SKLZ - NYSE

Good afternoon, and thank you for attending today's Skillz Second Quarter 2022 Earnings Call. My name is Jason, and I will be the moderator for today's call. [Operator Instructions]. I would now like to pass the conference over to our host Charlotte Edelman of Skillz.

Good afternoon, and welcome to the Skillz Second Quarter 2022 Earnings Conference Call. I will proceed shortly by reading our forward-looking statements and non-GAAP measures immediately followed by brief introductory remarks and then question-and-answer session. Hosting the question-and-answer session today, we have Andrew Paradise, Chief Executive Officer; Casey Chafkin, Chief Revenue Officer; and Ian Lee, Chief Financial Officer of the company. We hope you've had a chance to read out our press release, which we published earlier today and which is also available on our Investor Relations website. Some of management's comments today will include forward-looking statements within the meaning of the federal securities laws. Forward-looking statements, which are usually identified by the use of words such as will, expect, should or other similar phases are subject to numerous risks and uncertainties that could cause actual results to differ materially from what we expect. Therefore, you should exercise caution in interpreting and relying on them. We refer you to the company's SEC filings for a more detailed discussion of the risks that could impact future operating results and financial condition. During the call, management will discuss non-GAAP measures, which we believe can be useful in evaluating the company's operating performance. These measures should not be considered in isolation or as a substitute for our financial results prepared in accordance with GAAP. A reconciliation of these measures to the most directly comparable GAAP measures is available in our second quarter 2022 earnings release. With that, I'll turn the call over to Andrew for some brief opening remarks. Andrew?

Thank you, Charlotte. Good afternoon, everyone, and thank you for joining us today to discuss our second quarter 2022 results. Before we begin taking your questions, I'd like to share a few thoughts. We entered 2022 with a change in strategy to focus on profitable growth, improving our efficiency and driving greater impact on our player and developer experience through product-led initiatives. I'd like to share my thoughts on what's going well so far, what is not and what we're going to do about it. Let me start with where it went well. In Q2, we continue to make progress on reaching profitability. We reduced our net loss by almost $90 million while reducing RAEM revenue after engagement marketing by only about $8 million. We lowered our used acquisition marketing spend by 49% quarter-over-quarter, and we also made progress in improving the efficiency of our user acquisition. This is a business that was running 6-month payback periods at the past days of our IPO. And we plan to systematically march our paybacks back towards those levels. As such, in Q2, we continue to reduce low-return engagement marketing initiatives that aren't yielding or do not have indicative data that they will yield, and we executed a restructuring of our workforce to better align our resources with our changes in strategic priorities. Overall, we reduced our net loss and adjusted EBITDA burn rate. We improved the quality of our revenue by increasing revenue after engagement marketing as a ratio against revenue. Over these last 3 months, we didn't make the progress we wanted to see in product development, but we still had a few wins, which I mentioned here more for the sake of balance as opposed to anyone thinking we're pleased with the progress this past quarter. We launched a public beta of our cloud gaming technology initiative as we discussed we would in Q1. We launched a public beta of the refreshed core user experience. Okay. Let me go into what didn't go well. We're positioning the company for profitability had a real cost. Revenue after engagement marketing, paying monthly active users, all fell sequentially quarter-over-quarter. We also have areas where we can and will get better. Several that I want to highlight here include our H1 product initiatives did not achieve our desired results yet. In fact, we've had slower product development velocity than we would have liked or anticipated. We've had too many new product features that aren't driving LTV and sometimes are even detrimental. We've had too many new experiments split tested that didn't yield LTV growth and also have sometimes been detrimental. On the marketing side, we still have more work to do on improving user acquisition efficiency. Moreover, we have more work to do on reducing low-return engagement marketing initiatives. We're raising the bar across the board on any and all experiments in marketing and product to grow our business. It's a systematic decline in quality of work that I'll personally address in product development. The primary causes of the sequential decline in revenue and revenue after engagement marketing was really lower retention for some of the mature cohorts of users on our system. And we think we've now caused that. The main drivers of this are a mix of instances of users cheating us on our platform and past product modifications that shouldn't have rolled from our test phase to full rollout. So when I speak of cheating, let me give you an example. We've seen instances of users cheating on the platform, including abusing system discounts and incentives in other countries such as in the Philippines, where users are abusing our friend referral program to get financial incentives without delivering real user referrals. So, here's what we're going to do next and why? We remain confident about the potential for product-led initiatives to drive a better experience for players and developers. It's frustrating especially the progress in product lately with so many opportunities to improve consumer and developer experiences on our platform. As I mentioned earlier, when we approached the IPO, we had a 6-month payback. I plan to get us back there. And the first step to fixing our paybacks is I'm returning to our product development to fix our products to help build superior paybacks and a superior LTV to CAC profile. This work has already started. There is a lot more coming in the days ahead as we fix and right Skillz business. Repositioning the company for profitability is far pretty easy from where we were ending last year. And as such, we need to lower our full year 2022 revenue guidance to $275 million. We're going to share more detail on all the things that are happening in our business in our Q2 stockholder letter, which will be published tonight. While the scope of our ambitions and the driver team as always leaves us wishing we had accomplished more. We did make some progress towards our long-term goals this past quarter by rationalizing our cost structure. There's still much more work ahead to do, and I'm more determined now than ever to do it. Our focus remains on a path of durable growth and efficiency in the years ahead. This means we're going to continue to reduce costs. We're going to continue to drive improvement in marketing efficiency and we're going to increase the velocity of new product development. It's incredibly exciting to think about the long-term opportunity for this business ahead to transform the existing $92 billion mobile gaming market through competition. It's truly a huge opportunity, and we're steadfast in our belief that through competition, we can enable everyone to achieve their greatest potential, both the player and the developer alike. I wish I had better news to deliver today. I can only tell you that we are pushing very hard to execute the transition strategy that we've outlined last quarter, and we will outline again this quarter. And I thank you for your support, your patience and your understanding now more than ever as we continue on this turning this path. With that, let me open up for questions.

[Operator Instructions]. Our first question comes from Jason Tilchen with Canaccord Genuity.

Just going back to the guidance, and I think you obviously, you laid out some of the reasons why the results came short. But just thinking back to a quarter ago, reiterating the revenue guide of $400 million and then just 3 months later, reducing it by nearly 1/3. Just wondering if you can maybe provide some additional color on what maybe is different now versus 3 months ago? And what gives you confidence that as you laid out in the press release, you expect for, obviously, users after the decline in the next 2 quarters to resume growth next year? Just wondering if you maybe touch on that as well.

Absolutely. So first of all, thank you, Jason, for the question. We're lowering the guidance because of the user retention trends we've seen on the system and the slower-than-anticipated velocity of new product development and the delay in achieving the expected results from the recent product launches. We're also assuming that the macroeconomic environment is going to be challenging throughout the rest of the year.

Just going back again to next year, obviously, expecting a resumption of growth there with a much lower base of marketing spend. Just curious if there was anything specifically you can call it in the product pipeline that would maybe give you confidence that you're going to be able to achieve that resumption of growth as we head into fiscal '23?

Yes. The primary driver from lower retention from the mature cohorts. First part is we -- with going public, I don't think we quite anticipate the level of attention the company would get that attention is also landed on the cheating and fraud community. And we've seen a lot more bad behavior and bad actors entering the ecosystem. So users cheating on the platform, I think I mentioned 1 in my overview, but our friend referral program, seeing a lot of abuse, particularly from countries outside the U.S. attempting smoothing and/or actually executing in their home country to literally try to steal money out of the system. We've remediated the issue in countries like the Philippines. We're continuing to identify and address other instances of users of spikes in cheating and fraud. The second piece, I think I mentioned past product changes over the last 18 months that really have not been landing the level of LTV that we anticipate. We have multiple different initiatives there that we're modifying that I think give us confidence that we'll get back on track both on retention, engagement and monetization. If you think about Skillz as a platform, trust is a very important part of our platform's promise. I think trust and safety and almost any marketplace-type business is really important to how it operates. And you have to look at the cheating and fraudulent behavior as damaging not only because that revenue is not real. But even beyond it, the people that are being defrauded and cheated, it's damaging their retention and engagement on the platform. So I think to summarize, I'd say it's very top of mind for us to get this under control immediately.

Our next question comes from Drew Crum with Stifel.

So your outlook commentary assumes further declines for paying mouth through the balance of the year. What type of decline are you anticipating? Should it be similar to what you experienced in 2Q or something different? And Andrew, is there a catalyst you have in mind to reaccelerate this metric next year? And then I have a follow-up.

Okay. Thank you, Drew, for the question. I'm going to turn it over to Casey to talk about Marketing growth.

Thanks, Andrew. Jason, the -- or sorry, Drew, the way to think about this is as we extract some of the discounts from the system, we are -- we have a smaller number of people who are depositing low amounts of money or playing with system incentives who have deposited in the past. And so as we pull those incentives out, that becomes the largest contributor to the decline in paying monthly active users. And we're not quite done reducing our engagement marketing spend. So ostensibly, what we're doing is becoming more profitable on a user-by-user level across the business. And so the way to think about that and the confidence in the reacceleration of growth is after those incentives have been completely pulled back, the retention profile of our cohort -- each cohort never actually dies. We don't see that cease playing at all on the platform, they decline and then level out. As we pull out some of those incentives, we're pulling out low or no value users who are still counted and are paying MAU calculation. And so the way to think about that is continuing to decline at a reducing rate in the back half of the year. And then as those incentives are completely pulled out, adding -- just adding new cohorts will drive the user growth in the future.

Got it. Very helpful. And then my follow-up pertains to the adjusted EBITDA margin that's implied in your guidance. It looks to be the same in the second half as the first half. And with the planned reduction to engagement marketing spend and presumably some savings from your reorg, what prevents you from showing margin improvement in the second half?

Ian, would you like to jump in on that question?

Yes. Thanks, Drew. I think the focus here is really, as we mentioned here, we are really focused on continuing to optimize around the marketing spend. We took the restructuring action this quarter, I think, continue to just focus on overall operating expense reduction in the second half, so I expect that to continue. Obviously, with the declines in the top line and revenue after engagement marketing, that is, I think, going to be, I think, where that nets out. So I think we'll see what happens in the second half, again, with the reduction in operating expenses very clearly that's focused on, again, that may be netted against the declines in the top line. So I think that's where the guidance came out, Drew.

Our next question comes from Jason Bazinet with Citi.

I just had a quick question on those 6-month paybacks. You talked about sort of at the time of IPO. Is it fair to say that it's easier to get a 6-month payback if you're running a business that does $50 million of revenue or $100 million and then if you're doing $400 million of revenue. And the reason I ask is even if some of the initiatives don't work, is there some natural improvement in your paybacks that should occur just as the revenue gets smaller?

I think in many businesses that can potentially be true -- all sorry, I don't mean to be rude, thank you for the question. I appreciate it. I appreciate the engagement. I think that can be true in many of our smaller markets. And I think that question kind of leads to the consideration is the TAM being dried up, if you will, and that's why you're seeing worse paybacks. I'd point out, we were, I believe, $230 million of revenue -- net revenue. And so not that all scale. And as we scaled upwards, we would not say that is what we've been seeing, basically tapping out the market driving our prices. I think for a lack of a better way of putting it, we have been tripping over our own feet on scaling human capital in COVID. And it has resulted in underperformance in a variety of roles across the company, where we are working very hard on getting the right people in the right roles and a strong understanding of how the system works so that they are accreting value against LTV as opposed to detriment it. And I wish there was a better way -- better reason this. I guess the 1 good piece here, I think, is that we have no reason to believe from everything we're seeing that the TAM has been consumed. Casey, do you want to talk a little bit about marketing and where we see TAM?

Yes, absolutely. And Jason, I think that the behavior that you're talking about is not uncommon in many businesses, though in the case of our business and particularly when Andrew was running product for us and the lead up to our initial public offering, what we saw was that our cohorts we're becoming sequentially more and more valuable, which is the opposite of what you see in many businesses and is evidence of the real network effects that are observable in this space. The challenge in the last year to 1.5 years is that we haven't seen that. And we don't believe that, that's endemic weakness in the system itself. So much as operational opportunities that we can and will correct as Andrew has mentioned, the gain, we called the tripping over our own feet. But I think getting people together for in-person collaboration, improving the analytical rigor behind the new products and future rollouts that we bring to market. These are -- these become operational opportunities for us. And so I think it was -- I believe it was -- actually it was another Jason, who asked the question about confidence in the future. The confidence comes from seeing as we make those changes, we can see the data secure some of the new cohorts that we're bringing on board.

And if I can add, I think one of the things Jason with our business is we were an entirely in-person culture, a more than 9 to 5, by absolute in-office culture as the COVID approached. And we scaled from 190 people to 750 over these last 2 years to support the revenue growth and to be able to grow from $384 million revenue year last year on to $1 billion. I think what's happened is we've really seen the difficulty of scaling 5x or 4x, call it, people in culture. As we've guided on a lot of people remotely, and it has not been smooth scaling in COVID. And to that end, the immediate action we've taken is an RTO plan and execution of getting all of our people back in offices 5 days a week. I wish I could tell you that our -- everyone on our team agreed with that. We've had a fair amount of turnover and dissatisfaction with getting people back in person to work. It's been challenging. And I talked to a lot of other CEOs of other companies, and I am hearing similar statements across the board about the transition from a work-in-home environment to an in-office environment. But I really do believe that for our level of creativity and our type of company to generate the kind of results that led to our IPO, we need to get people back together. It's just too difficult to teach people about a novel system and a novel business area that we're building remotely. We can't achieve it fast enough or at a quality level that works for our business. So we've really taken what might be considered a dramatic action of going immediately to RTO at 5 days a week in office.

Our next question comes from Clark Lampen with BTIG.

I had a couple of questions. First, I wanted to go back, Andrew, maybe I guess, the point around cheating. You mentioned that you wanted to sort of address the issue and instill like us confidence amongst the, I guess, if you will, cleaner player cohorts on platform. And I wanted to see if you could elaborate on maybe the ways in which you might be able to do that moving forward? And then there was a comment in the release about sort of growing profitably in '23? And I wanted to see if you could be a little bit more specific on whether that means we should expect 1 quarter of profitability, maybe sort of improving incremental margins put a finer point around that, if possible. And also, as we think about Skillz returning to a place where you are growing at the rate you hope to and seeing the payback period, you'd hope to, what sort of capital investments need to be made in the short run. I think you talked about in the prepared remarks, one of the hiccups shorter term being product development, I guess, sort of efficiency and also the quality of products that was coming out of the development team, What, I guess, can you guys do or what do you need to do to get that back to a viable level going forward?

Sure. So, I think first, we try and recap that a little bit for everyone who's listening. First, cheating, can we elaborate and what we're doing to mitigate. Second is profitability. What to expect in 2023 and beyond. And third is capital investment to build the products that we need to be successful. So let me hit cheating first. There are a number of different techniques to increase the level of identity around a user to ensure that it's more difficult to duplicate accounts and defraud the system, which is actually one of the primary ways we've seen cheating and fraud is some level of duplicate account, trying to reset your player ratings on a given game, handing your device off to another person on a duplicate account but kind of think about like multi-account fraud. That's one of the top things, so we've built out an identity team that is implementing a number of things, but everything is kind of industry standard like a 2-factor SMS authentication. So, forcing that we know it's a real phone number, not a scooped account on a laptop, I can go on and on. But there are lot of ways that we are forcing identity down to a single actual person as opposed to someone being able to more easily pretend to be multiple people. That's one way that we are enforcing and changing cheating and fraud capabilities of the system. There are many others in the just some time, I'd be happy to chat more offline on the things we're working on that if there's interest. Maybe second, I can hit profitability, what to expect quarter-by-quarter. And let me ask Ian, if he'd like to talk about the numbers and are where we're looking for 2023 and 2024.

Yes. Thanks, Andrew. I think the key is here that we continue to look at focus and improve to reduce costs and improve efficiency across all our clients beyond 2022, obviously into 2023. We obviously haven't given a specific guide to that yet. But the goal is still, as we've discussed previously, to reach breakeven by the end of 2024. So obviously, that means continued improvement through the course of next year. But we haven't, again, provided more specifics on the particular margins yet in 2023.

And then third, on capital investment that we need in the short run already represented in the financials that we put out. I think we are very determined to get our business on a meter beat culture in terms of our financials, one of the most disappointing things for us. I think it has been moving off of that and missing revenue forecast. I will point out that one of the things we did for the first half of the year is we outperformed on cost and what we said we would spend for the first half. And I think the first steps that we are looking at is we can control our costs. We can't always control if our investments into product development and experiments yield. I can tell you that the capital investment that we are making in the product development, I am personally going back into overseas at a much closer level and we'll be outlining more details on that in the coming days. But with the level of investment we're making in product development and my background in building new products, I feel like it's definitely one of the best uses of my capabilities for the business. Anything you would like to add, Casey?

Just, Clark, on the question about cheating. I think it's a helpful clarification that Andrew is bringing up a I think people often think of cheating as things like hacking or hacking the game to gain an unfair advantage, whereas the -- what we're talking about in this case is, is something that we believe is much more controllable in the payments industry will be referred to as friendly fraud, but this is people using discount codes, incentive marketing programs, things like the friend referral programs. And so creating geographic restrictions on those programs. making those programs more targeted. These are things that we're already doing to reduce the expenditure in what ultimately becomes unprofitable revenue streams. And likewise, the kind of cheating that we're talking about, the most common form of cheating is people trying to gain more player matching system and identity and they ensuring that it's -- we're truly at 1 user per device per account. And an account per person is something that we can very much control. And that will improve both the quality of our revenue and the quality of our cohorts on a go-forward basis.

And I was just going to add on to what Casey said, which is I do -- I would point out that the per account cheating and fraud technologies that we built are actually working quite well. But multi-account and spoofing is a different type of attack. And clearly, our defenses are not strong enough, which, of course, we're rectifying immediately.

Our next question comes from Brad Erickson with RBC Capital Markets.

I guess first is on the path to profitability, maybe setting aside the cheating commentary and all that. Can you just talk about some of the new products or maybe characteristics even if new products you're working on where you think you can drive this better retention over time?

Sure. Thank you for the question, Brad. So in terms of how we're changing and improving retention. I think, actually, one of the things we're doing first is we're reverting and removing experiments and product features that the data does not support their broader rollout. So it's not actually building new products. It's product features and experiments that were rolled out that we really were really not supported by a deep data analysis. So rolling those back actually is already showing improvements in LTV. I hate to say we're paying to undue work that we paid to have done, but that is the reality of what we're seeing and what we're doing first. I think when I left a very direct engagement on product development to really focus much more on taking the company public as CEO, we have not been able to invest our dollars into product development as efficiently as we used to. New features that we've been working on, such as social, such as improving our lease technology given the scale we're at now, such as refreshing our core loop of our product experience, the execution of each of these things is as critical to improving LTV as actually building these things. It's -- I hate to say it's such a detail-oriented problem, but it's a double-ended details problem with product development, where I think we are missing some of the key product needs in each of these features, and they need to be revised carefully to generate the level of LTV that we'd expect from these features. And I'm in there and doing it with the teams. And we are solving the problems on, I would say, somewhat back to the basics product development that needs to happen right now in our company. And I think the best way to think about that is our plan right now is to return our cohort strength to where we were pre-IPO in March 2020. And I'm -- pre-pandemic, pre-IPO cohort strength is what I am moving us towards. And I think that's an achievable goal in the next 12 months, but it's quite a lift from where the product has evolved to.

Yes. Got it. And then maybe a question for Ian. Ian, I think the filing isn't out yet, but I guess you guys spend every quarter with maybe $4 million or $5 million on the balance sheet of customer funds? And then I think it's something like 80% or so of the GMV comes from existing customer accounts. So I guess that implies the customer balances turnover like something like 100-plus times a quarter. Can you explain how that works, how that math works, anything we may have wrong in there?

No, I think that's -- I think you -- appreciate the question on that one. So I think we discussed a little bit before that people are playing, obviously with new funds. They're playing with funds that are already in the system as well and prior winnings and entering into more games continuation from that. So the funds, as you mentioned, there's a fairly limited balance on the balance sheet, as you mentioned, that $4 million to $5 million because people tend to deposit but then actually use that funds to go play games. So again, they're using that plus prior winnings on the platform to enter and continue playing games. But I'll open up again for Casey, Andrew, to give you more, you want to add on that.

Yes. So Brad, the way to think about it is not so much of the turnover of funds multiple times so much is the fact that players today are not using Skillz as a long-term stored value account. So players aren't putting a lot of money into their Skillz account and then leaving it there for a long period of time. They're adding money to their account and then consuming that through the competitions they play in to then adding more money to their account, consuming that through the competitions that, that they play in. And it's something that people have actually highlighted and we believe is one of many long-term opportunities for the business, is having people hold on to a larger account balance inside their accounts over time because, of course, then we could make money off of that money in the same way that insurance companies do. But the way to think about it is that people are depositing money with their credit cards and playing through it in real time.

Got it. That's helpful. And then maybe 1 follow-up, and I apologize I jumped on the call late. I missed some of Andrew's opening comments, but on this cheating issue, I guess can we take any part of the guidance reduction as a proxy for how much of the business that represented? And I guess maybe percentage-wise, sort of how far are we do you think through this, call it, a cleansing process and when you think that might be completed.

I think we are not done with cleaning. I do think we have a very good idea of the size of the problem and how to deal with it. And I think the first step we always engage in is size and understand the problem before we try to solve. And as such, I think in the guidance we're giving out the revised guidance, we are accounting for removing everything back down to a normal level of -- there's always some level in any business of bad actors. And you just need to be managing it below a certain threshold and our threshold is far too high. And I think if you take a retail business, for example, the shrink in retail goes over 1% you may bankrupt a store. And I think we are looking at our business right now, and there are a number of bad actors. And not only that, they're disrupting the experience of the good ones, of the good paying customers. So I think you have 2 effects in solving this problem of duplicate account and duplicate account fraud and cheating, which is you can both remove what is unprofitable revenue. And simultaneously, you can improve the retention and engagement of your good customers who are actually just trying to shop in your store and use analogy. And so I think it would have been great if we could have caught this faster. And certainly, the next step after solving this problem is improving our systems for faster detection of this kind of issue as we scale upwards and onwards after we finish cleaning.

[Operator Instructions]. Our next question comes from Ed Alter with Jefferies.

I had -- wanted to dive a bit deeper on the payback period kind of where exactly that's coming from to sit on the CPI side, where I know that's been challenging in the last year? Or is it more on moving the spending itself back to those [indiscernible] faster?

Ed, thanks for the question. I think I heard you, it's a little bit challenging to hear you, but I think you said -- you asked -- your questions were on the payback period and where exactly is the opportunity? Is it lowering CAC or improving LTV, if I'm stating it correctly?

Exactly.

Actually, we think both. So we've been cutting back inefficient user acquisition programs. We are now actually at the best CAC we've seen in years. And I think we will continue to get to better and better CAC. The LTV decay has primarily been from a disruption in retention. And we -- I think we've talked a fair amount about cheating. We've also been talking about product features that have led actually, unfortunately, to lower LTV as opposed to higher where there is data integrity around the testing of the product features that led to a test feature being rolled out. And so what we're doing is we're looking at any feature, any program that was rolled out as we've seen the decay in LTV, and we're basically rerunning the data and verifying whether or not it is a valid good change to our products. And that's enabling us to quickly move our LTV back to where -- to levels that we were very proud of as a company as we are approaching our IPO. So I think there's opportunities on both sides. So Ed, I really appreciate that question. And Casey, if you'd like to comment on either particularly on CAC, I think, as well as your business that might be viable?

I was actually going to say just that that's what is particularly exciting for us at this moment in time, Ed, is that our CAC is around our local low point from -- that we haven't really seen since the beginning of COVID, and we still think there's a big opportunity to drive it lower through app store optimization through investment into organic channel development and continued media optimization, we think we can bring our customer acquisition CAC even lower. And on the LTV side, while the development of new features, such as leagues, continued investment in social is the things that Andrew in about represent attractive upside that we still very much believe in. Simply getting back to the user programs that we were running about 12 months ago will give us an increase in LTV and move that return on investment calculation and payback in.

There are no further questions waiting at this time. So I'll pass the call back over to the management team for closing remarks.

Well, thank you, everyone, for joining our Q2 2022 earnings call. We very much appreciate your support. And as always, we will be vigilant on building towards the future of competition. We'll talk to you next at our Q3 earnings call until then, thank you.