Maplebear Inc. (CART) Earnings Call Transcript & Summary

Before we get started, I have to read the research disclosures, all important disclosures, including personal holdings disclosures and Morgan Stanley disclosures appear on the Morgan Stanley public website. They are also available at www.morganstanley.com/researchdisclosures and at the registration desk. Some of the statements made today by Instacart may be considered forward-looking. These statements involve a number of risks and uncertainties that could cause actual results to differ materially. Any forward-looking statements made today by the company are based on assumptions as of today. Instacart undertakes no obligation to update them. Please refer to Instacart's Form 10-K for a discussion of the risk factors that may impact actual results. Chris may also reference certain non-GAAP financial metrics, and reconciliations are available on Instacart's Investor Relations website. The last part was unique. That was custom.

So well, nice to meet you in 3D. So you've been at the role now for about 9 months.

Yes, that's right.

You've been handling a lot of investor discussions, internal meetings, getting to know the company. As you sort of sit and sort of compare and contrast those, 2 questions. One, what has surprised you most or what you've learned about the company sort of going division by division? And then sort of like match that up with what do you think is the most misunderstood aspect of the company externally?

Yes. I think -- okay, so a couple of things. First of all, it hasn't been quite 9 months. It's been 6 months, I think, but I was appointed 9 months ago, and I've been effectively running the company, I would say that since then. So starting on kind of the thing that surprised me the most, like the underlying strength of our business has really and the -- strength of our strategy has really been a pleasant surprise. Obviously, I've gone -- I went very deep on the strategy when I got the role. And I do think that people don't necessarily think about the strength of our business the way that we think about it internally. Most people just think about the fact that we are -- they think of us as a consumer marketplace when really we are so much more than just a consumer marketplace. We are the leading grocery technology company in North America, and we're continuing to extend our lead. And that's something that's very different and differentiated from what anybody else does. No one else is doing what we're doing. No one is building the type of integrations that we're building. And the market that we're going after is just enormous. Obviously, the grocery TAM is there, but also as evidenced by the number of partners that we're working with. And the genius of our strategy is everything that we do, we're building a system that is reinforcing itself and is all connected to one another, which allows us to kind of invest in one area and then get the benefits platform-wide. So let me tell you kind of what I mean by that. So obviously, we do have a large and growing marketplace. The marketplace is a very big part of our overall business. But we take all of the learning and the scale from that marketplace and we offer it to retailers in the form of an enterprise business. So that strengthens our enterprise business. And then our enterprise business, in turn, helps us on our marketplace because we have integrations with retailers and we have quality work streams and we have SLAs with the retailers. The enterprise business also gives us access to the ad surface area, powering retail media for retailers. It gives us access and a right to win on in-store with our Caper Cart. We have an in-store physical AI-powered cart. If I take the Caper Cart as an example, the Caper Cart is operationally excellent. Retailers love it. It's also an ad surface. It's also going to help us strengthen the marketplace because it's collecting in-store signals. So those are just a couple of things that -- a couple of examples. But really, the point is that we are a grocery technology company that is leveraging our advantage across enterprise and marketplace to stay ahead of the pack. So that's a key strength area that I want to highlight. From an unexpected surprise, unexpected challenge, I think you said. Like, look, I think the grocery category on its own is extremely complicated, and it presents a huge opportunity for us. Outside of the fact that the TAM is enormous and it's only 13% penetrated, it's actually a very attractive category because it's a recurring category. It's an essential category. It's habitual. That said, it's also an extremely complex category to get right because it's local, it's perishable, it's low margin. And I think a lot of other players think that they can come in and just win this category, but they can't because of the complexity. And all of that complexity is actually where our strength is. We're very strong because we've already built the retailer integrations. We've built the density. We have all of the data. And so we can essentially power the intelligence layer in an AI world. And so all of the complexity of this category is the reason why our growth is sustainable, it's durable, and it's the reason why we think we can continue to extend our lead.

And that point on the complexity. So there's -- there are a lot of factors you're bringing together with the algorithms, the delivery people, matching everything so the orders on time are accurate. Maybe sort of walk us through what have been the biggest drivers of the recent acceleration in growth? And how do you think about what will sort of keep that faster growth durable throughout '26?

Yes. So the -- if I had to summarize what we're trying to achieve, our goal for 2026, it's accelerate, and we are accelerating to your point. So we just posted our best growth as a public company at 14% GTV growth last quarter. We just provided our best guidance as a public company, 11% to 13% for this quarter. And that's despite there's been tons of competitive headlines. There's been -- competitors have actually been building out. And that's because we're playing a fundamentally different game. And because we're in the lead, we're able to press our advantage across multiple priority areas and multiple dimensions on -- across our business. So there's our marketplace, as an example, our core marketplace, we have -- we're investing to make sure that we grow that marketplace. In fact, we have been accelerating here, too. We accelerated every quarter through 2025. We view it as our job to provide the absolute best customer experience across all consumer -- all elements that the consumer cares about, including affordability. I've talked a lot about affordability and the importance of affordability since I've gotten this role. And one of the areas that we're tackling now is moving price parity to stores with retailers. At our last earnings call, we announced that Hy-Vee already this year has gone price parity. Raley's has gone price parity. We're about to go price parity with Fairway on both Storefront Pro and on the marketplace. We know this is important because retailers that are priced at price parity, they grow faster, they retain better. But that's just the marketplace. Then we have the enterprise platform where we work with retailers to grow their business. And as they grow, we grow. And we just continue to scale this side of our business. We are up to 380 storefronts. So that's where we actually power the e-commerce storefront for the retailers. And we're also -- we also continue to expand our enterprise offerings with all of our existing client base. So as that client base grows, we sell them more products and services. So that's an important element. And then the 2 other big priorities that we're leaning into is obviously advertising and continue to compound our ads and data advantage. We just continue to scale here, right, 9,000 brands. We are at 310 retailer sites that we power. We know that brands are looking for a partner that they trust to scale across multiple surfaces, a player that they trust on performance and measurement, and we are that partner. And then the final area that we're leaning into in a big way is obviously AI. We are -- our ambition is to build the absolute best agentic experiences and the best AI grocery assistant directly on Instacart. Of course, we're going to work with others like Gemini and OpenAI and others, but we want to build the best one there. And then again, consistent with what I articulated upfront, we're going to take that to our retailers in the form of an enterprise product. So we're taking it to Sprouts and Kroger are the first retailers that have signed up to work with us on AI.

There's a lot of projects. You have a lot of different balls in the air items in the basket, I guess. Remind us just sort of what is the company's philosophy about sort of how much it chooses to invest, where it chooses to invest? What are sort of the analytics or the ROIC they're doing on the investments? And how do you sort of balance we want to invest versus flow through the profitability?

Yes. We have -- we operate under -- we have internal financial guardrails. And we -- obviously, because we're so early in this category, we look for every opportunity where we can get a return. The good thing is that the opportunity for us to invest in new areas is large for us. So for example, we're in the process of taking our tech internationally for the first time. And so we -- this is an investment that we're making, but we think it's a long-term growth lever for us because of the TAM that we can unlock. Europe has a larger grocery TAM than North America. So this is -- I think this is a great example. We recently launched with Costco in Spain and France. We announced that we were going to start going internationally just when I took the role, and we've already entered countries. And so we evaluate opportunity by opportunity. Obviously, we have guardrails where we believe that the investment is a responsible investment. But because we have such a giant opportunity because we're in the lead and because this market is so big, we look for lots of opportunities to drive growth.

Maybe I'll double-click on the international part because I think there's a -- walk us through what you're investing in international. How are you building it out, asset-light, asset heavy? And sort of we're sitting here 1 year from now, what will be viewed as success with the international expansion?

Yes. So our approach to international is very much an enterprise-led strategy. So that means that we're going with our existing products. We're going with Storefront Pro. We're going with Caper. We're going with FoodStorm. I do think that our enterprise products are going to travel very well because grocery complexity is universal. We've been in the market. We've talked to partners in various other countries across Europe as an example, and they're trying to solve the exact same problems that we're trying to solve that retailers here in North America are trying to solve. So we think our technology is going to really meet their needs. And what's amazing is many of the retailers in other markets in Europe, in particular, they don't -- they haven't built e-commerce capabilities or they have a site and the site is not transactable. And so as a result, many of these markets have really low grocery penetration, much lower than the U.S. And so when we look abroad and we look at the size of this opportunity, we think we have product market fit and the markets are ripe for us to go in and help drive online grocery penetration. From a cost perspective, we think we can go internationally in a very cost-disciplined way because we're going with our existing tech stack. We're not building new products to go international because we're using Storefront Pro and Caper and FoodStorm. We're also, as part of our approach is we're going to -- geographically is we're going to target major markets where we can have an anchor partner similar to what we've just done with Costco, and we'll localize around that anchor partner for that market. So as you would expect, we have target markets across other large attractive markets. But if we start working with a major retailer in that market, we will localize there first. And so that's how we're going about international in a highly cost-disciplined way. I will say that we were very happy to be partnering with Costco internationally because we've got such a deep interesting relationship with them. We've already been working with them. We've launched Storefront Pro in the U.S., and then we launched it in Canada as well. So this was a natural extension of that strategy. But that said, we are also pursuing net new retail partnerships in these markets. We're going to be looking for retailers that are trying to solve technology needs around grocery e-commerce, and we'll be pursuing them as part of this go-to-market. Success in a year would look like us continuing to build this out in a meaningful way with net new partners.

Got it. Okay. Let me ask one about competition. Your core business right now is really, really cranking at a high rate, your retention, your cohorts, your frequency, everything is going in the right direction. And yet there is this almost bogeyman out there where Amazon, Walmart, DoorDash, Uber, this litany of companies are all sort of getting bigger and bigger into grocery. And eventually, there's this view, well, one or all of those are going to be able to compete more aggressively against Instacart for the big weekly shop. That's going to be a problem for Instacart. What's your reaction to that? And sort of what gives you confidence in the durability of your growth even as these other players come in to go after your shoppers?

Yes, it's a great question. We don't obsess about competition internally because for a variety of reasons, but we do obviously watch it very closely. What I'll say is, you're right, large baskets are 75% of the market. That's where loyalty is built and that's where lifetime value is built, and that's where we are particularly strong. We're also strong in small baskets. It doesn't mean we're not strong in small baskets. We are -- in fact, we also convert small baskets to large baskets at faster rates than others. But when I take a step back and I look at the overall landscape, obviously, the market is big enough for multiple players to win. But when I look at the specific offerings that are out there, most of them are geared towards small fill-in orders, I think probably incremental use cases versus the weekly shop. And if you look at Amazon as an example, which you raised, Amazon has a few thousand SKUs in their switching stations or their warehouses. It looks like they have 4- to 5-hour delivery windows. When you look at the data, it would suggest that their baskets are under $50. If you look at the restaurant delivery players and the traction that they're trying to make within grocery as well, when they launch a new retailer, they gain an initial amount of share because their share was 0, but then they start to plateau and their baskets stay small. And that's despite these partners being after this business for multiple years in the case of Amazon, multiple iterations. And so none of this surprises me because smaller fill-in orders are much easier. The large weekly shop is what's extremely complex, and this is what we've cracked, and we've been at this for almost 14 years now. Our AOVs are over $100. And the reason is, is because we know what it takes to win in these baskets. And it requires deep selection. So we're at 2,200 retailers, but also 100,000 stores. And we also have 22 million unique items. You can get pretty much anything you want on Instacart in the grocery category. We've got -- we invest heavily in quality. Our perfect order fill rate was up another 5 points last quarter. We're very fast. So we deliver our on-demand orders, which are 75% of our orders. That's anyone who's clicked priority or the first available ETA. We're delivering those on average in 90 minutes. Now a lot of them are delivered in under 30 minutes. And we have these retailer integrations. And I don't -- I really want to emphasize how important these integrations that we have with retailers are to the overall experience. And this is something, again, that none of our competitors have or are building out, right? We work with retailers to improve their owned and operated sites. Retailers like Kroger, where we power their deliveries on all of the Kroger properties, when we work with them to improve that experience, it's also improving marketplace. When we work with retailers on the depth of understanding the inventory and what's on shelf, we're also getting that benefit on our marketplace. So that -- these are things that really differentiate us in the experience. It's the reason why we have success in large baskets and others are struggling to break through, and it's durable and sustainable because others don't have those integrations and are not even pursuing them.

All right. Well, let's talk about agentic. I'm sure you have heard me. I know Emily has heard me explain my Utopian grocery dream of the Saturday morning grocery shopper that says, good morning, Brian. Here are the 30 things you and your family order. I found some new dip that goes well with your favorite chips. Here is a coupon for a new 2% milk. Do you need a more Topo Chico for the Super Bowl party you're throwing? And is the Monday morning delivery time what you want? And I will say, heck yes, this is amazing. How far away are we from that? Is that something that is 12 months away, 24 months away? And sort of what are the most challenging parts of getting that right on Instacart?

Yes. I think our imagination is being validated in real time with what's going to be possible from a consumer perspective. I think let me tell you how we're approaching this agentic space and how we're navigating our own strategy versus the strategy of partnering with others. Our goal is to build the absolute best agentic experiences and AI experiences directly on Instacart. And we do think because everything -- all of the examples that you just used requires data and it requires intelligence and it needs to understand you. We think we're going to be in the right position to be able to build the absolute best experience because of the data. Grocery complexity exists even in an agentic world. And so it requires you to understand millions of items, local preferences. There's dietary preferences, budgetary preferences perhaps, in-store intelligence. And so our objective is to use all of the data that we've amassed from the 1.6 million orders that we've done to date and build the absolute best intelligent assistance so we can do exactly what you described. That's our objective. So that when you come on, we will know what you typically buy, but we'll also know what you intended to buy. We will combine that with all of the data signals that we get from in-store, like what is in stock, what's on promotion. It will build in your preferences, for example, again, your specific diets and your specific dietary preferences. So with all of that, we do think that we're going to be successful building the gold standard of agentic experiences directly on Instacart. The way we're going about that is from an experience standpoint is we're going to build the AI assistant. So when you go to Instacart, you're going to be able to prompt an agent directly on all of the things that you're hoping to kind of achieve in your shop as you build and refine and optimize your basket. But we're also just going to build agentic experiences throughout the consumer experience. So to give you an example, if you don't want to come to Instacart and engage in an agent directly, what you could do is you could just fill your cart normally and then ask it questions throughout and have it optimize your grocery basket throughout. So at the end of the shop, you could say, is there any gluten in my basket or how much sodium is in my basket? Or are there other alternatives from a sale or price perspective? And so we're going to build those experiences. And then we want to, again, extend all of the innovation and technology out to our retail partners in the form of an enterprise product, which we're doing now. And we believe by getting this right, Instacart will still be the primary destination for these types of experiences that you described. People will come to Instacart for their weekly shop because the agentic experiences will be the best. Now when you take a step back, a giant step back and you say, okay, well, the landscape is broader than just Instacart and obviously, there's these AI platforms. The way we think about this is we want to be wherever consumers shop. And we want to be the grocery intelligence layer and the grocery technology layer on those surfaces. But we want to work with those platforms to help shape the experience versus just react to it. So we're operating under a series of principles. One of those principles is we want to co-create the grocery experience. And in our partnerships with these AI platforms, we're able to do that because we are the experts in grocery. We understand the grocery journey. but we also have the intelligence layer and all of the data. So we're able to do that. The other thing that we want to do is protect that data because for obvious reasons to be able to control the experience in the future. So we're surfacing our data to these platforms in an extremely controlled way. But the goal here is to make Instacart discoverable on these surfaces and when they discover us, have that experience be amazing. So regardless of how you come into Instacart, you're having a quality agentic experience. And right now, the referral volume that we're getting from these platforms is still really small, and we're not seeing any material changes from a consumer perspective on how they engage with Instacart. But what I'm actually hopeful for is that this is going to help us drive incremental users and incremental use cases because, again, coming back to your example, there are so many use cases where if you're conversing with one of these platforms, you might just want to order a couple of ingredients for dinner or fill in order. And if that grows the online grocery category, that is a huge win for Instacart because, again, remember, $1.2 trillion market, 13% penetrated behind almost every other retail category, there's so much upside. So if we're discoverable on these surfaces and we have a good experience and they're driving incremental use cases, that's more e-commerce sales. That's good for us, both on our marketplace and for all of the partners that we power.

Let me go back to the on-platform agent, though. Just that -- so we understand the biggest challenge, like the hurdle to build that agent. Is it compute capacity? Is it just sort of engineers need to build it out, you have the capacity? Is it getting the couriers? Is it supply? Like what's sort of the hardest part that needs to be completed?

I think the most challenging part of all of it is having all of the data signals, and we do have that. So at this point, it's just a matter of making sure that we launch with the absolute best experience. We're not in a hurry to be the first one out of the gate. We want to be the best one out of the gate. And really, the challenge -- the reason why others will struggle where we will succeed is because of the intelligence layer that's reinforced with the physical signals. You have to remember that we're a logistics company. We're a software company, we're a service company. We're also in-store, we're a logistics company. We gather all of these signals from in stores. We know when something can be delivered. We know when something is going to be in stock. So it's stitching all of those data layers together in order to build an agent that could do anything through refining and building the grocery cart.

I think you're quite important to some of these early horizontal agents. When we think about sort of these horizontal agents that are trying to be built out ad dollars or commissions are going to follow wallets, consumer spend. Grocery and CPG is the largest bucket of offline spend remaining. I actually think sometimes if they don't capture food, there could be a problem with their long-term monetization algorithm. You don't search for carrots the same way you do running shoes as one example. So as you structure these deals and sort of let them have you on the platform to enable grocery behavior, how are you sort of ensuring that you have safeguards in place so that you can still win those customers, bring them back as direct Instacart customers. You don't sort of get disintermediated by an agent that wedges itself between you and them?

Yes. Again, I think we're viewing these other agents as lead gen in incremental sales. And by basically controlling the data that we're surfacing, they're sending us the query and then we're basically supplying the answers to that query. It's always going to be a good use case and a great base use case, but the best use case is going to be on our platform where we have all of that additional data where you can in real time interact with the agent. I think when you talk about from a brand perspective, I think there's going to be lots of opportunities for us to work with brands in this world for a couple of reasons. One is because we're maintaining ball control of the agentic experience directly on Instacart, we're going to be able to work the consumer angle with the kind of media and ad angle in parallel. So we're going to -- we actually are working those road maps in tandem so that we're able to appeal to CPGs. And we have a lot of experience and I think a right to win with CPGs in the way that we go about this because, again, they're bought into our performance. We have a history of delivering ad formats that really move the needle for our brand partners. And the other thing is we're very strong from a relevancy standpoint. Our underlying data models help us with the relevancy of what we're putting in front of consumers. And with that, the digital shelf will change. What we're showing to consumers will change over time because it will be tailored to each individual person. Once we have that relevant shelf in front of consumers, we're going to be completely indispensable to consumers, but brands are also going to want to be part of that digital shelf. So we have ongoing ideas on how to engage with CPGs on how to drive their products and drive discovery for their products and throughput for their products on our platform. I'll tell you, every conversation I have with the CPG over the last 6 months has been how do we think about the future here, and this is how we think about it.

Yes. The ad business is doing very well. It grew 10% last quarter. You added thousands of brands and different advertisers to the platform. As you sort of think about 2026, what are the key areas you're investing in, in the advertising business this year other than agentic to sort of continue to get not only new advertisers, but also scale that spend per advertiser this year?

Yes. So you're right. We're very pleased with our ad results. So we had 10% in Q4 adds and other growth, and we just guided to 11% to 14% in Q1. That's on top of 14% comp from the prior year for Q1. And look, I think this is a reflection of a strategy that we've been building out that's working. We want to be -- we want to build the best and most complete ads ecosystem for CPGs. And that's working. That's driving really healthy diversification from -- on an advertising basis across supply and demand. And what I mean by that is on supply. So obviously, we have our growing marketplace. We have 2,200 retailers. So when an advertiser comes to us, they get access to a large growing marketplace with lots of retailers. But we're also now rolling up and powering all of these retailer sites, 310, what we call Carrot Ad sites, where we're extending our tech and our demand on other services. Brands love that we're doing that because it saves them from having to work with individual retail media networks. They can come to us, they trust our performance, and we scale there. We've also started to scale on the actual physical Caper cards, the cards that we have in store. So all of this incremental supply, all of these surfaces allows us to attract an influx of demand, that and all of the AI tools, which I won't touch on, but are an important part of that equation as well. So with that, we just continue to attract more brands. So we're up to 9,000 brands versus 7,000 brands the year before. And we're attracting more and larger budgets because we're using those budgets across a larger surface area that's performance. And so that is -- that strategy, that underlying strategy is one of the reasons that we're successful in driving growth. And that's all on platform, what we would consider on platform. That's growing -- that's driving the bulk of our growth, but we're also now expanding off-platform. So what I mean by that is brands can come to us and they can use our first-party data to plan their campaigns on search and social. We've struck partnerships with Pinterest and TikTok and Google and Meta and others. And that allows them to use our high-intent audiences. They can refer that traffic back to Instacart. They can get closed-loop measurement. So it's highly strategic. And one of the things I want to address here because we've gotten it a couple of times over the last couple of days is, yes, some of our off-platform ad spend comes with a higher cost of revenue. We know that, but that's very intentional because what we're realizing is that the ad spend for off-platform is incremental because it's coming from search budgets and social budgets, it's coming from a broader digital budget. So we're actually driving incremental profit dollars through this strategy, which is working. So what you're seeing, the results that we're driving from an ads perspective is the strategy, the ecosystem strategy working for us. And it's why we believe that we're going to be able to hit our long-term targets of 4% to 5% of GTV.

Got it. Great. Let's talk about how you're using Gen AI sort of internally to drive more productivity. You've talked about 40% increases in output per engineer, 4x faster on some projects or the software being developed faster. As you think about this year, how do you think about further Gen AI savings down the P&L to either reinvest or let flow through to shareholders?

Yes. Well, I mean, this is not a unique to Instacart observation, but it's remarkable what's happening here. And that stat that you referenced, the 40% higher output per engineer, that was a stat from 2025. What's even more remarkable is that 10% of our engineers, our output has increased double that at 80%, and that just continues momentum into 2026. The 4x faster software projects, that's where we have an entire team using AI. So if we start a project where every kind of pillar within the project is using AI, we can complete that project 4x faster versus individual engineers using it. So the velocity of what we can achieve here, it's not incremental. It's quite transformative, and we're leaning into this in many ways. So just to give you a few examples of how we're using it internally to kind of to layer on all of the productivity gains with some of our critical advantages as a business to move faster is we're using it on our core marketplace to improve order quality and fulfillment efficiency. We're constantly running experiments on Instacart, as you would expect. We can run these experiments faster. We can measure the experiments faster. We're onboarding retailers faster than we have before. But this -- the velocity of our retail launches on the enterprise side has accelerated. Last year, we did 70 versus just 30 the year before. We can launch them faster and we can launch them with more features. It's not -- in the past where we might have done an MVP, we stood up the site. Now we can give retailers more of what they're looking for out of the gate, more white glove service. It's helping us go international. the difference between going in international now that we're starting this build-out this year versus even 12 months ago, it's going to be remarkable about how fast we can actually move on the tech side. So basically, the way that we're viewing this is we're leaning into agentic AI across the entire company, across all functions, and we're accelerating our momentum in all of these areas.

4x faster. sounds very encouraging, but when I can see my grocery shopping agent because if it's 4x faster, it should be years. So I'm optimistic I'll be able to order my Topo Chico automatically over the next 12 months. Talk to us about the capital allocation and sort of the share repurchase program. You've been pretty my word, aggressive in buying back the stock over the course of the last year or so. So just philosophically, how do you think about capital returns in the right amount?

Yes. So our strategy from a capital allocation perspective is, one, the first thing that we always look to do is reinvest in the business to drive momentum. This comes back to the fact that we are the leader in a growing market that's still early. The most responsible thing for us to do is look for every opportunity to invest in the business, and that's what we do. Our second priority is then to save some firepower for M&A. We've had some successful M&A. Last year, we bought a company called Wynshop, which is a grocery technology company. In the past, we purchased Caper, FoodStorm, which is a catering company. We bought a company called Rosie, which provides retail tools for small independent retailers. This has been something that we've done well in the past, and we want to save firepower to be able to do that going forward. And then the third strategy is we want to opportunistically buy back our own shares, which you saw happen last year. We bought $1.4 billion worth of our own stock back in 2025, and much of that was concentrated in Q4. So in Q4, as our business accelerated, we bought back $1.1 billion worth of our own shares. And so we expect to continue to do that on an opportunistic basis going forward. We always approach that as an opportunistic approach.

Great. All right. Just one more. If we sort of -- there's always this discussion about large baskets, small baskets, where are you better positioned, more challenged over the long term? How do you think about 3 to 4 years from now, does large basket, small basket matter? Or are you just sort of we're going to go after all of it?

Yes. We look at this data quite frequently. We're not -- I don't see a shift between large baskets and small. What I see is incremental use cases coming in for small baskets. And because the category is so early, I think lots of small basket use cases are going to pop up as consumers get more comfortable with ordering online for their grocery needs, which is, again, lagged over time. The large basket, the idea that a consumer is going to place their grocery order for the week, I think that's a durable habit. I think it's been like that for 50-plus years, and I think that it's going to continue to be like that. We see the spikes when consumers should -- like Sunday, as an example, is a big day for us. We can see that families are placing their grocery order for the week, and we expect that to continue.

All right. Well, anxious to see what happens over the next 12 months. Hopefully, I'll have my agent ready to go.

Awesome. Thank you so much.

Thank you.