Diebold Nixdorf Inc (DBD) 2017 Q2 法說會逐字稿

Good day, everyone. Welcome to the Diebold Nixdorf Q2 2017 Financial Results Conference Call. At this time, for opening remarks and introductions, I would like to turn the conference over to today's host, Mr. Steve Virostek, Vice President of Investor Relations. Please go ahead, sir.

Thank you, Allen, and welcome to Diebold Nixdorf's second quarter earnings call for 2017.

Joining me today are Andy Mattes, President and CEO; and Chris Chapman, Senior Vice President and Chief Financial Officer.

Per our custom, today's webcast is being recorded, and a replay will be made available later this afternoon. For your benefit, we've posted presentation slides to accompany our discussion on the Investor Relations page of dieboldnixdorf.com.

Slide 2 is a reminder that we'll be referencing certain non-GAAP and pro forma financial information, which we believe are helpful indicators of the company's performance. We've reconciled these metrics to their respective and most directly comparable GAAP metrics in our supplemental schedules of today's earnings release and the slides.

On Slide 3, we remind everyone that certain comments may be characterized as forward-looking statements, and that there are a number of factors that could cause actual results to differ materially from these statements. You may find additional information on these factors in the company's SEC filings, including our Form 10-Q, which will be filed next week. As usual, this forward-looking information is current as of today, and subsequent events may render this information out of date.

And now, I'll hand the call over to Andy.

Thanks, Steve. Good morning, everyone, and thank you for joining us this morning. Two weeks ago, we made significant revisions to our 2017 outlook. I am personally very disappointed, and I know the management team feels the same.

Given that our books for the quarter were not yet closed, we were very limited in the amount of information we could share with you. Today, we will review our second quarter results, walk you through the elements that led to our recent revision and outline the actions we're taking.

Looking at market demand in the second quarter, we delivered 4% sequential order growth. Our book-to-bill ratio was greater than 1 in all 3 regions. The Americas region provided the strongest order growth and book-to-bill ratio due to large project sales for branch automation. On a global basis, we booked another strong quarter of recycling orders, with 6 large contracts for approximately 6,000 recycling units across all regions.

Our backlog grew 6% sequentially to approximately $1.2 billion. Now let me take you through the main factors that led to our revised outlook, and Chris will provide additional detail by segment.

The biggest change is to our top line, where we see full year revenue coming in about $300 million less than previously anticipated. The weakness is predominantly tied to our banking hardware volume, which also has a direct negative correlation to our installation services business. And while the recent order activity demonstrates the market acceptance of our new solutions and the competitive advantage we bring to our customers, the pace of orders to date is still short of our expectations.

Additional contributing factors are for: One, prolonged installation schedules causing delayed backlog conversions; and secondly, earlier than expected contract runoffs from legacy Wincor base in the U.S. and a few niche IT service contracts in Europe.

We also reduced our profit outlook due to the impact of the lower revenue and higher costs in our service business due to a complex timing challenge. This includes lower utilization of service techs due to the accelerated runoffs in the above-mentioned contracts. This was coupled with our strategic investments to maintain our competitive advantage in break-fix services and grow our market -- our managed services and ATM-as-a-Service business.

Also, as the NextGen multifunction Diebold series reached critical mass in the Americas, we applied incremental resources, invested in additional spare parts and technical training to retain high customer satisfaction levels. In addition, like other North American companies in the IT services industry, we're experiencing higher wage inflation and attrition rates among service technicians.

Now shifting to our action plans, we are taking aggressive steps to solidify the top line and adjust our cost structure in the near-term revenue realities. We're boosting our sales transformation and sales excellence programs by upscaling our teams on the broader portfolio of services and software for banking and retail customers. We are investing in additional sales training and new hires for software and project management. For example, as part of our service growth activities, we launched targeted sales initiatives to ramp up service attach rates on the installed base of ATMs. We've had some early wins, adding about 5,000 ATMs under contract today.

We're also gearing up our ATM-as-a-Service, our sales checkout and our retail store life cycle management programs in key markets.

With respect to software, we're seeing increased opportunities to sell software into our installed base of customers. As an example, we had more than a dozen wins at U.S. regional banks in Q2. We also formed a partnership with Kony, the leader in mobile application software, as recognized Gartner and Forrester. Consequently, we're making a concentrated push to train our sales force to sell cloud-based mobile software platform, software toolkits and apps to current customers in every region.

We are enhancing our product excellence program, which, amongst others, consists of a holistic quality initiative across all lines of businesses, which will ultimately improve performance for our customers and lower cost of both our systems and services portfolio.

With respect to the product portfolio, we will have streamlined the number of terminals from 96 at the beginning of our combination to less than half by the end of this year. In the same time frame, we will have also renegotiated about 70% of our direct procurement spend, and we're beginning to benefit from volume pricing on direct materials as well as optimization of freight costs. These actions have resulted in an improvement in systems gross margin.

With respect to our overall integration, we have completed or are in the process of completing, several actions which should benefit our P&L by the end of the year. For example, we have taken all necessary actions to close our legacy Diebold plant in Hungary and our distribution logistics center in The Netherlands, and we expect to complete these closures by year-end.

We are well on our way to reducing redundant stocking locations globally. As of today, we have closed some 200 locations and expect to close another 100 by the end of the year.

We have also made good progress on integrating the field service organization, which includes unifying the legacy IT systems and logistics support.

With our work nearly complete in about 70% of countries, we expect to drive increased operating efficiencies through fewer calls and faster fix rates. We're also making greater use of lower-cost call centers, both offshore and nearshore, to improve both the cost and performance of our global delivery infrastructure.

On the organizational front, we will continue to remove management layers, reduce headcount and improve our decision-making processes. We've reduced our global headcount by more than 600 FTEs, and we expect another 300 employees will exit the business by year-end.

Finally, we continue to review our portfolio of businesses to reduce complexity and emphasize our more profitable entities. During the quarter, we completed the sale of our Electronic Security business in Mexico and reached an agreement to divest our legacy Diebold business in the U.K., opening the door for us to pursue major customers in this country with the full breadth of our portfolio.

Our DN2020 initiatives continue to point to significant scale benefits for the new company. As a result of the increased visibility on our cost structure, we now expect net savings of $240 million by the end of 2020.

And now, Chris will provide additional detail on our financial performance.

Thanks, Andy, and good morning, everyone. My comments today will focus on our non-GAAP results from continuing operations, unless otherwise noted. In addition, we are providing select pro forma information for the year-ago period to help facilitate more meaningful comparisons. I will first comment on the results for the quarter, then provide additional details on our recent change in the full year outlook.

In addition, given the accelerated timing of our earnings release this quarter, we are changing our practice of filing the 10-Q the day of earnings, and expect to file next week.

Starting on Slide 6. We compare total revenue for the second quarter 2017 with pro forma revenue from the year-ago period. On a constant currency basis, revenue decreased 11%, primarily due to our systems and services lines of business for our banking solution.

Looking at the mix of revenue on a GAAP basis, our mix was largely unchanged for the first quarter -- from the first quarter. Services and software accounted for 61% of the business, while systems accounted for 39%. Our geographic mix of revenue was 52% in EMEA, 34% in the Americas and 14% from Asia Pacific. With respect to our solutions, banking accounted for 74% of total revenue, while retail was 26%.

Moving to Slide 7. We compare key non-GAAP profit metrics for the quarter with pro forma results from the prior year. The $56 million change in gross profit is primarily due to the systems and services businesses, which I will discuss in greater detail later in my comments.

Operating expense is down $11 million, compared to the pro forma 2016 results, showing the benefit of our integration efforts and our overall DN2020 program. Operating profit was down approximately $45 million, with operating margin of 3.5%, reflecting the impact of the lower year-over-year volume, partially offset by the reduction in our operating expense. Our adjusted EBITDA of $74 million was down approximately $47 million from the prior year pro forma results, reflecting the lower operating profit performance.

Turning to Slide 8. Non-GAAP services revenue decreased 7% in constant currency on a pro forma basis, with growth in retail more than offset by a decline in banking. The revenue decline is due to lower product-related installation activity as well as lower contract services volume, both in the Americas and in EMEA, tied to multi-vendor and IT contracts that were not renewed. The lower volume, combined with the previously mentioned service investments, resulted in a gross margin decline of approximately 240 basis points year-over-year.

Looking at Slide 9. Total systems revenue decreased 16% in constant currency compared to the prior year pro forma period. Looking at the performance in more detail, banking systems revenue was down 22%, with the declines mainly in EMEA and the Americas largely related to completion of large projects in the prior year period.

Partially offsetting the declines in banking, retail systems was up slightly year-over-year, coming off of a very strong prior year performance, with solid results in EMEA and Asia-Pacific.

Systems gross margin decreased 150 basis points to 20.2% compared to previous year pro forma. The change in systems gross margin is largely a reflection of the volume decline and customer mix, partially offset by the initial benefits of our integration efforts, as outlined in our DN2020 program. On a sequential basis, our systems margins increased for the third consecutive quarter as a result of our continued focus on deal quality and procurement initiatives.

Turning to Slide 10. Our software line of business delivered revenue, $110 million, which was a decrease of 8% in constant currency compared to prior year pro forma. The year-over-year change is primarily the result of a tough comparison as we completed a large banking project in the Americas in the prior year period, partially offset by improvements in retail software solutions. Software gross margin increased slightly year-over-year.

Moving to Slide 11. Non-GAAP EPS was $0.08 for the quarter, which is a reduction of $0.35 compared to the previous year, reflecting the impact of the lower operating profit and higher year-over-year interest expense. As a reminder, the deal-related interest expense was excluded from the non-GAAP results in the prior year, pending approval of the combination with Nixdorf.

The non-GAAP EPS for the current period excludes restructuring expense of $0.19 and nonroutine expense of $0.78. The nonroutine expense consists of $0.29 of acquisition integration expense, Nixdorf purchase price accounting adjustments of $0.56 and other net nonroutine benefits of $0.07, mainly related to a gain tied to the divestiture in the U.K. The impact for restructuring and nonroutine items, inclusive of allocation of discrete tax impacts, was $0.48.

During the quarter, the non-GAAP effective tax rate was a benefit of 2.7%, with a year-to-date non-GAAP effective tax rate of approximately 20%. The non-GAAP tax rate in the second quarter largely reflects the year-to-date adjustment for lower forecast performance across several high-tax countries. Items impacting the company's non-GAAP earnings arise from a number of different tax jurisdictions, and the final amount incurred can result in variability in the non-GAAP tax rate. As a result, we continue to hold our full year non-GAAP rate at approximately 30%.

On Slide 12, free cash use was approximately $134 million in the second quarter, reflecting an increase in use of $37 million. The increase is primarily the result of integration and deal-related expense payouts in the current period, which included an obligation of approximately $20 million in stock compensation for employees of legacy Nixdorf. This largely brings the deal-related expense items for the Nixdorf transaction to a close.

In addition, interest expense payments were also higher by approximately $20 million versus the prior year. On a year-to-date comparable basis, free cash use is essentially flat.

On the right side of this slide, we provide highlights of our liquidity and net debt position. As of June 30, we reported cash on hand of $528 million and gross debt of $1.9 billion, resulting in a net debt of approximately $1.4 billion.

The previously announced Term B repricing became effective on May 9 and will reduce the interest expense by approximately $5 million per quarter, with the full benefit starting in Q3.

To provide a perspective on our net leverage ratio, if you looked at the trailing 12 months pro forma adjusted EBITDA based on publicly available information, the net debt to adjusted EBITDA is at approximately 4x.

Continuing on Slide 13, I will further detail the primary drivers behind our adjusted outlook for 2017. We have included a revenue and adjusted EBITDA walk to highlight the major movements that Andy previously summarized.

Starting with revenue, approximately $225 million of the change in our expectation is systems-related revenue, which is inclusive of the software license and installation services tied to our hardware sales. From a solutions standpoint, this was roughly a 70-30 split between the banking and retail businesses, respectively. The change in banking is coming from both the Americas and EMEA, whereas the revision in the retail outlook is primarily from the Americas.

Additionally, there was a change of $75 million in expectations for services revenue, largely in the Americas and EMEA banking segments, where we were unable to secure new contracts to offset the contract runoffs that Andy previously mentioned.

From an adjusted EBITDA view, the profit associated with the systems volume reduction is approximately $50 million. This is a fairly high fallout rate, given that changes in revenue came from our higher-margin businesses in the developed markets. The profit on the change in the service business is impacted by the reduced volume at a blended service margin rate of around 40%, combined with the additional investments and higher labor costs impacting service cost of sales.

Moving to our outlook on Slide 14. Full year revenue is expected to be in the range of $4.7 billion to $4.8 billion, with the anticipated currency impact of less than 1% on a full year basis.

The company expects a net loss of $110 million to $125 million on a GAAP basis for the year. Adjusted EBITDA is expected to be in the range of $360 million to $380 million. This assumes realization of around $50 million of cost synergies, includes depreciation and amortization expense of approximately $110 million and share-based compensation of approximately $30 million.

On a non-GAAP basis, we expect EPS in the range of $0.95 to $1.15 and a non-GAAP effective tax rate of approximately 30% for the year.

Looking at the remainder of the year, improving on our execution will be key to meeting our second half expectations. Based on our current scheduled backlog and timing of our cost synergies, we expect a modest sequential increase in both revenue and adjusted EBITDA in the third quarter and a significant ramp in revenue and profit in the fourth quarter.

Our free cash flow outlook for the year is now expected to be near breakeven due to reduced earnings outlook. This also could be influenced by the timing of certain integration and restructuring costs, which we estimate to be around $100 million for the year. In addition, we expect capital expenditures of approximately $90 million.

With that, I will open up the call for questions.

(Operator Instructions) And we will take our first question from Paul Condra with Crédit Suisse.

I guess, just for the first question, I just wondered if you could kind of revisit the DN2020 targets from the Investor Day, the $3.50 in EPS by 2020?

Yes, let me first go through the overall change in the DN2020 cost targets. So we had outlined approximately $200 million at that time, and we've increased it to $240 million. You can hear from the expectations in '17, we're looking to run at about an additional $10 million through. And if you think about it from a sequential standpoint, we're looking at pulling some things in from the 2020 standpoint into '19 and accelerating some of the activities where we've seen very good progress. If you break it down by the major categories that we highlighted previously, I'll just take you through the 3 main buckets here. We'd indicated service cost of sales benefits were roughly $50 million. We're increasing that to $65 million. On the system side, we previously had approximately $60 million, and we're increasing that to $70 million. And then on the overall op expense side, we were at $90 million, and we're increasing that to around $105 million. So that's the rough breakdown of what we see. With regards to the $3.50, that remains our outlook for the long term, and we are looking at pursuing all appropriate pathways to achieve that over the next several years.

Okay. And I just wondered, I mean, in terms of the conversion delays, should we think of that more as kind of the new status quo? So when you were looking at these -- looking at your pipeline maybe 6 months ago, now when you're looking at it, you kind of think even things that are a couple of years out are going to take longer to convert to revenue. I'm just kind of trying to think about, when we look at 2018, whether we should think of some of that revenue that you're expecting then should also be pushed out as well?

Paul, from a market point of view, I will consider this the new normal. The market has completely swung towards a big project market. The market has completely swung from what used to be, 3, 4 years ago, a market that was aided and driven by BRIC, to a market that is predominantly focused on the developed markets. Now the good news is, these are higher-value machines. Those have much higher software attach rates, these markets have much higher service attach rates. At the end of the day, they're more margin-accretive. The flip side thereof is within the developed markets, financial institutions that are buying are the big institutions, basically, the top 100 banks, and those turn into bigger projects, with longer installation cycles, but also longer decision cycles inside the bank. As the projects get bigger, more levels in the organization have to approve, so we would expect these longer project-related cycles to be the new norm, and we're adjusting our thinking and our planning accordingly.

So I guess, just lastly, I mean, just on the deal slippage, can you talk about your confidence level then around those deals closing in the second half and being similar in size and value to what you thought they would be originally?

Let me just take through, there's -- we have 3 elements. Yes, we wanted to sell even more than what we sold in Q2, but that's the smaller of the buckets. The more important one is, of the deals that we closed, some of the deals were in the works for months and months. By the time we got down to not just the frame agreement, but to the installation schedules, and those came in very late in the quarter, the installation of these deals pushed out. So deals we had in the funnel, where we had anticipated to have a rollout in the second half, are now beginning rollout after the blackout time, starting February next year. And then the third bucket is that within our existing backlog, we had a few projects where, for a multitude of reasons, the financial institution asked us to push the deployment into the next year. And when you bundle all of that up, that's the deviation from the assumption that we had 3 months ago.

Okay. Yes, I just wanted to clarify. It didn't sound like any deals that had -- significant deals that has been lost or that now were in question?

We've lost some deals, especially in Asia, to local competitors, but that's also part of our conscious decision to not chase revenue for revenue's sake. We're very much focused on deals that are margin accretive, and some of the Asian markets have gone into questionable territory when you look at the profitability, and that's where we didn't go full throttle forward.

And we will take our next question from Paul Coster with JPMorgan.

A few quick ones. So it sounds like you've lost a few service contracts that ran off. How many in total, and is there a common theme to why you lost them? Or is that already encapsulated in your comment around the APAC move to local competitors?

No, Paul. No, this is Americas and Europe. We're talking a handful, and we can be very precise on those. The legacy Wincor installed base that we took over, especially in North America, that's the base that ran off. And we knew it would run off eventually, but that has accelerated. And then we had some IT outsourcing contracts in Europe, basically on the fringe, they weren't in our core portfolio, that either came to a completion or ran off. So those are the 2 main issues. To give you a data point, the legacy Diebold installed base in North America has been rock-solid, around 90,000 units, for the last 3 to 4 years. And conversely, the legacy Wincor installed base in Germany has been rock-solid, around 40,000 units for the last 3 years. So the main contributors of our service and our service profitability are very much intact.

I guess, in the projects run -- insofar as these service contracts ran off, I anticipated that you'd step up and rebid on them and have a higher win rate on the rebids, but have you just chosen to let it go or is it an actual competitive loss?

Well, look, Paul, if you look back at what happened in 2016, and we've now got enough data, we have to concede that we lost a couple of points of flow share in 2016. And when you say where did that happen, it very clearly points to the weaker of the 2 parties in the time between deal announcement and deal closure, had market share loss, i.e., Wincor in North America and one financial institution and one of our competitors has also made this very public, that they won that deal last year, and we knew that, we had admitted that at previous calls, and vice versa, legacy Diebold on -- in the European theater of operation. So it's -- we knew it will happen.

(inaudible) Andy, I'm okay with -- then that makes great sense to me. What doesn't make such great sense to me, though, is that it should therefore be something of a surprise that the service margins would get impacted by the lower utilization rate when you knew that was coming. Or is it a completely unrelated issue?

So let me take you through the service margin issue. And it's through the U.S. because that's predominantly where this hit us, and it's down to 3 elements that are driving that. Very simple, a tech is not a tech, meaning, once -- when you combine portfolios, not every tech is trained on every technology. But if you think back 2 years ago, our Diebold techs had to know one product, which is our Opteva product line. We then took over 2 products from the Nixdorf side and we rolled out 2 products in the NextGen Diebold series, which means we went from 1 to 5, which meant we had to cross-train, which meant we had people in training, which lowered their utilization. But more importantly, people are in different local geographies. So let's say we have a runoff on a legacy Nixdorf product in California, and we had growth with new contracts on the East Coast or in Canada. I still need the guy in California because he has the Wincor Nixdorf domain expertise on the product. I cannot let that person go, but at the same token, he is underutilized, while at the same point, I have to staff up in other parts of the country. And that's when the scheduling gets so important. As long as schedules are rock-solid, you can plan for that. The minute that the customer accelerates one and slows down the other, our scheduling gets out of sync, and you have underutilization and margin pressure.

That helps. Okay, one last question then, please. The new guidance, to what extent would you argue it's de-risked the outlook now? Perhaps you can give us some sense on the puts and takes in either direction at this point.

Yes, I guess I would start just with the learnings that we've had here in the first 6 months of 2017. And I think it's very fair to say that we underestimated the amount of distractions tied to the integration, the amount of change that we introduced into the organization and also the size of the hole that we dug ourselves as we exited '16 and came into '17, with the order activity in that second half of '16. And so as we looked at that second half of the year and we had to acknowledge some of these distractions, we had to acknowledge our lack of execution, frankly, that we've had, and so we've tried to take a very prudent view of the second half. If I think about this across the 2 quarters, the third quarter, we have roughly, on a full revenue basis, roughly 10% sell-in revenue as of July 1 that we'd need in the overall quarter. So we've taken a very thoughtful view there. Then if you think about that, also then from a fourth quarter standpoint, it's roughly 25%, 30% of sell-in revenue for orders that we have to secure here in the quarter. So we -- execution is still key, and we have to improve there, but I think we've taken a fairly prudent approach on the second half outlook.

And we will take our next question from Matt Summerville with Alembic Global Advisors.

Chris, just a follow-up on the last question. Typically at this point in the year, how much do you still have to sell for Q3? And how much do you still have to sell for Q4? I guess, after this sort of massive reduction you've had in your top line forecast, I guess at the end of the day, I'm just trying to figure out why we should believe this latest forecast, which still implies yet again a big ramp in fourth quarter revenue, if you're going to even come close to the high end of your revenue range?

Yes, I would say for the first part of the question, Matt, the sell-in revenue expectations that we have are very much within our norm of what I would expect from the combined company right now. The other piece that you have to keep in mind when you think about that fourth quarter ramp as well, as we start to get the compounding effect and the benefit of our cost takeout of our synergies. So we're going to see a little bit more of that benefit. And again, when you think about the overall range, the banking activity and the banking activity and the overall order intake that we get here in Q3, is going to determine whether or not we're at the upper or lower band of that. And the last piece that I would just highlight, when you think about Q3, we've also tried to factor in the normal European vacation or holiday schedule impact that you typically see, a little bit of that lull from the overall August impact. So we try to factor all of these items in to come up with a prudent set of numbers for the second half.

And then, if possible, I'd like both of you to address this. When I look at kind of where the stock is, and if we think about a company that still has the potential to earn $3.50, why have we not seen management step in and buy shares, since this thing has been announced? I know there's blackout periods from time to time, but I've still been very surprised in particular that neither one of you have come into the market. And can we expect after this big drop in share price that we've seen, your behavior to change?

Matt, first and foremost, we were absolutely in blackout, and there's no way that we were allowed to do anything. And as you can imagine, our lawyers have us on a very short leash when it comes to what's material information and what's insider information. Second, let me just point out the majority of the top management's personal net worth is tied to the Diebold stock. And also, let me point out, the majority of our long-term compensation, basically, 80% of our long-term equity compensation, is tied to the stock and the metric here is the TSR, or the option price. So we are definitely in this thing, and we're hurting, just like any other investor, personally big time from the drop of the share price, and have every motivation in the world to do right for shareholders.

And then just lastly, when should we expect service gross margins to trough? Or should we be thinking about the second quarter as being at that trough rate, given cost savings, given utilization rates, all the sort of things that you've touched on throughout the call?

Yes, I would say that if you think about it, you're hitting some of that trough now. As I just highlighted, you get a little bit of that seasonal effect of utilization in Europe typically in Q3, and then you start to get that higher utilization as we start hitting that higher ramp on the overall system side. So that combination ultimately impacts it. And then also, as we move forward, when we start to realize some of our integration savings, there should be some step change benefits that we get in future quarters as well.

And we will take our next question from Justin Bergner with Gabelli & Company.

A couple of questions here. I'll try and run through them quickly. On the service contracts not being renewed, when you talk about accelerated runoff, could you maybe help me understand what that means, did the contracts actually end prematurely?

These were systems that the customer took out of commission and replaced them with a different product that was not under our service agreement. And any service contract that you have, they have a bandwidth in which the customers are allowed to do that. And depending on the age of the system, some of the contracts had shorter duration times, so the customers were very well within their rights to do so. They did press it to the bottom end of the commitment that they had towards Diebold. And as I said, they did that ahead of the schedule they had previously communicated with us.

Okay. If we sort of rewind a year or 2, multi-vendor service agreements were sort of a big revenue push for Diebold and you were making progress on those. I mean, has that momentum stalled? Or have any of these service issues affected that? Or is that mainly been on the margin side? Do you think you can continue to use multi-vendor service agreements as a source of good revenue growth for the company?

We will definitely continue to use multi-vendor services as a growth opportunity for the company. However, what we did notice is the stickiness or you can say the loyalty that these customers have with us is vastly different than the customers who've made a long-term commitment into Diebold Nixdorf technology. And on hindsight, we probably weren't nuanced enough in our anticipation on how you run these numbers, because if you look at where the runoff came from, it's all in the multi-vendor space, because when we started this project, Wincor was still a competitor.

Okay, I appreciate the honest clarity there. Just finally on cash flow. I think the guide coming out at the end of the year was for $50 million of free cash flow. It seems like EBITDA tax -- if you tax effect that, that's down by $60 million, CapEx is down by $10 million, restructuring is up by $50 million. That would sort of take me from positive $50 million to negative $50 million. Is my math off? Or how do you get sort of closer to breakeven free cash flow?

There's 2 things there. Number one, on the working capital side, we've actually had some very nice progress internally, improving the overall metrics, looking at working capital as a percent of revenue, down another 2% in the quarter. We were also down in Q1. So there's still significant opportunity there. And with the down revenue and profit, obviously, we're going to look to wring out of working capital, that's one, to get some offsets there. Number two, when you look at the overall restructuring, the timing of the actual expensing that through the income statement versus paying that out of the cash flow, that's a variable. Right now, I see some of that payment coming more in the '18 time frame, but it is in our best interest as a company to accelerate and get that out. So some of that's outside of our control. We've got to work with some other constituents, works councils and other things to get that finalized. And so that's, I would say, the biggest wildcard we have that could swing a little bit, and I will look to accelerate and it could have a negative impact, but right now, I would say that's more -- some of that's going to show up in '18 versus '17.

Okay. Would it be safe to sort of assume sort of 75%, 80% in '17, the rest in '18? Or what's the rough split for the cash?

Probably closer to 50-50. That's where I -- I'd still say we have roughly $50 million, $60 million of the restructuring payout this year, and that's something as we progress a little bit further and I have a little more clarity, I'll provide more insight on that in Q3. Again, some of this is outside of our control.

And we will take our next question from Kartik Mehta with Northcoast Research.

Andy, I wanted to go back to your comments on market share and you've acknowledged that there was some market share loss in 2016. And I guess, if you look at the numbers for EMEA and the Americas, it would indicate on the banking side that you've lost some market share. As we go forward, what has been done to make sure that you don't lose more market share? Or do you have any evidence that suggests that, that trend is reversing?

Kartik, the first evidence, of course, is our improved book-to-bill ratio. If you take a look at our second half last year, we were around about 0.7-ish, and we're now clearly through the 1x marker. So definitely more sales, meaning, we're back. From the competitive information that we have, from the deals that we see, we basically win the deals that we -- that our eyes on, and we've been very successful, especially in the Americas, to regain some of the territory that we had lost. Canada was a very successful market for us as of late. And the -- so that piece I feel really good. The second piece, lot of initiatives around training on the software side. We see software sales go up. And the nice thing that we see as a change is, in previous years, we would basically sell software with hardware shipments, and of course, we still do that, but we're now able to go to the broader installed base and sell new software -- new software functionality, multi-vendor software, new monitoring software, and we're finding very receptive ears in our customer community. So software business is growing very nicely. So those are the 2 main elements. And then let me reiterate, we will -- we have and we will take a conservative stance on margins. We will not go down the rathole of doing revenue for revenue's sake, we're going to be chasing margin-accretive deals.

And then, Andy, as you look at the Americas, and you talked a little bit about that starting to improve. But the market share you've lost or the deals that you lost, are there any commonalities? Was it large banks, regional banks? Was it a specific type of solution? Were there any commonalities in those losses?

Well, first of all, I'd go back to what I said previously. Regional banks in the developed markets are woefully low in their procurement efforts. So anything we do in this industry is large project, and predominantly, it's zero or hero, you either win them or you don't. Now what happens is, as I said, the weaker side in each geography lost. And if you step back and say, what does the customer normally do? Well, you normally go with the more incumbent brands that you have. So it's no wonder that in Europe, Nixdorf was the more reliable side to go with. And it's no surprise that in the Americas, Diebold was the safer pair of hands to go with. But in this process, when we had a lot of uncertainty, and needless to say, our competition helped with that, we did lose some deals last year, but as I said, I think we're on a very good recovery rate. Let me add another element to it. The fact that we weren't allowed to combine our business in the U.K. for over 12 months was a, unanticipated; and b, created a very awkward absence of our company in what's the third-largest developed market in the world. So with the opportunity that we now have, we literally just had our day 1 with our teams in the U.K. My calendar is filling up with invites from large British banks who want to talk to us. We were completely absent. We were not allowed to talk to these people and with us being back in the market, I would also expect us to have a better opportunity to participate in that huge opportunity, because the British banks still are amongst some of the most active when it comes to technology refresh.

Then, Chris, I want to -- when you, in your prepared remarks, I just want to make sure I understood something. I think when you gave the new guidance, you said that shortfall from a systems standpoint, you thought 70% was banking, 30% was retail. I just wanted -- was this retail that you were anticipating on winning in the Americas and it's just taking longer? Or are these deals you were anticipating winning, and they just went to a competitor?

It's a lot more of the former. With regards to the deals taking a bit longer, we also had a couple of things in that legacy Brazil other business, that roll-up into retail that also given the environment in Brazil have pushed out. So it's a bit of what you would call traditional retail and a bit of what you would call that legacy other, that now rolls in there from a consolidation standpoint.

And then just last question, Chris. Your net debt, I think on a trailing basis, you are at like 4x, maybe on a forward basis, you're like at 3.5x. Any issues with covenants? Or any other issues, because the net debt ratio is maybe a little bit higher than you had anticipated?

No, from a debt covenant standpoint, looking at leverage, we're at roughly 3x when you look at the overall agreement, which that's all public information as well. So when you look at that, and where we're at from a forward-looking standpoint, we have no issues with bumping up against our leverage covenants or the other key items.

And we will take our next question from Joan Tong with Sidoti & Company.

I would like to ask you about that $240 million in cost saving for that -- of synergy for that plan DN2020. Since like, you have accelerated or expanded it obviously, the total dollar amount, and how should we think about the trajectory now compared to like a couple of quarters ago? You talked about the trajectory, how we should be able to benefit or realize those cost savings. Is it accelerated? I'm just more interested in 2018, how much like cost-saving synergy you can realize in that period of time.

Joan, Chris is going to give you a little bit of a more of a breakdown in a second. Let me just give you one of the main levers, and it shows you where our teamwork is really starting to pay off. We had said that we would take the number of terminals in our system business from 96 down to the mid-30s by the end of 2020. Given the work that has been done, we are able to accelerate this by a complete year, which will not only benefit our hardware business, it also has a very positive flow-through to our service business. Because you have the flip effect that I tried to describe earlier, less systems out there, less systems to train people, less parts, et cetera, et cetera. So if you're looking for drivers over and above aggressive cost management, that's one of the very big levers that we're able to pull, and Chris can you give you a little bit more breakdown on how it manifest itself in the P&L.

Yes, and I would add one additional example as well. If you think about it on our service side and looking at the opportunities, we clearly see some major opportunities for cost out in Europe. Some of these areas, we are exploring and we have or are in the process of getting the appropriate agreements to do more nearshoring, especially given where we've seen the overall profit out, and that's going to take some time, and there are also some systems-related efficiencies that we're looking to get. And these also take some time to implement. So some of this is you have to invest a little bit and you've got to work through some of these agreements before you get it. That's why you see some of this layer in more in that '19, '20 time frame from an overall ramp. And so if you think about it year 1, we're looking at roughly $50 million and around 20%. Year 2, that's going to ramp and be somewhere roughly 40%, 45%. And then as we get some of the full benefits of the systems initiatives and finalize on the headcount out, unfortunately, certain markets are not like the U.S., where you can take a decision and effect it in months, it takes quarters at times to get some of the people out. And that's where we see it then ramping more so in the 2019, 2020 standpoint to get to the full $240 million.

Okay, got it. And then, maybe just stick to the market share and then on the sales side of things. So obviously, you lost some share in late 2016. But then, in the past couple of years, we also have seen, like some of those Asian vendors have come to America and take some like business away from you guys and also your competitors, like in the Western world. So I'm just wondering, with the winning of the patent litigations from Hyosung, I'm just wondering like how fast we would see some positive impact there, maybe winning some market share back on that front?

The biggest market where we were not active and were, as one competitor gained a lot of share, was in the ISO market in the U.S. We have had our first wins into that market late last year, early this year. We are definitely gearing up our product portfolio for this market segment and you should see more of Diebold Nixdorf also in the ISO space going forward. Now having said that, those are also longer sales cycles to get these partnership agreements in place. But that's definitely the area where we are going to fight back and are looking for additional markets that were not on our radar screen in the past.

Okay, any sort of potential opportunity in the larger bank customers? And we know like some of them actually have used Hyosung products in the past couple of years. Any potential to get back some of those business from them?

We're very aggressively pursuing all the large banks, Joan. And while you never know in which one you will have a breakthrough, just given the level of interaction, the level of interest in our portfolio, I would expect us to also break into accounts where we weren't that successful in the past. But I cannot give you a breakdown at this point in time in earnest, to say where that would materialize first. But in general, let me give you another data point, because the question behind your question is, why do you think you're winning? And one of the biggest things that we've done is we've truly refocused our company in the last month on customer engagement and customer interaction. And yes, you can blame us for having been too inwardly focused, especially in the second half of last year. We've taken very aggressive actions to change that. We've also taken very aggressive actions to invest into innovation. If I take a look just at our R&D budget today, we will break through the 50% barrier on the downward slope of how much of our money goes into maintaining existing systems. And by the way, in the past, for either legacy company, these numbers were probably pushing closer to 70. So as we reduce terminals, the amount of money that we spent on legacy, maintenance and all that stuff, goes down rapidly, and the amount of money that we can invest into go-forward topics, into recycling, into software, into cardless ATMs, into connectivity, into cloud-based solutions, is going up dramatically, and that's one of the main drivers for our future growth opportunities as we sit here today.

Right. And then finally, like your 2020 -- DN2020, you mentioned that 2% to 4% top line growth. And knowing that you have a new normal right now and things seem to be -- projects seem to be more competitive, but shouldn't like the secular driver there, it should be still there and so we shouldn't change -- think of there's any change to that like sort of like your target, the 2% to 4% top line growth a year?

I guess a couple of comments here. First of all, we're getting a little more detail on this, whenever we started talking about 2018. When we look at the $3.50, as I mentioned earlier, we're looking at a balanced path to get there. And obviously, some level of low single-digit growth is helpful to drive that overall. And we just need to see where some of the things are at in the market here and we'll update accordingly. Beginning to see great opportunities, there's some very interesting step change type of projects that are out there. And it's too early to, I think, get into the long-term details about 2020 on the top line, but we've got a lot of opportunities on the bottom line to drive the overall earnings accretion.

And we will take our next question from Jeff Kessler with Imperial Capital.

I know you've talked a bit about -- you've talked a bit about customer service and the loss of customer service agreements, the margin effect, and what I'm getting to is it seems clear at this point in time, that overall solutions that you are providing require the customer to love you, and to love you for more than 1 or 2 or 3 years, and to create a long-term trusted partnership. What are you doing specifically? And you've laid out a couple of things that you -- a couple of points, but what are you -- what have you done in the last 2 or 3 months to create an infrastructure that will reinforce, and [in long] the whole customer service process? How much more training people do you bring out? How much more service people do you bring out to teach the younger folks how to do things? And vice versa, for those employees you have who are not IT, who are not up to IT snuff? What are the things that you are doing specifically? And where is that spend is going to come from, to make sure that your customers stay with you for a longer period of time on that service and software side, as that becomes a bigger and bigger part of your company?

Yes, so all great questions. This is exactly the reason why our service margins are where they are today versus where we probably would've wanted them to be in the near term. So let me go through -- give you a few data points. Today versus this time last year, we've added approximately 150 service techs in the U.S. alone. If you talk about support and customers loving us, when there's -- if there's ever a time to overinvest on techs and on-site support, it's when you roll out new products, which is exactly what we did. You then take into account that the labor market for IT service techs is super tight, there basically is no unemployment. We are looking at attrition rates that are probably pushing 10%, so we actually, in totality, hired and replaced people, so we probably brought some 500 folks back on the payroll to also replace attrition. We put all of those through training. We train them not on one machine, but on 5 machines, to make sure we do not have the utilization issue that we showed earlier. And by the same token, we've invested heavily on the tools side. So all our new products have diagnostic tools on them. We're going down the whole route of preemptive maintenance, predictive maintenance, we're putting a lot of software out, we're piloting very innovative data warehousing, Big Data software analytics with some of our top customers to not only make sure that we do great break-fix service, but to make sure that we truly can provide 99.9% availability for our customers. We talked about that earlier, that shift from break-fix to always-on is both a huge opportunity, but it's also a big investment. If I take a look at all the service data that I see from where we used to be last year to where we are today, our first time fix rate is up dramatically, our call rates per machines are down, our customer sat rates are up and we also were the leader in this market. So we're up from a very high springboard, so all the parameters are pointing into the very right direction. But you're absolutely right, Jeff. The best way to counterbalance volatility in our business is with more long-term service contracts, which is why it was so important that we renewed our 2 largest outsourcing contracts earlier this year. And the more we have 3- or 5-year service contracts under our belt, the higher the predictability of the revenue stream in that area.

Is that going to be found in your gross -- in your cost to -- in your cost of sales? Or is that going to be found in G&A?

No, that's -- the majority of what we do on the service side, all these actions, but the majority, you'll find in our SG&A and our service margins.

And we will take our final question from Josh Elving with Lake Street Capital.

So most of my questions have been answered. I guess I just had a question about software. I was hoping you could maybe break out the mix between license and professional services within software, and then maybe talk a little bit about how you see recurring revenue in that line?

Very rough numbers, we've always said that license is around about 30% of our software business. We do see license revenue going up, which is the good news. But we also see professional service revenue to go up, especially when you integrate software into the environment. So I would say the ratio probably will stay the same for quite a while, while the overall software business in general is growing. So no major shifts between the elements. And that's also what you see in the margin profile. Keep in mind, our software margin is a blended rate between the software license margins that are normal in any software business and the PS margins that are also normal in that type of business. So moving up, similar ratios.

Sure. And how do you think about the recurring nature? Are these all new? Go ahead.

These are recurring, and the other thing is, and that's what I said earlier, we're investing very heavily into SaaS and cloud models. And our Kony partnership is a very big enabler for us to do so, because that will all be SaaS revenue going forward. So working very hard in increasing the recurring nature, not only on the service side, but also on the software side of our revenue stream.

And we will take our next question from Arun Seshadri with Crédit Suisse.

Just a couple of questions from me. I just wanted to understand, in terms of legacy service contracts, Wincor in North America, is there any way you could quantify how much is left in the revenue base today? And how we should think about that rate of attrition, and the effect for this year? And maybe if you can give any color around the number of (inaudible) that you're talking about?

Yes, I mean, a lot of pieces there. I think the easiest way to look at it, if you break down and you look at the roughly $2.4 billion of service, $2.4 billion, $2.5 billion of service revenue, roughly 70% of that comes from annual contract base, and that's largely been stable across our major markets. I would say one of the big areas where you've seen a little bit of a change is the legacy China, because of the new model there. But I don't think we've typically provided the overall units across that global universe that are out there under contract right now. But I would say roughly 70% of the overall contract or revenue, excuse me, would be contract-related. The rest would be your installation, your other managed and outsourced services that come through as well. So a very large stable base of annual contract revenue that exists out there still. And it's been stable.

Got it. So is there any way -- there's no way you can sort of give us some sense for that legacy service contract, Wincor in North America, how much is left in that revenue base today?

It's fairly -- yes, it's fairly small. There's maybe 5,000 units or so. I mean, it's relatively minimal at this point.

And then the other thing I wanted to understand is software went from plus 6% ex foreign currency in Q1, revenue down 10% in Q2. Is that just natural lumpiness in some respects? Or just wanted to understand what percentage of software revenue is truly recurring? And how we should think about normalized retention there?

Yes, I mean, the big driver of that net change year-over-year was a software project actually delivered in the U.S. in North America in the prior year that did not repeat. And so you get some of that large project lumpiness. And so when you have the 70-30 split between your professional service and then your software license and your software maintenance activities, you do get a little bit of that lumpiness as you deliver and finalize on some of those big projects. And some of that's also driven by, unfortunately, the rev rec treatment that you have on some of those as you finalize the completion. So you will get a little bit of that lumpiness. But the base of the recurring still is the smaller portion. And so that's -- it's not large enough to give us the level of stability that we would have otherwise.

Last thing for me is cash taxes for this year. How should -- is there any way you could quantify your expected cash taxes?

I'm going to hold off on giving a definitive answer on the cash tax, and this is largely because we are in the middle of a very detailed large project around our legal entity consolidation. And as we go through that, that's going to actually drive certain tax obligations. We're looking to do things in a tax efficient manner. But there are still a lot of moving pieces there. So I'm going to hold off on that one and I'll provide a little more clarity as we get further in the year on the final outcome of that cash tax payout. Thanks.

And it appears there are no further questions at this time. I'd like to turn the conference back to our presenters for any additional or closing remarks.

I'd just like to thank everyone for participating in today's call. If you have additional follow-ups, please give us a call at Investor Relations. Thank you.

And ladies and gentlemen, that does conclude today's conference. I'd like to thank everyone for their participation. You may now disconnect.