T has talked about its plans to virtualize 30 percent of its network by the end of this year. Does Sprint have any NFV/SDN plans in the works to virtualize your own network? If so, what are your target numbers and how would it help you?

Ottendorfer: So this is one of my favorite topics, you have to know. I’m a big believer that network function virtualization is an important paradigm shift for our industry. And we are working on it.

At Sprint we are testing currently different solutions in lab environments, that’s one important thing we’re doing. The other important thing is that I’ve started NFV Open Doors for all employees and we have created training courses for all employees. That’s because what I want to do is say NFV is a journey for telecom operators, I invite all employees in the technical field to join that journey, to get educated about it, and I want to create an open mindset with our technicians for NFV solutions. That’s really important for the future because if people have the right mindset, we will in the future have a much easier way of introducing NFV solutions into our network.

I expect from NFV several benefits. One is greater flexibility and shorter time to market because it will be easier to launch services on the NFV platform than it is currently. I expect it to be more cost efficient because it will have better utilization of those platforms, and it’s also opening up the door to more innovation because it’s easier to bring innovative services to the market when you have an easy environment to try them out and to integrate them into your telecommunications infrastructure.

WW: So you said Sprint has solutions in lab testing. Is this something you guys have in the works for maybe 2017 or 2018, or is it at a point where you can’t really say?

Ottendorfer: Well, look, it will be in the next years and it will be in the years you mentioned, but I don’t want to box myself in. I’d like to come forward with more about it when we have more to show practically.

WW: Fair enough. Let’s turn to 5G. Sprint has been doing some work with Nokia and Ericsson with some tests at the Copa America soccer tournament. You’ve focused a lot on 2.5 GHz but we haven’t hear much from Sprint in regard to some of the higher frequency bands the other carriers are looking at. Can you speak to where Sprint is on 5G and what you’re looking at there?

Ottendorfer: At the Copa America we were there demonstrating on 15 GHz and on 73 GHz with Ericsson and Nokia with download speeds of more than 2 gbps and more than 4 gbps. So that’s great.

We are the carrier in the U.S. who has the most experience working on the high band, on one hand because of the 2.5 GHz. I always say the 2.5 GHz band will be the low band of 5G. So in the world of 4G, 2.5 GHz is considered the high band. In the world of 5G, 2.5 GHz will actually be the low band because all other bands will be higher. So we have a lot of experience in the future low band.

But we also have a lot of experience in other high band spectrum because due to our history of Clearwire and Sprint having big microwave networks, we have thousands of cell sites currently connected in the 11, 18, 23, 28 and 80 GHz range. Especially in New York, we have quite a lot of links in the 80 GHz range.

So we have a lot of practical experience because we use these links everyday. We know how to operate them, how to plan them that they are not influenced by adverse weather conditions. So we are the carrier with the most experience with the high band, and we will continue to use that knowledge and build on that on the journey to 5G.

The trials at Copa America were the first step for us, but there are other steps that will follow.

We have a good foundation with the 2.5 GHz band, and for the rest we are waiting on the next steps from the FCC.

WW: Excellent. Any final thoughts?

Ottendorfer: I believe with our strategy we are pioneers. And as pioneers you try concepts that haven’t been proved in the past, so you have many people that are skeptical. I think what makes me excited again is that people in the industry are seeing our strategy working on the 2.5 GHz and that the densification, building the network is really working.

We’ve built a good foundation with our 2.5 GHz, but we’ve got plenty of room to grow. So I’m a big believe in the phrase ‘you ain’t seen nothing yet.’