Five9, Inc. (FIVN) Earnings Call Transcript & Summary

Hello, welcome, everyone. I guess good afternoon for some, good morning for a few others. Welcome to our 24th Annual Growth Conference here at Needham. My name is Scott Berg. I lead our Enterprise Software and SaaS Research efforts here. Today, we have Five9 with us. We have the company's CEO, Rowan Trollope; and the company's CFO, Barry Zwarenstein. Welcome, gentlemen. Before we get started, 2 points in housekeeping. The first is, we will take audience Q&A when we are finished with our questions here. If you'd like to ask a question, feel free to input that into the Q&A box that's on the presentation window and I will moderate those as we get through. And then secondly, Barry would like to start off by talking about the safe harbor statement. Barry, it's all yours.

Thank you, Scott. I want to remind everybody that today we will be making forward-looking statements concerning events and trends that may affect our industry or the company and its operations. Actual results may be materially different. Please refer to our most recent Forms 10-K and 10-Q under the caption Risk Factors for detailed information about certain factors that could cause our results to differ materially from those described in such forward-looking statements. Thanks, Scott.

Of course. So let's start off with the 1 or 2 people on this that don't know who Five9 is, how about a brief overview of the company.

Sure. I'll kick that off. Five9 is a Software as a Service company. We are operating in the contact center space. So this is the part of the business that receives all of the inbound communication from customers and it's the place where you engage and connect with your customers in a business as your contact center. Now increasingly, the contact centers of all that we all might think about are the sort of like vast seas of cubicles where people just sit there sort of answering phones. But the modern contact center, as powered by Five9, we're completely in the cloud, is much more about digital channels and automation and providing a really delightful level of service for a business to deliver a delightful level of service to their customers. We've been public for a number of years, more than 5 years. We are about $700 million in revenue, give or take, and growing quickly as businesses are looking to replace their legacy on-premises phone system-based support platforms with much more modern contact center as a service platforms that are all powered in the cloud.

All right. Let's get started. We'll start with the, I guess, the A topic a little bit and I realize this subject might be getting a little bit old, but I still get a lot of questions on it. Now that we start moving past the Zoom transaction, with it not going through, does it help or hinder your business? Or do you think there's no real change to what you all are seeing, just in general demand trends?

Yes. We have not really seen any impact on that front. There was an opportunity with Zoom to be able to access a new kind of buyer in our category and that buyer would be the IT buyer. So in some companies, we see that IT buyers are looking to replace their legacy phone systems and on-premises meetings infrastructure. And when they do that, they often need to replace the contact center as well. So the opportunity with Zoom was really incremental, the opportunity to access that new buyer. As a stand-alone company, we continue to prosecute the business that we always have, which is selling to the line of business buyer and so there hasn't been any slowdown on that front. I would say during the transaction, a lot of people had heard about Zoom, not as many people had heard about Five9. Some of our larger enterprise customers saw the Zoom transaction as a positive, if they were a Zoom customer, but it didn't really affect their buying from us. So we didn't lose any deals when we canceled the deal. And then the opposite was also true. We had customers who actually viewed the Zoom thing as a negative. Maybe they say, oh, I already chose Microsoft for my meetings platform. So on balance, it's been awash and the business continues to do really, really well as evidenced by our last earnings call post Zoom breakup. And we didn't lose a beat on execution, that's for sure. So we're still doing really, really well.

Fantastic. Last question on kind of the transaction, but it's really not about the transaction, it's just how I view the space a little bit. You know, Rowan, you and I have discussed, I think, several times, publicly at least, is my research in the space has highlighted what I've always felt is a divergence between the contact center from the UC environment, especially upmarket with larger customers that you all focus on. On the transaction call, I had a chance to ask this, and you talked about this trend coming together a little bit. But at your recent Analyst Day, you really kind of not so subtly talked about how contact center needs to be close to CRM or service desk solutions and vendors here going forward. These 2 concepts are a little bit in competition with each other. Help us kind of sort out what's going on in the market there?

Sure. So if you just look historically at this market, the leaders of the market, the on-premises leaders are really Avaya and Cisco, #1 and #2, in terms of market share in contact center seats. And what Avaya and Cisco sold was a physical PBX, often a hardware PBX, software but running on appliances. And alongside that PBX came a contact center. So first of all, the buying cycles are aligned because when people have a contract with Cisco or Avaya, it's a UC plus a CC contract. They don't have typically 2 different contracts. And the buyer/influencer of Cisco and Avaya hardware is typically an IT person, right? It's someone in IT infrastructure or network infrastructure. Increasingly, what's happened that when the cloud came along and sort of disrupted this space, the contact center kind of pulled away from the stand-alone PBX and so what companies like Five9, and NICE is another example, or Genesis with their PureCloud offer, what we all offer is a stand-alone, no PBX, cloud-based SaaS service. And so what that has enabled is us to go directly to the line of business buyer, who are increasingly empowered to purchase sort of software. Now they always will involve IT at some point, but they are more empowered than they've ever been. And they have been able to -- what we've been able to do is carve out this business of selling directly to the Head of Customer Service or the Head of Contact Center Operations and then we deal with IT on the back end. So there's really 2 separate swim lanes. There's folks in IT who are driving a purchasing cycle that is typically a communications platform, which includes a contact center, and then there are those line of business buyers who are in the contact center space, who aren't thinking about a UC platform at all. They're just thinking about how do they get their next-generation customer service experience delivered ultimately. So for us, what we are seeing is, as we move up into the strategic enterprise, it tends to be more of a joint decision. The line of business buyers are not making multi 7- and 8-figure sort of decisions on technology without involving IT. So it actually ends up being much more collaborative and we see IT showing up much more. So the partnerships we have with the UC companies are really important. And we do see that at least 1/3 of buyers, and this was in a recent study by Piper, I think, who basically showed that around 1/3 of CIOs are looking for a converged solution. They're looking for a communications platform that provides them with all the technology to engage with their customers and that number is on the rise. So IT is getting more influence in this space. But it's approximately 1/3 right now of the buying public are really IT buyers, who are driving these decisions, whereas 2/3 potentially LOB buyers making these decisions.

Really interesting to follow the trends and see how that works out going forward.

You see they're both positive. I mean at the end of the day if somebody wants and it's an IT wants to buy a contact center, we would like to be there and offer them that. That's not our current go-to-market sales motion. In order to access that sales motion, we need to partner up with UC companies. So for example, with AT&T, they have a partnership with RingCentral for office at hand and they have a partnership with us. And that's a place where we can actually meet in the channel and deliver an IT buyer of converged solutions, so that we can access that route to the market as well.

And you all have spoken about your successes with that AT&T partnership recently, which speaks to that point.

It's doing well.

All right. Market, let's talk about the market a little bit. I think my early work in the pandemic, maybe 18, 20 months ago, talked about contact center sales pipelines that have effectively shifted to 100% cloud due to the perceived business risk with implementing an on-premise solution, especially upmarket. I guess first, have you seen that same thing? And I guess the second question there is, contact center replacements seem to occur because of some new business requirement or existing system that's, of course, old and depreciated. Can the pandemic pull forward some of those time-based decisions based on this perceived business risk of the on-prem model as today?

Well, I'll start out with pre-pandemic. We were already seeing an acceleration in our business before the pandemic and we had been reporting that in our quarterly earnings. And the pandemic did accelerate, maybe it was pulling forward deals that were already in the pipe, maybe it was customers sort of considering, as you said, cloud is really the best option because of work from home. Those things were all true. But ultimately, that was a sort of a -- we haven't seen that demand slack off post-pandemic. We really have seen this increased attention paid towards digital channels and better customer experience. And so Barry did some really good work to quantify what the sort of the onetime additional tailwind we got from the pandemic, which was I think low single digits. And we're now through that period. We've lapped it and I think the business is more on its normal pace. And what we've said is durable growth in the 30s of our enterprise subscription business, which is what we foresee. And of course, that's also reflected in our proxy. So net-net, I think it helped. I think perhaps what the pandemic did more than anything else is sort of put the final straw on the camel's back of, is cloud a good option for enterprise contact centers? The answer is a resounding yes. It makes our lives a little easier because when we go into these businesses, we're no longer having to convince them that cloud is an acceptable option. We're having to convince them that Five9 is the best cloud option. And that's a different -- it's a more efficient model for us. And hopefully, we'll continue that acceleration of on-prem to cloud transition. But frankly, as you know, we're still early. I mean 15% was the latest sort of assessment of the amount of systems that have been transitioned. That number might be higher now. I'm not sure, but we're still barely at the beginning stages of transitioning these on-prem systems to the cloud.

Plenty of runway there, no doubt. From within the contact center, I know a product area that we've spent a lot of time last couple of years discussing, and you have directly publicly, is around automation. And you've had some very strong attach rates to your automation modules. And at the Analyst Day, you talked about how AI in your strategic segment had an 88% attach rate while workforce automation -- optimization, excuse me, had 100%, for example. But I guess from a higher level, what's really driving the interest in these automation products, what's still early stage for several of these.

It is early stage. A few things. Let me just start out by saying what we include in that AI and automation category. And essentially, it's everything that we provide to help you optimize your labor spend. So if you think about a typical contact center, for every dollar that's spent on technology, you'll spend approximately $10 on labor. So from a contact center operations -- and by the way that budget, even if IT controls or has a big influence on the $1 of technology spend, back to our previous UC, CC conversation, the contact center operations people are the ones who control the labor spend and really have that in their budget. And so what our pitch to those customers has been is, hey look, we want to really help you optimize that labor. And there's probably some additional tailwind in this portfolio that we've seen through the pandemic due to labor shortages. We've certainly heard that from customers. They just can't hire enough people into their contact centers and so they've been coming to companies like Five9 and saying, how can you help me with automation technology, so I don't have to have as many people. So that's the part of the portfolio that kind of answers that question for a customer. And we do sort of a couple of very discrete jobs. One, we have an intelligent virtual agent, IVA, which is a conversational-based system not dissimilar to an Alexa or a Google home assistant or SIRI, where when you call a contact center, that's what you speak to instead of speaking to the legacy IVR, which -- IVR, interactive voice response; hello, thanks for calling, push 1 for sales, push 2 for support; that's something an experience customers don't like at all. And these conversational-based technologies powered by AI is what we've really been leading with at the forefront of this automation push because they're now at a point where they're much more pleasant to deal with. They're really good at recognizing folks in terms of speech to text and so on, and so that's the new technology we're selling. And customers are looking at that as an opportunity to not only replace that IVR that customers don't like, but also to defray even more calls from landing at agents because if you can do simple kind of transactions with the IVA, the intelligent virtual agent, things like; thanks for calling, I see that you have an order with us because we've looked up your phone number and we see an order; are you calling to find out about shipping status? If so, I can let you know that blah, blah, blah, blah. So you can sort of head those calls off at the pass and not have them end up being handled by human beings. And the net result is, as can be for some businesses, a very significant savings on their labor spend. And that represents a very large opportunity to grow our TAM because we're no longer just looking at the TAM of tech spend, we're looking at an arbitrage of labor spend using technology to drive those efficiencies.

And I believe you all have defined that as roughly a $30 billion opportunity.

$34 billion. Yes, $34 billion. And that's just a piece. It's a fraction of the labor spend in the contact center market globally, which is more than $230-something billion spent on labor. So we think that we can capture some of that labor spend as technology spend in terms of just shifting value from labor to tech.

Sure. So within the contact center market, you have several competitors that are making noise today about kind of their own traction, if you want to call it that. This was a space 3, 4 years ago. You guys were only really compared to NICE and inContact after they acquired them. You had a competitor that went public last year. I believe you had SPAC. You have 2 more that might try to go public here in the near term. But has anything really changed competitively? Or is there just maybe that much more cloud contact center business out there to be had today?

Look, I think when you see the success that we -- I know you mentioned one of our other great competitors, NICE, the success that we've had, I think it's really opened people's eyes to this opportunity. So it isn't any surprise that you've got a lot of sort of new companies into the space. I mean, I think if we didn't see that, I'd be wondering, are we breathing our own fumes here? Or is something really going on? And that's the reality. The reality is, no, the market is super-hot. And of course, it's attracting a lot of competition. And nothing has sort of fundamentally changed from a product and offer perspective in terms of the companies that are coming after this space. So we have a really good lead at Five9, right? Not only our reputation and traction we've had in this space for essentially 2 decades, before it was even called cloud, and that makes a difference for customers. That kind of experience and the trust that you need when you're making that leap from on-premises to cloud, you definitely want to go with a name that you're -- especially if you're moving off of like a trusted brand like a Cisco or an Avaya, you need a name that really is going to engender the kind of confidence and trust that a contact center leader is going to put sort of their career in. If they make a bad choice here, it could really affect their careers. So they look for that trust. And so I think Five9 certainly checks those boxes and we've built up that reputation more recently in the last few years by moving into the Enterprise and more recently the Strategic Enterprise. So we now have some of the largest contact centers in the world running on Five9. So if it's good enough for a Fortune 50 company to have their 10,000, 20,000 agents on our platform, it's probably good enough for you as a small company with 200 agents or 20 agents, so on.

All right. On the demand side, just in the industry as a whole and I guess your specific success, the company's growth in the $1 million plus ARR segment has been really impressive. Your number of customers over $1 million has increased from 49 in the third quarter of '19 to 123 in the third quarter of '21 based on your Analyst Day comments. And Barry, I was disappointed we didn't get your Minecraft chart at this past November Analyst Day like we did 2 years ago, but maybe that will be at the next one. But doing something that...

Thank you, Scott, for being pretty much on top of this.

Fair enough. But doing some quick math, I think the growth in your $1 million plus ARR segment represents almost half if not more of your organic revenue growth over the last 2-year period. What's driving the strong success specifically in the segment?

Yes. I think back to -- we discussed with our shareholders 3 years ago, a strategy for Five9 to invest in the Enterprise segment. So there was a plan that we put in place to invest in both product and innovation to expand our portfolio. So if you're going to go after these larger enterprises, you need to have a complete portfolio. That's number one. So we invested in R&D pretty significantly. We also invested in public cloud, so that we could easily drop new data centers around the world where, if you think about the Fortune 50 or Fortune 500, these are all multinational companies and they've got agents around the world. So our ability to sort of show up and respond to their RFPs and drop data centers wherever they are in the world, that's been one of the keys to success here. The second thing is that we invested in our go-to-market machine. So beyond technology, it was also about go-to-market and our partner investments. So we hired a veteran partner leadership team, now it's today led by Jake Butterbaugh, who formerly led the partner channel for Cisco's collab business, who have really invested and built up that partner channel for us. They are very, very critical for winning these large customers. I would also say our investment in systems integrators has been absolutely key to winning in the Enterprise and Strategic Enterprise because there are very few large companies who make these kinds of strategic transformations without engaging outside help. And so the investments we've made there are really paying off. So you can sort of say that we've executed well on our strategy. We've also been helped candidly and been lucky by the fact that I think the natural tendency of these markets is that the largest companies wait for the smaller companies to try these technologies and say, hey, is that working? And so what we've done is sort of continually grown the size of company that we've been able to serve and the investments we've made on the technology side have made it so that we now can serve contact centers with tens of thousands of agents. That's not something that we could have done 3 years ago. And I think as you see, those customers really do wait for reference customers. Nobody wants to be first. They want to see a contact center with 1,000 reps go first and then 1 with 5,000 or with 10,000. So as we've consistently knocked those milestones down, it's opened up more and more of these doors to these strategic enterprise customers. And there's not a single customer in the world that I could think of today who probably isn't considering moving their contact center to the cloud as a next step, maybe not this year, maybe not next year, but in the near future.

Sure. Kind of continuing on that line a little bit. You've made a lot of investments around data centers and you talked about partner environment. You've had the other benefit with both of those being critical to your expansion in Europe and EMEA, in particular, right? Because the internal data center operations probably want to scale quite as well there and of course, those customers buy mostly through partners. Is that strategy something that you expect to further capitalize maybe internationally going forward in other regions that you have not had as much success in yet?

Yes. It's been a clear investment area for us and you've finally seen us start to really move the needle internationally as international grew as a percentage of revenue, sort of outpaced the rest of our business, which we need to happen because we're still, I think, is it 9% Barry, 10% international revenue contribution?

9%.

9%. So we still have a lot of opportunity there. And we now have got over 100 people in EMEA and we're really going hard at the European market, taking a focused approach, looking at Germany and sort of taking -- we started in the U.K., we've now targeted Germany. We're expanding in a very sort of focused approach as opposed to kind of going broad and shotgun approach over the whole continent. So we're now doing that and there are big opportunities where it's just an untapped market and we haven't even really scratched the surface with APJ yet. So lots of opportunity to grow this business internationally. And especially in a world where at least half of the contact center agents in the world are located in companies that are headquartered outside the United States. So it's an important part of our growth strategy. And you can see that in our proxy by the way as the international growth ends up being much more of a contributor over time in the next 5 years.

All right. Look forward to seeing that play out as well. Moving towards, you mentioned the word proxy. Let's talk about the growth rates that were in that proxy. So probably the most frequent question I've had over the last 90 days, and I'm sure you have as well, is the growth rates in those documents and the company's ability to meet or exceed them. You all sounded generally comfortable with them at your Analyst Day. Barry, you characterized it as a solid 50% shot towards a [indiscernible], but a higher multiyear growth rate at a higher scale often seems like a big mountain decline. I broke it down as more partner impacts, new global traction, stronger upsell, cross-sell, especially within your larger customers, as that will become more impactful as bookings mix continues to favor those large customers and maybe a market that's more ripe for cloud contact center. When you boil all that together, that suggests that those assumptions are not behind the sky assumptions, that those are actually reasonable. But what am I missing from this ingredients list to help drive that higher growth rate at scale that you're looking at today?

Barry, do you want to take that one?

Yes. Scott, you've summarized it extremely well. Just to kick it up maybe 1 notch, as Rowan mentioned a little while ago, he's been saying as the CEO for many years now that we see durable, sustainable LTM enterprise subscription growth in the 30s and we've delivered that; the most recent quarter was 51, which was on the high side compared with the 30, but still there you have the number. So that continues. Another key factor is our installed base. It's gotten bigger. If we rewind the movie, say 3 or 4 years ago, the contribution from new logos would have been about 2/3 and from installed base, about 1/3. Now it's closer to 50-50. And we also have the benefit of, within that installed base, these bigger customers who tend to expand further. And that's why we've said that with fluctuations, the 123% LTM dollar-based retention rate that we reported last quarter will gradually increase into the high 120s. And we feel pretty comfortable about that. We have the benefit also potentially down the road. We just don't know exactly when and how much of ARPU increases, especially from the AI and automation parts of our portfolio, which are doing so well. And then thirdly and finally, Scott, is the fact that we have international that we were talking about a moment ago. We're 9% right now and we can see that growing into the mid-teens. If we grow higher, but the domestic business is growing pretty healthily as well at the same time, sort of slowing that proportional growth in the international. So you put that all together and the things that Rowan mentioned earlier on, and our direct sales force is doing extremely well and it's now being very well complemented by those partners, which are helping both domestically, but especially internationally.

So Barry, that booking shift is, it's kind of a big shift, right? 1/3 of your new bookings historically came from customers, now it's closer to 50-50. How do we think about that change? Is it more driven by customers just buying more seats or they buying more modules to move that number?

Currently, it is mainly more seats and that will be the case for the immediate future. But given the huge disparity in terms of the AI part of the portfolio, when we're talking about the IVAs, which have gotten such prominence, especially with the bigger customers who can afford it and we have this imperative of hiring more people, which is difficult in this environment, we're talking about $400 a port on average and that's approximately twice what you would pay for a live agent.

One of the things that we've done to go after that, because we are just scratching the surface in the installed base, we've created a dedicated overlay sales team, put one of our top sales leaders into that position. If you go back in time, Scott, just to give a little credibility to the sales leader, remember when I started, our commercial business was -- we were projecting that to be flat. Of course, embarrassingly, it broke our predictions and ended up growing at 30%. The person who really made that happen, one of our top sales leaders, has now moved from that commercial business and is leading up an installed base and net new sales team focused exclusively on AI and automation. And so look, we're organizing to go after this opportunity and drive that installed base traction up on the automation portfolio.

Fair enough. Last question for me before we get to audience Q&A, because I do have a couple of questions here that have come in, is we've talked a lot about product, but the company's spend on R&D as a percentage of revenues has actually declined to 12%. The comment sounds a little bit tongue-in-cheek because 12% sounds great, especially with all the innovation that's come out of the company, but are you spending enough on R&D, because pure SaaS companies with similar growth rates tend to spend closer to 20% plus or minus than 12% of revenues.

Well, this is slightly embarrassing for both Barry and I, but we did make -- as you know, I think 3 years ago, we came out and provided new estimates in terms of R&D growth. And we did drive several percentage points of increase in R&D growth and that story really still remains the same. I think what changed though is 3 years ago, 4 years ago, we were growing in the mid-20s. But as we expanded and coming out we kind of -- we're obviously in the quiet period, can't talk about this last quarter, but for the last year, that growth has really accelerated and so we've had a hard time keeping up on the R&D hiring front. I mean that's the truth. We also have been leveraging low-cost engineering sites and so you have to hire even more head count. So look, when the company is growing in the 30s and 40s, it does get harder to keep up with the revenue growth and that's the embarrassing part of that. But I think fundamentally, yes, we do need to spend more in R&D. I'd like to see that percentage of revenue. I mean, I think we're in the right band. We don't need to be at 20% as some of our competitors are, but we would like to see that number being slightly higher than it is and we have pretty aggressive hiring plans for R&D in this next year and leveraging some of our low-cost sites now as well to bring in new talent. So that plus remote hiring I think is going to also help as we're able to access talent in more parts of the world.

As a side note on that is the remote talent, given the environment that we're in and how companies are more comfortable hiring remotely, does that actually give you a chance to keep those costs a little bit lower over time because you don't have to hire so many that are in the valley, which tends to be a little...

Yes, we had this discussion and it turns out that at least the folks who we're interviewing throughout the U.S., it's pretty much valley rates everywhere throughout the U.S. I mean we haven't found that there's much cost savings to be had just by hiring remote folks. And look, we're looking at the top, the creme de la creme of the engineering sort of ranks. We're trying to find those very tippy top kind of most senior employees. We're not talking about junior entry level. So I certainly think over time you definitely should see the effect of being able to reduce costs by hiring in lower-cost locations in the U.S. But I don't think the market has had time to adjust to this. And so we're sort of fundamentally, as we hire remote employees, we are paying just kind of the same rates, not really helping us from the cost perspective yet. That might change.

All right. Let's take the first question from the audience. I'm going to add in a couple of words here, I think, and wordsmith it a little bit. But can you elaborate on your, it looks like, upsell or cross-sell opportunity, maybe the quality or potential of the existing installed base? So I don't know, maybe the penetration rates are $1 and you think you can take them to $5, maybe help characterize that a little bit. Then the second part of the question is, what's the timeframe for a typical land and expand process? Is it 2 to 3 years?

For a land and expand? Yes. Okay. Well, maybe I'll take it in reverse order. So the sales cycle, roughly speaking, you've got 6 months -- for a typical enterprise, 6 months to a year potentially for those typical sales cycles to kind of go through the RFP and close, and closer to a year for the largest customers, sometimes a little more. And then from there, you have 6 months plus from a deployment perspective, 4 to 6 months. So we are known in the industry as one of the easiest to deploy, and the most successful sort of deployments come from Five9. So we try to get those customers going as quickly as we can. But you're still talking about, let's call it, 18 months on average for a sales cycle plus a deployment cycle. And then, once they get to the cloud, then the opportunity opens up to start to sell them more. Now what we've seen recently, especially as we move into the large enterprise, is that they're actually taking more of the portfolio upfront as opposed to sort of, hey, just give us the minimum and we'll get started and then we can add more. We are seeing more of these large strategic enterprises who are saying, no, give me the full boat when I start. And so that's really helping us right out of the gate to get them on all of the technologies. That's WFO. That's our IVA technology, which has, I think grown in adoption at least 5x sort of growth of that IVA technology. So we are seeing more upfront adoption. But then following that, that new motion we're now putting in place to go and mine the base, as we grow that base, that should really start to kick in. And I think probably that 2- to 3-year estimate is roughly right to see those customers start to adopt new technologies. So I think it's a lot to bite off to move your on-prem to the cloud. And so getting them on to the base technology, set, and then coming back in with the add-ons that we can bring those customers is going to be a really important key motion for us over the coming years.

Got it. Next question is, what type of uplift do you get when you sell an IVR bot? Barry, I think you talked about that. The pricing is about 2x your typical seat today, 400 versus 200. But the second part of the question is, and how much of the success was offset by labor replacement and less agencies?

Barry, do you want to take that?

Yes. So to be perfectly candid, we don't yet know exactly how much is replacement versus reduction through deflection, as Rowan was talking about earlier on. We did do an analysis and assumed, somewhat arbitrarily, that 10% of a, say, 100-seat contact center becomes automated rather than live. In a scenario like that, it's a win-win situation for our customer and ourselves. We get about a 10% uplift in our ARPU and the customer gets about an 8% cost reduction, taking into account the fact that they spend $2,000 a year on labor versus $200 a month in the technology that they paid for us -- Excuse me, $2,000 a month, not a year, versus the $200 a month for technology.

Next question, which might be all that we have time for is on acquisition philosophy. The company has executed predominantly smaller acquisitions historically. Would you look at doing something on a more transformational basis?

I think all options are on the table for us. As we think about the market that's currently shifting over to the cloud, there is a desire to buy more. We have really done a great job over the last 3 years of expanding our portfolio. So the playbook we put in place was look, we have to buy into these near adjacencies, which we did. We did 3, I would say, sort of smaller acquisitions, and those have been incredibly successful, the most recent one being our IVA. And so I think we've built up that muscle for inorganic development. I also am much more of a build it, if you can, kind of a person. I'm an engineer, so I like organic development. It certainly has higher returns and reliability for investors. And acquisitions are certainly always high risk. So we always prefer to build stuff than buy it. But if there are assets that are out there and we can leverage our position in the marketplace to make bigger moves and those things make sense, we certainly would do it, but no immediate plans on that front.

With that, that's all we have time for. I want to thank everyone for joining us. Rowan, Barry, thank you so much. Good luck with your meetings today and we will chat on your fourth quarter call.

Thank you, Scott.

Thank you.