Yelp Inc. (YELP) Earnings Call Transcript & Summary

Okay. I think we're going to get started with our next session here. Thanks, everyone, for joining. It's my pleasure to introduce Yelp. We've got David Schwarzbach, to the CEO -- CFO of Yelp. So I'm sorry. You know I need my coffee. David, I'm sorry. CFO of Yelp, not to undercut Jeremy. CFO, Yelp. David, let's start with -- you're going to start with this statement first. So you go first and then we'll get into it.

All right. Thanks, Eric, for having me at the conference. We'll be making some forward-looking statements during the conversation today that are subject to risks and uncertainties. Please refer to our SEC filings for more information on the risk factors that may affect our results.

Thank you, David. Okay. And now we can get into it.

I think one of the big picture issues we were talking a little bit for a few minutes offline before we started the conversation. I think last quarter, what we saw was a lot of volatility in the broader digital advertising landscape. Macroeconomic conditions remain very volatile. Probably the number 1 topic here at the conference has been the macroeconomic environment. Talk a little bit about the positioning of Yelp's digital advertising business. And how you, Jeremy and the team think about the positioning of local advertising, Yelp as a platform when operating in such a volatile environment?

Yes. So a few thoughts there. And it's probably worth just stepping back for 1 second because I think the strategy plays to the answer. And as folks may already be familiar over the past several years, we did set the strategy for ourselves to focus on 3 areas. One was to really acquire and expand advertisers through our most efficient channels. That's both national accounts. We call it multi-location, it's also mid-market and also increase self-serve. So that's one component. The second was really improving monetization of our services traffic. And the third may be self-evident, but very, very important is deliver value to advertisers. At the beginning of this year, we did add a fourth area of focus, which was around the consumer. But just going back to the Q2 results and maybe some of the separation between different ad platforms and why we think there was separation. There's certainly a component of it in that we do serve the local economy. Obviously, we think that our trusted content is highly relevant to consumers. And we think that businesses want to reach our audience, and our audience is more affluent. Nearly 60% of visitors to the site come from households with income over $100,000. So you always have to start, I think, with as an advertiser, who do I want to reach? And that Yelp Audience is not just affluent, but they're also high intent, and they're bottom of the funnel. So the second thing is how are we interacting with them? What's our content versus the content on other sites? Our content is around that decision-making. It's lower funnel. And it's about the query. The person is actually telling us what they are interested in. I need a locksmith, it's 2:00 A.M. Are they going to get here quick? Is it going to be a good experience? Is very different than having to infer what the interest is? So I think that's a critical part. The second is, of course, the type of advertising that we provide. We're click-based. So that's important. It's performance marketing, it's not brand. And then I think there's just some broader themes, which is we didn't ride e-commerce up, and we're not writing e-commerce down or up or down. And also brand spend, although we have our Yelp Audiences product, and I know we'll talk about that which is more brand-oriented, we are definitely performance marketing. And then there's just the whole amount of spend on app downloads. You're not advertising on Yelp to drive app downloads or crypto or, or, or. So just fundamentally, we're playing at the bottom of the funnel. We have a high intent, valuable audience that advertisers want to reach. And the things that people are looking for on Yelp, I'm going out to eat, I need a service pro out to my house, and that's just day-to-day life.

Right. And the secondary point, maybe you'll agree with this. I think day-to-day life is also resuming back to a normalization. So the idea of going to a restaurant today versus going to a restaurant a year ago or going to a spa or getting a service professional in your home. There's also that dynamic from a tailwind perspective, right?

Absolutely. And we definitely saw over the course of COVID that when people were not confident, they did less. And when they are confident, they did more. I think at this point, everybody is like I'm sort of over COVID. There may be a risk, but I'm not paying attention to it. I think the thing that's also interesting, we've read a lot about Goldman going back to the office. Are people coming back in the office? Are companies requiring people to come back to the office? And I think that's still playing out because a lot of folks have found that being able to work from home gives them that flexibility that they prefer. And so there's certainly some intention with employers, but it does mean that, for instance, here in downtown San Francisco, there are just fewer people coming into downtown San Francisco or midtown Manhattan or name your City Center. So that has not fully recovered, and I think it's going to take some time for that to play out.

We've got a number of investors this week saying this is one of the busier 2-block radius is in San Francisco in a while over the last 3, 4 days. So I think that's probably right. But let's take one just further step back just so investors could understand if we did go into a sustained period of lower economic activity, what typically happens on a platform like Yelp? Just so people can better understand who your advertiser is and how they might change their behavior just so there's a level setting there?

Definitely. There's obviously both sides. So let me just talk about the category diversification. What are people coming to Yelp for, matters a lot. And so what we saw over the course of COVID is obviously Yelp has, for a long time, been known for restaurants, in particular, that we do beauty and retail in those other categories, and we've now broken that out for you. But what we saw was that this diversified set of categories is a real asset because as people worked from home and did more on the services side, we saw services as a percentage of revenue grow considerably. And is at a little more than 60% today in restaurant, retail and other pull back. So what we see is that across those categories, we think we have the opportunity to participate and that you get that portfolio effect. So that's certainly one component of the consumer dynamic. The other thing is we can't ignore the backdrop today of inflation. And I think what is -- what we're seeing from the credit card players is consistent spend. We were just talking about how much savings people still have. I think people did a lot of revenge travel this summer. And so people are certainly going out, but things are more expensive. So if you're spending the same amount and something is more expensive, you cannot do it as often. So that's definitely a dynamic that we saw play out definitely in the first half of the year. So that's the consumer side. Now businesses, of course, they have different objectives. And let's just take the services categories for a second. I mean, if you were a Service Pro in January of 2021, life is very, very good. In fact, it can't...

If you needed house renovation.

Yes. I mean you're talking like 4 months to wait to redo your kitchen or bathroom, that's a bit slower. And so what we believe has happened is that those services pros want to be busy, and they're willing to spend money on a lead. And because, again, we're cost per click, it means that I'm only going to pay if someone clicks. We spend a lot of time on really continuing to refine the quality of that lead that the service pro gets because you've got the click, of course, but if it's not a good lead, it's less interesting, of course. And if it's a good lead, then that's real value for you. So I think there's an interesting dynamic about staying busy if you're a service pro. Restaurants, it has been a tough slog for them because people wanted to get back out, but then you don't have enough people to work. I don't know if you've had this experience. I've been to restaurants where they still -- even though the lines out the door, they don't have enough servers, so there's empty tables, oddly enough. And certainly, the cost of the inputs has gone up as well. So they're managing against that backdrop. So I think that what we see overall is that maybe just -- it depends on category. It depends on certainly your particular circumstances. And what we continue to believe is most important is we have to deliver value to an advertiser for them to continue to advertise with us. And if I can just go back in time for 1 second, we had a bumpy transition to non-term contracts. But it had a huge benefit for Yelp, which was the speed of signal. If we weren't delivering value, we knew because people left after a month. Whereas when you had annual contracts, it was a long time before you knew whether or not you were delivering something that was meaningful to folk. So that's where we're lined up is the stuff that's in our control, continuing to deliver value to those advertisers across this very broad diversified set of categories.

Okay. Super clear and really, really interesting and helpful. Maybe let's stick in the moment and bring it back to Yelp specifically. What's the current state of your advertiser base? We get a lot of questions. I think people understand the local component of who you're trying to reach. But is it national advertisers trying to find a local audience, SMBs trying to find a local audience, other elements of brand versus DR, you've been very clear on the direct response bottom of the funnel piece? But just help us understand the advertiser base and density that Yelp has today? And how the company thinks about continuing to evolve and grow that advertiser base over the next couple of years?

Perfect. So we have -- we think about go-to-market as 3 channels, 3 ways to reach those advertisers, but it's really local and multi-location. So let me -- why am I making that distinction for a moment. We really wanted to get way more efficient in the way that we interacted with small businesses. So we have been very focused on self-serve. That self-serve portion of our business in the second quarter grew about 30%. And our multilocation business grew about 30%. Together, those now account for about 49% of total advertising revenue. So that's worked. But you're asking me a slightly different question, which is, hey, I'm not -- like whether a rep talks to someone or the person who came in, did the work themselves, they're both local. So 30%-ish is multi-loc, 70% is still local. Now those local businesses do tend to be -- or then you want to look at, hey, among all of our advertising locations, which skew for pain advertising locations between those larger advertisers and the smaller advertisers. So we do have more of those pain advertising locations coming from the national advertisers. So I'm just trying to paint, it's a little bit of a more complex picture than it may appear on the surface. So then why are they advertising on Yelp? What are they trying to achieve if they're a national advertiser? They still care about store business. It really is that simple. I have this asset called a store. I still have it even though we're deep in the e-commerce because I think it delivers something valuable to my customer. Now if I have that asset, I want to ensure that someone's coming into that. And you may not be aware. We have something called Yelp Store Visit. So first-party data is also a theme, but we have the Yelp Store Visits. So we're actually able to do the attribution piece of that for national advertisers. So they know if they're getting a return on ad spend. And so they're coming to Yelp, they want to drive folks in, they want to deploy their dollars, and we think that we help them to do that. And then if you're a small business, one of the first things you do when you start a business is you definitely claim that business on Yelp. And they want to grow their business. They want to have some interaction with their customers. They want to be able to tell their story. And so we believe that we play a very important role there as well. And then I'll just, as a last point, come back to and they want to reach that affluent high intent audience that is making a purchase decision in the moment. They don't want to miss that opportunity.

Got it. Talk a little bit about the investments because obviously, you now have different go-to-market strategies. Talk a little bit about how we should think about some of the key investments you want to make to grow and sustain the advertiser base and reach your sort of broader goals on scale of advertisers going forward?

So by adding multi-location in particular, national accounts, if you know Yelp reveal years ago, that just wasn't a focus, but it's enormous camp. The amount of ad spend in national advertisers is enormous. So first of all, there's just the pure like more dollars in budgets to spend on Yelp. So we wanted to go after that. Now national advertising requirement is very different than your sole proprietor. Sole proprietor is like they're not running a lot of sophisticated analytics. They're like I spent a few hundred bucks on Yelp. I got a few more leads. It seems like it's working. I'm good to go. The mid-market players, they're actually pretty sophisticated. They're hiring agencies. They're doing a lot more work. They are actually balancing budget across different ad platforms. The national folks, as we all know, super sophisticated. They're looking at the analytics. They are making these media buy decisions. And what we have had to do in order to serve that customer is to expand our product portfolio. Because in my time at eBay, I used to have a really close partnership with the marketing team and spend a lot of time on this. Once you figure out that a channel is productive for you, you want to put as much spend as you can through that channel. There's a lot of infrastructure that goes into that. There's a lot of analytics that go into that. And you're always looking for ways to productively spend. Now if we only have one product, you can't spend that much. But if we have lots of products, and we've dramatically increased the product portfolio for these national accounts, then they spend more with you as long as you have the attribution. So what we saw with Yelp Audiences, which is definitely geared to national accounts. That's our product. We are able to enable national accounts to reach our audience off of Yelp. They love that because, first of all, I'm spending on Yelp. Now I can expand that. I can spend off Yelp. On Yelp, it's very high intent. Off Yelp, it may be more brand oriented, but I'm having a conversation with Yelp across the funnel or up and down the funnel on and off Yelp. Think about it from our perspective. That sales rep is now more productive. It's the classic case where you want an enterprise sales rep to have as many products in the bag as possible so that they're more productive, they're serving the customer better. So that's been a huge area of investment and focus for us.

Okay. I wanted to turn a little bit to users and engagement. My two cents you have sort of different platforms where user growth and time spent are more important and rising utility and rising commerce are more important on other platforms. How do you frame up for investors on how they should think about user growth, engagement and use cases evolving on Yelp as a platform?

Absolutely. I mean folks are not spending 22 hours a week on Yelp. Well, maybe some very die-hard Yelpers are spending 22 hours, that's hard to do. But when they come, it's very high, very high intent. And on other platforms, it's low intent. And you have to infer, as I mentioned, what it is that they might want to purchase. And occasionally, it's clear, but a lot of times, it's not. For us, because it's so a high intent and the person is going to make a decision, they're actually pretty quick in the amount of time that they spend on Yelp. So that doesn't deter us. Like that's perfectly fine. We have the -- we're relevant to advertisers for that very reason. But we do think that there's an opportunity to help consumers understand they can do more on Yelp. It's not just for restaurants. It's also for services. And we've spent a lot of time if you use the app, you'll see that in that banner the hero image at the top is often now about services. And so we really have a concerted effort to make folks aware that they can do more on Yelp. So that's one piece. Another piece is just the experience. You can make it more engaging, and we all love to eat. If you're on Yelp, you care about food typically, you're often going to look for restaurants. And so we've redone the home feed, and we're putting a lot more in there. We have a product called Connect that enables advertisers to tell their story. And we see higher actually ad engagement for those advertisers who use that product. We're spending time on Android, where, candidly, we really focused on iOS, and we left a lot to be done on Android. That's a clear opportunity for us. And just acquisition, we only spent $11 million on consumer acquisition in 3 years, '19, '20 and '21, $11 million. I mean that's it's really low, obviously. And so there's clear ROI for us to continue to drive that. But it's always, hey, we want to acquire more customers, but it's always easier or less expensive to get the next session as opposed to trying to get the next customer. So we're spending a lot of time on that. And I would just say, as a word, we'd like that experience to be even more complete. I want -- I know the hours. I love the reviews. They're easy to engage with. I use Request a Quote, service pro contact me. It's easy for me to have a conversation with them. We just introduced something called the Request a Call. We're just trying to make it so that we have that more complete experience even if you're only spending minutes a week, those are important minutes to a decision that you're about to make, and therefore, it's important to an advertiser.

Understood. Okay. I do want to ask one more in this area. Obviously, a lot of consumer behavior changed through COVID, and we're now coming on the other side of that. How do you think about what you've learned about how users interact with Yelp as a platform? And how do you and the team think about aligning product development and your rising utilities you talked about against what you've learned through the pandemic?

So one overarching thing to know about Yelp is we are experimentation-driven agile development. And why is that so important? Experimentation-driven means we develop hypothesis based on the information that we're seeing. Then we run an experiment. If that experiment has yield to it or a signal, then we're going to do more there. And if it doesn't, then we redirect the resources to something that's more productive. So we're always on the lookout for how our -- how is consumer sentiment changing? What is an interaction that's more engaging and more relevant for a consumer? And that is always shifting. So the things that I would say haven't changed about consumers coming to Yelp is I'm coming to Yelp because I trust the content, that it is relevant to me in making a decision. And I know that almost seems self-evident, but it's really quite fundamental. But how do we then get folks to write a review when they may be using other platforms and it's just a photo or it's -- I'm going to take a video. That's actually the way that people want to express themselves as opposed to writing 40, 50 words about a business, it may be that I just want to take 4 or 5 photos of that business. But how do we still prompt them to write something about that experience because the content is really consumed when people visit Yelp. So it's more about how do we continue to drive the reviews. Let's step back where user-generated content site. And if users aren't generating content, then we have a challenge. So we want to make sure that, that's part of it. The other thing is, of course, if you're writing reviews on Yelp, you're way more engaged. So that's a big part of the experience. What's interesting is while we see certainly that our reviews are showing up on TikTok and you can see reviews in Instagram, you got to go back to this completeness thing. Am I really going to go to Instagram and Snap -- obviously wants to use Snap Maps for this. Am I really going to go to those platforms if it's, hey, I want to know all the Thai restaurants near me? Not just that Thai restaurant, that is, at the moment, the hottest Thai restaurant and everybody did a video about it. So I think what we know is that we still need to be able to surface all of that in a way that engages consumers hopefully prompts them to write a review, but certainly continues to deliver value to them because the content is relevant to a decision they're about to make.

Got it. Sticking with that theme, there's a lot of end market verticals you sort of play, and you compete for digital advertising dollars, you're competing for consumer utility, commerce. How do you think about the broader competitive landscape and how you operate within that landscape?

So it is a very competitive environment. One of the things that Jeremy said as a strategic objective, I suppose. More than a decade ago was really ensure that we are getting people on the app. That is really fundamental. And we certainly have our mobile web product. We certainly still have our desktop product. People use those products for different reasons. But at the end of the day, you got to be downloaded, and you got to have an experience that people when they open the app, it delivers value to them. And if you're not doing that, you cannot survive in this environment where, like us, other platforms are experimentation-driven agile development. And they're looking at all the data, and they have their hypotheses and they're running their experiments and they're figuring out what's going to be most effective. And then you have new platforms that rise and present a different set of challenges in the way that people want to interact with you like TikTok. So we believe that at the end of the day, we're a publisher and you got to have good content. And I have a very supply hypothesis, which is as a publisher, if you have good content that attracts a valuable audience, you have a durable market position.

Got it. Sticking with this theme of competition and a new area in the last couple of years for Yelp. How do you think about some of the investments and the competitive positioning you put the platform in, in the services space? And how we should be thinking about the services environment and your own services offering continue to evolve in the next couple of years?

So absolutely essential to being successful as an ad platform is efficiently, effectively matching consumers with the right business fundamental. So Request a Quote is our survey form, intake form, I guess, you could call it, to really determine what is the job that someone needs done. And the example we like to use is if you need a pool liner, that's different than having your pool cleaned. So we've spent a lot of time refining the questionnaire to determine what is the exact project. We've also spent a lot of time on developing the technology infrastructure behind that matching. We run auctions. We have hundreds of parameters that are going into our algorithm. And we think that we've gotten better and better and better at that. And in services, that's so important. Maybe a different theme, and we've had to differentiate the experience between services, restaurant and retail and other. The need is very different for both the business but also for the consumer. So a differentiated experience with information gathering, with an underlying set of technologies that enable us to more effectively match consumer and businesses, we've poured enormous amount of resources in that.

Okay. All right. And then you referenced earlier Yelp Audience and you've given some disclosure about annualized run rate in that business. How should we think about what you built and how that's changed the dialogue in terms of budget, decisions and advertiser density?

So again, Yelp Audiences, that's a product where we are enabling advertisers to reach our audience on Yelp that is coming to Yelp, off of Yelp. So a couple of things that are very important there. First of all, lets us reach new advertisers. So advertisers that would not have an obvious way to advertise on Yelp. Say, one of the car OEMs as opposed to an auto dealership or repair -- car repair. So it lets us reach them, financial services, correct, all of those folks. So first of all, it just opens up new advertisers for us. Second, as I already mentioned, we want to be able to provide a breadth of products to those national advertisers because they want to put as much spend as they can through. Third, we have a valuable audience, I want to reach them, not just on Yelp, that certainly matters to me as well. And what we fundamentally believe about this Yelp Audience product, maybe one thing that's very important is lower margin for us because we're syndicating it. So we have gross margin is lower. So we do look at it in the context of the overall profitability of a national account. So we want to expand how much they're spending with us. Obviously, the rep wants to have more conversations with them, but we're also thinking of it not just in isolation, but what it does for the overall business. And we have seen it have about, I'd say, one point impact on gross margin.

Okay. Sticking with this theme of sort of broadening out the advertising offering. Advertising, as you know, always comes down to what are you delivering for advertisers, what's the broader landscape for ROI and pricing. There's a couple of different topics maybe mind in here. Number one, first give us an update on where your ad tech stack sits today? And how you think about some of the investments you want to make to sort of improve measurement attribution, some of the things advertisers sort of demand from platforms now? Maybe we'll start there and then go down a few different themes.

Absolutely. So each national advertisers, it's almost unique to every one of them, their attribution model. That presents a set of challenges. There's a technology component to it. There's actually just a negotiation, what is going to be the metric we're going to use to determine our return on ad spend. Then you have to have someone you can have that conversation not just with the analytics team, but also with the marketers. So you've got to do a few different things. First of all, we did implement a pixel, so that helps us. That was very important. Because we are owned and operated first-party data, because to use Yelp, you're allowing us to track location. We have a lot of information ourselves, so we're not dependent on third parties, that's obviously important. And then we have to be able to surface that information to the analytics team, either in output like raw metric or input the raw data. So that also ends up being very important. So we've invested heavily there to continue to increase the technology side of attribution. But what's probably maybe less obvious is we've also had to increase the people side to have almost like sales engineer, but it's analytics -- sales analytics engineer kind of concept in order to have a dialogue with the analytics team as the national account, what data you need, what are you going to look at, how are you going to measure it? You have to geek out a little bit on the stats and figure out what are you doing with outliers and all of that and bots. So we've had to invest there as well. And the sales reps have had to educate themselves on some of these subtle pieces. So it's -- you can't just do the tech. You've got to have the organization in order to have the conversation.

And then maybe 2 follow-ups. The first one is, as you continue to iterate and improve, talk a little bit about, for lack of a better term, the flywheel effective as you bring more advertisers in because you iterate? And what impact that has on pricing and auction density? And how we should be thinking about sort of almost an inflationary impact to pricing because of that density?

Yes. So if you're not familiar, we actually get budgets from advertisers. They are not setting their CPC. And it's up to us to efficiently deploy that budget over the course of the month. And the way that we do that is we run that auction. We think that we're quite sophisticated in the way that we do that. And what we have to do from a density perspective is continually look at ways to drive not just the click, but the value of the lead without degrading the consumer experience, that's the constraint. So we spent a lot of time adding products, but we're also spending a lot of time experimenting and adding ad formats. And then we are increasingly sophisticated in the way that we're running that auction. In the past, we just ran the auction on unit by unit by unit by unit. Then you run it on a couple of units. We're on our way to whole page auction. That becomes really -- which ad format works best for this business, given this consumer who's visiting right now, that's a lot of work, not just in the infrastructure because it has to be fast, but also on the algorithm side, plus on the experimentation on the ad unit side. So you got to like put all that together. We are a vertically integrated publisher. And I just -- I think it's very hard going forward if you don't have that vertical integration. I think it's very, very hard to continue to hold on to your profits out of this business because if you're giving that away to someone else, then you're obviously going to be losing the game because the speed at which the innovation is happening is very fast. And so you got to be able to keep up. So we're super first-party data important. Obviously, we help folks to generate content. We curate that content. But this ad tech stack is absolutely strategic.

Last one on this is as the team goes out and dialogues with advertisers, what do you see as some of the big asks coming back to you from the advertising community that you have to unlock to continue to sort of have the operating momentum you want to achieve?

There is never an end to the attribution request like everybody is like help me understand the return on my investment, so that never ends. There's clearly on the Yelp Audiences side, hey, could you -- this great audience, can I reach them on CTV, like or OTT. I want to be able to a nonlinear TV, be able to reach that audience. That's certainly been an important area for us. It was interesting when we introduced the Connect product, that's the one where we enable advertisers basically to create a story, Instagram style story on Yelp. It's like, hey, I want to just be able to talk to consumers directly. Those are sort of 3 themes for us.

Okay. I want to turn to the investment side and the margin side of the equation for a minute. Obviously, you've got a lot on the agenda side that aim towards growth and expansion and scale. How do you manage the level of investments you want to make into growth versus achieving your margin targets? Obviously, there's a greater focus from investors today on margins than there is growth versus 18 months ago. How do you think about sort of striking that right balance? And how does that come back to either your philosophy about margins or the way you've framed margins for investors going forward?

I would say when we told investors a year or 2 years ago that we were focused on long-term profitable growth, it was less interesting. Now when we tell investors that we're focused on long-term profitable growth, it's more interesting. And so we definitely make those trade-off decisions extremely ROI-driven company. And because we're measuring everything, we're running the experiment, the implication of that is, hey, if we're seeing return, we're going to invest more. Even if it means that it's going to have a little bit of impact in the near term on margin because we think it's the right thing to do for the long term. But we are definitely trading off between revenue and margin in order to deliver sustainable long-term profitable growth. And that's the mindset that we go into it. We're running the planning process right now for '23. And that's at the heart of everything that we are thinking about. We have our strategic objectives, but it's how are we going to, over the next 3 years, deliver the business in a way that meets the needs of consumers and advertisers, of course, but also shareholders.

So one thing you mentioned recently is that you might lean in on the marketing side in the back half of this year. So against that backdrop, you just laid out of striking the right balance. How does marketing fit into the formula? And how should investors be thinking about what an increase in marketing or higher intensity around marketing will yield or what the output they should be expecting from those marketing dollars beyond just the second half of '22?

Yes. I think the hard thing about marketing, ironically, we're in the market. We're in the ad business. And we're also marketing our business to folks is -- you spend not just to get return in the current period but over a number of quarters. And so there's that mismatch. It does mean that you need to be spending consistently because the spend from 4 quarters ago is showing up now, right? It's different than e-commerce. Did someone buy this good right now is very...

Short purchase cycle, bottom of the funnel...

As opposed to, hey, we acquired a business, and we're doing our best to retain them, of course. And we expect them to retain for this period of time. And if we spend $1 now, we're only going to get a return over a number of quarters. So that's the tension in the conversation that obviously we want to have with all of you. And so you have to be consistent. I'd just say you can't -- for a business like ours, turning marketing spend up and down is not an optimal strategy. So sometimes we ask for a little bit of patience, and we ask for some room to do that investment. And that's how thinking about it here in '22. That being said, it's still very ROI-driven. We're constantly measuring it. If we learn more, and it's more productive, we're going to ship the spend around. That's what you'd expect of any media buyer. All that being said, we've gotten the question, well, you cut your local sales force in half. I didn't think that was going to work, but good for you, you did it. But now it seems like you're just filling in with marketing spend, like when is that going to end? And the short answer is we -- we're still committed to driving margin over the long term. How are we going to really do that? Yes, we're going to increase marketing spend. If there's a return to be earned and just telling we're going to spend the $1. But fundamentally, we shifted from that sales headcount-driven growth model to a product and engineering-driven growth model. And that's where the leverage is going to come. So we will come back to you and say, "Hey, we think there's an opportunity to spend like we did as we came into the third quarter." But at the end of the day, the long-term margin expansion is going to come from that product and engineering-driven growth model.

Understood. You've laid out a couple of key pillars for the way investors should think about capital allocation within the company. Maybe just refresh for those who don't know, what are the strategies around capital allocation? What causes management to sort of rethink those or how to lean into one versus the other? How should we think about the priorities around allocating capital?

So we -- the way that we approach it is we have a base level of operating cash requirements that we meet. If we didn't learn anything else from COVID, you better have a healthy buffer. And then obviously, we went from COVID, maybe not over but right into macro. So you got to have a healthy buffer on top of that. We do hold cash on the balance sheet because that's for tuck-in acquisitions. And then our commitment is to return capital in excess of that target balance. And we simply believe that we're stewards of capital and that it's incumbent on us to drive total shareholder return, and that's our commitment.

Okay. We only have a minute or 2 left, but I'd like to turn the mic over to our speakers. When you think about where the company is positioned, where you want to go thinking about the next 1 to 3 years. Is there anything you want to highlight that you think is either thematically or idiosyncratic to the company that you think there's a bigger disconnect between investors and what the company thinks or where the company think the platform might evolve to?

So maybe just in the spirit of being direct, the bear case on Yelp is Google will crush you. And I would just point out, we're 18 years into the journey, they have not crushed us. So maybe the surprise statement as we will be here in 3 to 5 years. That's the first one. The second is we think that we have an enormous opportunity ahead. And we think that we have really lined up the company to take advantage of that by shifting to this product and engineering-driven growth model. We think that we can be even more of a complete solution for consumers, and we want to deliver against that opportunity. And I would just say, overall, folks may know Jeremy, folks may not know Jeremy. Jeremy is in the business every day. He's in the detail. He's still extremely passionate about the role Yelp plays in the world, and he is very committed to driving that performance over the long term. And so you have a founder, CEO, who's been doing it for a long time and really cares about it. And I think we have a market position that enables us to do more for consumer in advertising.

Okay. Super clear. Well, David, thanks for taking the time. Really enjoyed the conversation. Please join me in thanking Yelp for being part of the conference this year.