Pegasystems Inc (PEGA) 2017 Q3 法說會逐字稿

Good day, and welcome to the Pegasystems Third Quarter 2017 Earnings Conference Call. Today's conference is being recorded. At this time, I'd like to turn the conference over to Kenneth Stillwell, Chief Financial Officer, Chief Administrative Officer and Senior Vice President. Please go ahead, sir.

Thank you. Good evening, ladies and gentlemen, and welcome to Pegasystems' Q3 2017 earnings call. Before we begin, I would like to read our safe harbor statement.

Certain statements contained in this presentation may be construed as forward-looking statements as defined in the Private Securities Litigation Reform Act of 1995. The words expects, anticipates, intends, plans, believes, could, should, estimates, may, targets, strategies, intends to, projects, forecasts, guidance, likely and usually or variations of such words or other similar expressions identify forward-looking statements, which speak only as of the date of the statement was made and are based on current expectations and assumptions.

Because such statements deal with future events, they are subject to various risks and uncertainties. Actual results for fiscal year 2017 and beyond could differ materially from the company's current expectations. Factors that could cause the company's results to differ materially from those expressed in the forward-looking statements that are contained in the company's press release announcing its Q3 2017 earnings and in the company's filings with the Securities and Exchange Commission including its quarterly report on Form 10-Q for the quarter ended September 30, 2017, its annual report on Form 10-K for the year ended December 31, 2016, and other recent filings with the SEC. Although subsequent events may cause the company's view to change, the company undertakes no obligation to revise or update forward-looking statements, whether as a result of new information, future events or otherwise, since these statements may no longer be accurate or timely.

And with that, I will turn the call over to Alan Trefler, Founder and CEO of Pegasystems.

Thank you, Ken. Though there are many achievements this quarter and year-to-date, we didn't execute as well as I feel we should having in closing some key Q3 deals, and I know we can do much better. As I've said, we're making enhancements to our sales approach and team, and it's taken some more time for some to adapt than we anticipated. I will discuss some of these changes as I go forward, and I remain confident in our long-term strategy and ability to execute.

We're also seeing faster-than-anticipated moves to the recurring flash/cloud revenue model. This is a very desirable long-term benefit that resulted in some short-term downside pressure in our quarterly revenue. If the mix shift hadn't happened as quickly this quarter, our numbers would have looked much better. Nevertheless, we're happy to see the shift accelerate, and you can see its benefit as well. Ken will provide additional financial details later.

Now at Pega, our motto is Build for Change, and we're actively making changes to improve execution to finish the year as strongly as possible and set us up for success in 2018. For example, we are simplifying processes and ensuring we have crisp alignment in our sales approach that enable us to execute with greater success and velocity. We are increasing our awareness campaigns. You'll be more likely to see us positioned as the real CRM alternative. For example, we've got an aggressive field marketing campaign running now in San Francisco in conjunction with Dreamforce.

We're making our products easier to try and to buy and making it easier to do business with us. We continue to grow the Pega community to support the growth and demand for our products, and I'm especially pleased to see the continued increase in training since we announced 3 online Pega Academy courses at PegaWorld in June. Since then we have seen a per-day enrollment rate increase of 47%.

And finally, we're changing some key staff members. For example, in sales management, to ensure we have the right talent to move us forward at greater pace. We've been very assertive about bringing in new talent, including a number of senior sales execs coming from our toughest competition. With increased market awareness, we're seeing a significant amount of excellent, excellent resources actively coming towards us. We're excited about the flow of talent from competitors who understand the value we bring in the industry and know how to position Pega in the front-office CRM market. We've made more than a dozen such hires in sales recently, and our mission is to combine their experience in selling into the front office with our superior technology to address any gaps and to accelerate our selling.

So let me talk a little bit about what's working. My confidence is high and reinforced by the many positive trends we see, not just this quarter but over the last 9 months. Let me summarize.

First, our business remains strong, as reflected in our growing backlog. The sharp move to recurring arrangements, an increasing amount of which are cloud, improve our long-term disability. And our focus on margin expansion continues to pay off, demonstrated in our mix-adjusted margin and operating cash flow, about which Ken will provide more color in a moment. And I'm confident we have the best software technology in the market. Our offerings continue to be highly rated by influential industry research firms and frankly, wow clients and prospects in demos. That momentum is reflected in the number of deals we've closed year-to-date, which has increased 40% compared to last year. We also continue to see many more new logos.

So we're doing a lot of great work and are seeing results. We've also documented our sales management practices this past quarter and revamped internal procedures to streamline our responsiveness to this dramatic increase in business volume. And we continue to see a good mix of business between existing and new organizations, between our applications and the core platform and across our verticals with strong inroads into CRM applications.

And most importantly, we continue to see clients using our software to make meaningful and positive changes to their businesses. This quarter, for example, a new business, Sanofi, selected Pega Customer Service on the cloud to better manage claims and requests coming from its B2B customers. Their goal is to significantly increase customer satisfaction to an outstanding customer experience while improving the employee experience and productivity. France will be the first country to be deployed, with another 23 countries to follow, a great competitive win over SFDC.

And government continues to be fertile ground. The U.S. Department of Justice expanded its use of Pega following formal evaluations, comparing us to competitors that included no cost and open source options. Pega was selected based on a unique [level] support for 0 co-development, our unified agile platform and robust offline mobile support.

In terms of great go-lives this quarter, client AIG Japan went live with their AIG Connect application. Built on Pega Sales, it serves 160,000 sales agents working for independent agencies as well as AIG Japan employees. This is a critical milestone in its merger with Fuji Fire and Marine and will streamline sales processes, reduce costs and improve productivity for the newly merged company. We're very excited of the significant role Pega is playing in the success of this merger.

And UnitedHealth Group went live with a mission-critical application to support their Advocate4Me program, an initiative to transform how they service members. The initial Pega robotics implementation proved so successful that the company has rolled out additional Pega robotics-driven solutions to support many applications. The user base continues to grow and is now more than 4,000.

So these successes are based on key differentiators, which are meaningful, not easy to copy and go far beyond marketing hype. I think fundamental to them are 3. First, real-time AI for customer engagement. We continue to make advancements that enhance our real-time artificial intelligence offerings for clients. We announced, for example, the partnership with Accenture to offer Intelligent Customer Decisioning as a Service to improve customer engagement and marketing for communication services providers. This new AI offers and leverages Pega's Customer Decision Hub, the only battle-tested platform on the market.

We also introduced an incredibly cool AI-based virtual assistant that enables faster and more effective responses to customer service requests across channels. It instantly analyzes incoming chat messages, reads e-mails, listens to phone calls to shorten resolution time, eliminate errors and improve customer satisfaction, all while reducing the burden on service agents.

And, most importantly, we announced the industry's first artificial intelligence transparency controls to help clients mitigate the risks associated with AI. As we discussed at PegaWorld, our pioneering transparency switch capability allows organizations to set appropriate thresholds for AI transparency, predefining levels that each AI model will use on a simple sliding scale from transparent to opaque.

This is especially important to clients in highly regulated industries that need to be able to explain how their systems are making decisions. For example, if a consumer bank needs to show why a particular product or service was deemed suitable or denied to a specific customer, they need to be able to explain it. Only transparent AI can do that. It's also timely with European General Data Protection Regulation, known as GDPR, scheduled to go into effect next May. This new regulation gives European customers more control over how their personal information is used and mandates that businesses are able to explain the logic behind AI models or face massive fines. Our offerings, and particularly this T-switch, will be very attractive solutions.

We're very well positioned in the AI space because our entire technology platform is based on the principle of software that writes software, now with robotic feedback loops and cloud-based machine learning that gauges the performance of staff to recommend process improvements. While our competitors deliver increased marketing hype, we are steadily delivering real business value to our clients.

Secondly, end-to-end automation, which is essential to increase customer engagement, provide a great customer experience and save money. We've all experienced handoffs between a company's departments with the information we provide to one department having to be repeated and all the frustration that comes with that. End-to-end automation and great customer experiences can only be achieved and maintained when the software's architected with that in mind, with a unified platform.

Our competitors attempt to position sets of disconnected offerings of "integrated platforms." However, we're in the enviable position of having the most unified platform architected from inception to support cross-application, cross-functional and cross-channel customer engagement. Last quarter, we talked about our leadership in Forrester's newly created category of digital process automation, representing what they see as this evolution from traditional business process management. And just a few weeks ago, Gartner also named us as a leader, if you look at the picture, clearly the leader, in their business process management intelligent systems magic quadrant for the 12th straight report.

I'm particularly proud of 2 Gartner observations. One, that Pega develops market-leading product capabilities well in advance of others, and that our customers have exceptional product satisfaction ratings. In addition, we were just named a Hot Vendor in Robotic Process Automation by leading analyst, Aragon Research, in that space. We're the only company to bring robotics seamlessly into a true end-to-end process.

And third, we're the only vendor to offer true cloud choice, which continues to be an important market differentiator. We can do that because our software writes the software because, we can generate different software optimized to be cloud native on each appropriate platform. Our Cloud Choice Guarantee delivers enterprise-grade SaaS on the cloud platform that works best for our clients and gives our clients control.

Now interestingly, Salesforce this week announced they have the option to run on Google or Amazon Web Services. But this is just a parlor trick. Smart buyers know would still mean that Salesforce, and only Salesforce, has their hands 100% on the controls. Only Pega offers the combination of offers -- options and support that's going to be needed for an unpredictable future. At the same time, our Pega Cloud is a very robust cloud offering that continues to be increasingly popular for clients who prefer to run their Pega apps on our fully managed virtual private cloud infrastructure. Though still a relatively small percentage of revenue, cloud is growing fast and strengthening.

Toward that end, we recently hired Frank Guerrera as Chief Technical Systems Officer to help drive optimal client success through our cloud services. Frank has spent the last decade driving the transformation and growth of the most leading Software as a Service vendors and most recently, at HPE. He'll direct our cloud strategy to ensure continued innovation, excellence and competitive differentiation. We're very excited to have Frank on board.

So in summary, we're happy with the fundamentals of our business. We're committed to improve aspects of our execution, and the management team is assertively using this commitment to accelerate our transformation to recurring revenue and greater predictability. We will continue to focus on go-to-market improvements and scaling our business system to support the massive CRM market opportunity at hand.

And there's a lot more to say, so I'm going to turn it over to Ken, our CFO, to continue.

Thanks, Alan. So we've discussed in previous earning calls that we always look at the performance of the business on a year-to-date basis because it's the most relevant view of our performance. The longer the time frame, the more it reflects the true trends and business performance. As Alan suggested earlier, we're really only addressing a fraction of the market opportunity that's in front of us. That said, there are many positive outcomes that we've seen through the first 3 quarters of 2017.

On the surface, our top and bottom line results are solid, but we know the opportunity is much greater than we've captured so far in 2017. So our go-forward execution against this opportunity is our primary focus. There's just too much opportunity to ignore.

Let me take us back for a moment to how we progressed through the year on a recurring mix. We started off 2017 planning for a 50-50 mix of recurring term and cloud versus onetime perpetual commitments, and we're actually pretty close to that through the first quarter of 2017. We then saw an acceleration to recurring in Q2, where we had somewhere close to 60% and then, we finished Q3 with an amazing 79% recurring. 79% recurring, wow. Year-to-date, we're at 62% recurring, which is a pretty big shift from where we modeled 2017. As I've said before, I hoped we could get to a much greater amount of recurring than 50%, so this is great news on where Q3 and year-to-date landed.

To remind everybody, a shift of 1% away from 50% would impact full year 2017 revenue by approximately $3 million. Although we can't predict where Q4 will land and the impact to the full year mix, it is likely we will land 2017 full year with more recurring commitments than onetime. Whether we land above or below the current 62% depends on how things close out in Q4, which is always our strongest quarter of the year. So the movement to recurring certainly is continuing. But at the current pace, this faster shift to recurring revenue would impact full year revenue by approximately $40 million.

It's worth mentioning that Q3 2017 is a bit of a challenging compare. Our year-to-date top line revenue for the quarter declined 2% from a year ago primarily due to this recurring mix shift we've been experiencing. A common discussion with our investors, prospective investors and analysts over the past year has been around this shift transition. The value of a business of Pega's size with an increasing amount of recurring and predictable cash flows is well worth the short-term pain in the transition. I think everyone would agree with that. Many investors have went a step further and suggested that we should move as fast as we possibly can in 2017. Well, so far in 2017, that exact outcome is occurring. Of course, all of the impact to this mix shift would drop right to the bottom line because our costs are largely unchanged regardless of a term, cloud or perpetual arrangement. In fact, our commission costs are slightly higher for term and cloud deals versus perpetual, so our revenue is actually lower and our costs are slightly higher.

Another factor impacting our year-over-year performance in the quarter is that we had no whales in the third quarter of 2017, though we had one in the third quarter of 2016. To remind everyone, our definition of whale is a client software commitment of greater than $10 million. Whale-sized arrangements, which just exaggerate the lumpiness, are still very likely in our business going forward. We have a number of whales that we are, in fact, chasing for Q4 and into 2018. And recent time -- in recent times, whales have been just as likely, if not more likely, to be recurring.

Our success in penetrating new accounts has led to an increase of 40% compared to last year in a number of deals. An interesting data point is that the average deal size of a new client is about half the size of an existing client. Also, as our customers move to recurring licensing, we see them starting with smaller and faster implementations, with cloud deals tending to have a duration closer to 3 years.

Our year-to-date professional services revenue growth was 27% year-over-year. We have anticipated growth in professional services of more in the 10% to 15% range and our cloud growth of at least 20%. Our cloud growth in revenue year-to-date has been 20%. The stronger services growth is a result of a small number of consulting service engagements where we're playing a bigger role than anticipated. This increase in consulting revenue does not affect our efforts in building our ecosystem and having partners perform an increasing amount of Pega services. Over time, we expect services growth to be more in line with our longer-term growth expectations as we enter 2018.

To repeat something I said earlier, our full year guidance was developed using a 50-50 split between recurring and onetime commitments from our customers. And each 1% shift away from this 50-50 has the potential to reduce full year 2017 revenue by $3 million. Year-to-date, we stand at 62% recurring and 38% onetime.

An important piece of information around this is that cloud is a driver in the move to recurring more so than in previous years. Our Cloud Choice message is really resonating with clients feeling confident that their investments are protected against future decisions they will make related to deployment environments.

As an additional piece of information, we began disclosing annual contract value, ACV, a few quarters ago. We believe annual contract value is highly correlated to recurring and predictable future cash flows. Our term license and cloud ACV grew to $200 million, up over 20% year-over-year. Our total ACV, which is term license, cloud and maintenance grew 17% year-over-year to almost $450 million. We believe this is a reflection of both the long-term value in Pega's solutions and the broader market trend of moving to recurring spend in the enterprise software landscape. The technology industry research community suggests this trend will continue and potentially accelerate. We hope to capitalize on that shift.

Our embracing of financial awareness continues to positively impact our bottom line performance. Our year-to-date costs have increased at reasonable and sustainable levels. Our operating margin improvement, if adjusted for the recurring mix shift in 2017, would be over 300 basis points year-to-date. So we're really making great progress if you take the noise out of the increase in recurring commitments.

We've reported both GAAP and non-GAAP results. A full reconciliation of all GAAP to non-GAAP measures is provided in the financial tables of the press issued earlier today and is available on the Investors section of our website.

On a year-to-date basis, our non-GAAP fully diluted net income was $0.59 per share compared to $0.57 per share in 2016. If we were to factor in the impact of this mix shift, our EPS would have increased by over 50% year-to-date versus 2016 year-to-date. This increase was driven through great execution of our strategy to increase costs at a slower pace than revenue while continuing to strive for 15%-plus top line growth.

Moving on to backlog and the balance sheet. We compute license and cloud backlog by totaling 2 components: deferred license and cloud revenue, as posted on our balance sheet; and license and cloud contractual commitments that are signed but have not yet been billed nor revenue was recognized yet. As a reminder, you can find detail of both components in our 10-Q and a summary table on our press release, both of which we filed earlier today.

We finished Q3 of 2017 with $548 million of total license and cloud backlog. Backlog grew over $120 million or 29% from the end of Q3 2016. Our average contractual commitment duration continues to average approximately 4 years, which is not a material change from previous periods.

From a cash flow perspective, we produced an amazing $114 million of operating cash flow for Q3 of 2017 versus $21 million for the comparable period of 2016. We finished the period with total cash and marketable securities of over $190 million, an increase of over $60 million from the end of 2016. We're really beginning to see some impact from our financial awareness initiatives and recurring commitments.

On headcount, we finished the period with over 4,000 employees, up 7% from September 30, 2016. The growth in headcount is disproportionately concentrated in lower-cost territories as we expand our global client base and enter new markets.

In summary, we're excited about our movement to recurring commitments, an increasing amount of which are cloud; our solid management of expense growth; our mix-adjusted results, top and bottom line; and tremendous operating cash flow production. We still have a lot of work to do in Q4 that will require focus and execution to ensure we capture our share of this massive market opportunity. We're laser-focused on finishing 2017 strong.

And with that, operator, we will open the call to questions.

(Operator Instructions) And our first question will come from Steve Koenig with Wedbush.

I've got a few questions, so I'll ask these and if there's still time on the call, I'll jump back in the queue. So Alan, you talked a little bit about the sales changes that are going on and execution improvements that you're working on. The -- so your actions to remediate the Q3 execution are welcome, but they suggest that there's more than just lumpiness. There's something a little more systemic that you need to tweak at least. Can you give more specifics on -- maybe in the quarter, what areas were weak? Any areas better? What sorts of issues are you seeing that you're working on? And I'd love to get color on a little more specifics on what changes you're making.

Yes, sure. I think that we've talked about how we've been moving from the traditional relationship model that often works very much with the folks we, frankly, in many cases, knew for long periods of time and was very much geared on having a customer share what their challenge is, what their issues are and from that, us really crafting a solution that would sort of meet your needs. That's the basic of traditional relationship selling, and it's how the business largely grew up. Starting probably about a year ago, perhaps a little more, we really started rolling out much more of a challenger messaging. The idea with challenger is, you've got much more out-of-the-box value props. You need to, frankly, be more assertive about finding and talking to people you, to be honest, may not know. And the style is what I would describe as much more proactive and much more engaged. I think that we perhaps underrealized how the traditional model allowed some -- certainly not all, but some folks to really rest on a behavior pattern of talking to the people they already knew. And we need and should have the opportunity to talk to whole collections of new people who are responsible for these front office functions around service, sales, marketing, where we know we've got a winner. I mean, the fact that we've got a winner is proven by the fact that we won a lot of business against well-established and well-known competitors. But some of our behaviors were not, I think, as assertive as they needed to be. So we brought some outside folks in from some experts in challenger selling. I've been optimistic this might have clicked a little more than it did, to be very transparent. And we, instead, decided in the last, frankly, 2 or 3 months, have brought in a number of very senior folks who come from what I would describe as a much more proactive and assertive background. We're reaching out assertively, not waiting for leads, is part of the culture. And that's something, I think, that they can both improve our leadership on and they can help train the rest of the organization. I don't want to go dribble names all over, but if you go to LinkedIn and do a search on people who used to work for the competitors you might think of and now work for Pega, I think you'll see some incredibly impressive backgrounds. And I was amazed at how much interest there was in talented, experienced, highly successful people from these companies coming to Pega because frankly, they've lost to us at times in the field and they know that our stuff is better. So we're -- we've, I think, recognized that we needed to be assertive. We've taken these steps. We have some more that I have hired but can't talk about yet because they'll come in, in Q4. And I think we've decided that we wanted to really be as forceful as possible so that we're not sitting going into 2018 wishing we had been stronger.

Okay. That's very helpful, Alan, I appreciate that. Maybe just following up on that explanation. Can you tell us a little bit more about what is it that you're hoping will make the challenger methodology and the new people contributing in Q4 to make Q4 a lot better? Is it a matter of the new people ramping? Or is it a matter of getting that challenger methodology accepted culturally through a broader part of the organization? And specifically, how -- can that happen quickly? And how can it happen in Q4?

Well, a couple of things. So one, we've got a tremendous amount of business queued up to close into Q4. We've had, in terms of the late stage of our pipeline cycles, what we refer to as decision and legal. We have massively more in there than we had historically. And we typically are quite successful in closing that sort of business. So I'm not counting on new heads coming through the door to cause us to execute in Q4. And historically, Q4s have always been really quite big for us, and I'm hopeful that this will be the case as well. However, I'm not settling for Q4. I've been doing this for a long time, and I'm committed to changing the velocity of this business. And my management team has recommitted to changing the velocity of this business, and that means we got to get out of the sort of this current envelope of velocity into one that's really much more suited to what we can do for our clients. And so, that's what these changes are about more than anything else. It's really about making sure we get a strong launch into 2018. And I can tell you that, in doing work in preparation for that sales kickoff, you're going to see us coming in with a very assertive new set of messages that build on what we've done historically but are consistent with the energy level that I'm hopefully conveying.

And next, we'll take a question from Greg McDowell from JMP Securities.

A couple questions and then I make it back in the queue like Steve and ask some more. But first, I want to drill into the year-to-date headwind. As you called out in the press release, it's around $40 million. And if I just walk through some remedial math here, the original guidance was $860 million. I take $40 million out, I get $820 million. So I'm just sort of wondering, what is that Q4 incremental adjustment to -- that I should make to the, what feels like, the new baseline of $820 million. How's it feeling? These big deals in the pipeline, how do you anticipate those deals will be structured term license versus perpetual license? So basically, 2-part question, first.

So I think -- I mean, your math is right. I mean, that's logical, what you said. This is the way -- so let me kind of tell you the way I'm thinking about it. We, as most software companies, we do have deals that close, close to the end of a fiscal quarter, right? No big surprise to you. And so we will have a lot of activity in the month of December like every other kind of enterprise, especially because it's the end of our year and salespeople get into accelerators, all the normal behavior that you would see. That said, we are absolutely committed to recurring deals. We are at 62%. That's a pretty reasonable jump over 50%, and we'd love to see Q4 come in above 62%. I think that would be great. And if that happened, naturally we'd even -- have even, if they came in, say, at 65%, we have even a little bit of incremental headwind. And so the way we're looking at it is that, that is a perfectly fine trade-off for us to make if that happened and then we would have kind of less revenue but more ACV and more backlog build in that scenario. Now here's the catch. The catch is, we can't predict where these deals are going to fall, and deals change, I mean, literally right up until the last part of the negotiation in some cases, where customers will decide to go cloud, they'll decide to go term and so we -- so we're perpetual. And so when we think about that, if we have a scenario where, say, Q4 was 62% perpetual and 38% recurring, like, exactly the opposite, you can imagine that we would -- our revenue would go up closer to at or above guidance or et cetera. So that's the kind of -- in my mind, it's hard for me to gauge where we're going to land. But I will tell you, I'll absolutely take a trade-off of a higher recurring for short-term revenue at this year, and I hope everybody kind of understands why I say that.

A bunch of this is gut feel, if you would ask me. I mean, this quarter was shockingly term given where we were.

Term and cloud, yes. 79%.

Term and cloud was 79%. It's hard for me to envision that, that will repeat. It's clearly not going to swing all the way back to 50-50 for the year. I mean, if anything, I think the year-to-date average probably represents, as a former statistician, probably represents the highly [varied and the best sort of guesstimate] that one could expect.

Yes, that's fair. And then, you talked about a strong launch in 2018. I want a strong model in 2018. And I was looking back to the slides from the June analyst day and just trying to think through the framework between growth and profitability. I don't want to say we can throw that out the window now based on the accelerated shift to recurring, but it certainly feels appropriate that it -- there may need to be some fine-tuning given the accelerated shift. So maybe it's a little premature to ask about 2018 and sort of a early glimpse of how we should think about 2018, but I'll ask anyway.

Well, we haven't backed away from the thought of being fiscally responsible. The idea -- obviously, if the shift to recurring happens faster, the math just tells you that, that creates a bit of a setback relative to short-term margins. But relative to the actual strength of the business, it's an enormously positive, particularly with the level of exposure we give, where we're so complete in showing what the backlog is by year. I mean, we really try to create a lot of visibility here as well. But the financial awareness work that Ken brought in when he took over and I think is very much vested in the company, I think has been well adopted by the culture. You're going to see us trying to make sure that we can be a responsible and profitable company by spending our money wisely. In fact, we have certain advantages, big advantages in that we're not trying to glue together or advance a Frankenstack of 15 products that are unified in name only. We've got some very, very positive things that should be able to help us as we sort of cross the $1 billion threshold, I would expect. But I do think it's premature for us to put a formal model together for next year, but we look forward to doing that.

Let me give you one thing to think about it, Greg, and for those listening because this will probably be a common question. When you go through a transition shift, it doesn't hold forever, as you know, right? Because you hit the points where everything that you're losing in a shift, you're actually gaining from the recurring from -- so we're a good, solid 2 years into this shift, and most shifts take somewhere between 3 to 4 years till you kind of normalize that. So I would say that although 2017 has a -- let's say, we landed at 62%, right? We had 2 years in a row where we had a pretty good move to recurring. The assumption that we would then move to 80% or 90% in future years is probably pretty aggressive in the models, right? So I'm not -- that's not giving you any guidance, but just giving you a perspective. So I kind of look at it like, at some point, the shift starts to normalize for you, right? And we just happen to be through 2 years where we've been fairly dramatic in that shift.

(Operator Instructions) Next, we'll take a question from Mark Schappel with Benchmark.

With respect to the execution issues or missteps in the quarter, were they felt mostly on the CRM side or on the BPM side of the business?

I think, it was mostly on the CRM side. I think, we're fresher in that market. We are not as, frankly, known or experienced in talking to some of the front office-type folks. We know that when we engage in the front office, we do extremely well. But I think to be honest, we may have overestimated the confidence and level of training that some of our staff selling to the front office have internalized. And we've doubled down on fixing that, and we've complemented it with some really good role models, which seems like the right combination to do.

Then Alan, staying on the execution theme, was it related to a particular geography that you saw some difficulty?

Europe was a little stronger than I expected, to tell you the truth, and parts of North America were a little weaker. So this is where I use the L word, which is lumpy. It's really hard to know. I think that -- and parts of Europe were very different from other parts of Europe. I think that we, as a company, and part of what I've done with my management team and more broadly in the firm, is had us recommit to execution as opposed to great strategic initiatives as what we need to do to drive not just the rest of 2017, but really being the center point of 2018. Because I'm pretty comfortable that, strategically, we've done and are doing the right thing. So I don't think that there's any one part that is so great because I feel it's all perfect and so horrible that I'm terribly panicked about it. I think we need to improve execution across the board.

Okay. And then, one final question, Alan. Q3 is usually a strong public sector quarter for the company. I was wondering if you could just comment on if you had difficulties in that particular sector.

I would say we had difficulties. Q3 can be reasonably strong. Things are weird in Washington, as you know. We did close some nice public sector business. I spoke a little bit about it during my presentation, but there was nothing extraordinary there.

And next, we'll take questions from Matthew Galinko with Sidoti.

Just a couple for me. Alan, you talked about kind of the AI model transparency switch. Just curious, sort of how that came about and if that's something that's kind of emerging as a differentiator as you go to market with these products. Or is it not quite there yet?

No, no, it's fresh. It's new. It really just hit the market this quarter. And so people are still learning about it, but there's a tremendous amount of interest and excitement. If you want to understand this, and I think this relates to the whole sort of non-BS aspects of understanding AI, the PegaWorld presentation by Dr. Rob Walker, who's been with us now for -- working on this now for 15 years and I think is one of the real thought leaders in the space, he did a brilliant presentation on the Tuesday morning on the importance of transparency in AI and how organizations can accidentally, very easily violate laws by asking sensibly innocent questions or gathering together innocent-sounding information. You can do things that either violates redlining laws from a racism, a whole collection of things. And it's real thought leadership elements. And because of our architecture, we've been able to implement this, tie this to process, so it's not just making the decision, to making the decision at the right place and the right channel and bring this to market. I would strongly recommend anybody interested in AI to spend the half hour and listen to that presentation. It's really insightful.

Got it. And then, also you talked -- I think it was a 40% increase in the number of year-to-date deals closed versus last year. Can you line that up kind of against your concerns on the sales execution, where that metric could have fallen if you feel you were firing on all cylinders?

Yes, I think -- I'm not sure I would believe it would have been meaningfully higher. I think we would've gotten a couple more deals that we thought were important. It is interesting seeing the business, the velocity of the business system, frankly, be put under that level of pressure. It's also given us some insight, not just in the sales organization but in Ken's organization and other parts of the business, about ways that we need to take a bunch of things that we used to be able to sort of get away with as one-offs and really start to make the things that will be important for the business continuing to run faster in the future. But I viewed that as very exciting and very important. If we had closed everything I wished and had hoped we would close, it would have been up several additional points, and some of those would have been larger, too.

And next, we'll take a follow-up from Steve Koenig with Wedbush.

The deals being up 40% and ASPs down, is that coming from the Global 2000 group? Or is that more the enterprise sales organization? And I also just wanted to ask as well, I'll throw this out there and then I'm done, on Pega robotics, so we're definitely hearing a lot of interest out there in the field, a lot of proofs of concept going on. It sounds like from Q3 that customers are still kicking the tires, or that just these deals are smaller based on your Q3 results not being better. So maybe some color there as well.

So I think on robotics, there's a fair amount of tire-kicking that's going on and, frankly, there's a fundamental problem in the whole robotics sector. It's interesting because McKinsey, about 2 months ago, put out a blog titled Burned by the Bots about how some of this robotic stuff isn't yielding the returns that, frankly, they were telling everybody to go after 2 years ago, which I thought was actually an honest observation. The real problem that folks have is that robotics, as the stand-alone robots running around, isn't the right way to automate a business. And we are perfectly positioned to be able to put robotics in the context of end-to-end processes. I mean, that's what we do, and we're the only ones in the space that have a top-rated process automation engine and as well as a top-rated robotics capability. So a lot of the robotics things that will lead to, I think, much bigger deals in the future involve some reeducation. So somebody who thinks that what they want to do is kind of drop random robots on people's desks to do little task automation needs to understand that that's a fine way to begin, but it better be in the context of end-to-end process, or you don't have any control, don't see what's going on. So I actually think the robotics deals have been smaller than they will be, and I think the robotics deals and the process automation deals are wonderful ways to interplay parts of our business together in ways that will help. And frankly, there's also an interplay there with CR,M, because so much of these can be used well in the customer area. Getting back to the other part of your question, yes, we're seeing a lot of excitement in corporate markets and whereas that had gotten off to a start but a slow start, we're really seeing an energy building there and that both makes me feel good about building a more efficient business system, which we need to do to support that, but also that we can have a more efficient selling system relative to some of the questions about margin earlier.

And that does conclude our question-and-answer session today. And I'd like to turn the call back over to Alan Trefler for closing remarks.

Hi, there. So we're working hard. We're very committed and very excited about the prospect for this business. So sometimes there are opportunities to get the organization galvanized, I want you guys to know that that's exactly what we're doing. And there's a lot of enthusiasm for being a company that can move forward at a faster rate. With that, let me thank all of you, and I look forward to talking to you soon.

That does conclude our conference for today. Thank you for your participation.