Nick Giovanni: And as it relates to the growth step-up, we are only guiding to Q1. And for Q1, we are guiding to GTV growth of 7% to 10% year-over-year. And this compares to 5% for all of last year. But through the quarters of last year, it was 2.6% growth in Q1 and 5.6% growth in Q2, 5.9% growth in Q3, 6.8% growth in Q4. And so what we’re calling for now is a step-up to 7% to 10% in Q1.
Jason Helfstein: Thank you.
Operator: Thank you. Our next question comes from the line of Michael Morton of MoffettNathanson. Your line is open.
Michael Morton: Thank you for the question. I wanted to ask maybe 1 a little bit longer term just about industry structure. You have like incumbents with the physical network building their own distribution businesses but then also marketplaces competing with Instacart who is the first mover and can do it with a larger scale. But I would love to know, how do you see the actual picking aspect of fulfillment developing over the year as order density increases? Do you think — there are some in the industry that believe it’s going to go to merchant pit, so then like kind of the Whole Foods model that’s just bags in the front of the store and the economy workers pick it up? Or does this continue along the Instacart picker model? Just any long-term thoughts there would be great. Thank you.
Fidji Simo: Thanks for the question. So, the reason retailers for now with us to do picking and delivery is because we can do that very efficiently, which allows them to save on cost and pass on some of that cost to customers, which then creates growth for them, and with high accuracy and high quality to really respect the relationships that we deal with our customers. We do that in a variety of ways. So, the model you’re describing where some retailers might be doing the picking and then with the delivery, like some of our retailers are already using our own picking app to stage their order, for example, for pickup. And so again, we see our role as really developing all of the technologies to help support that industry, and one of them is great picking technology that retailers are integrated with our own system and are currently using inside their store.
Now, we have also seen models thinking about like automated picking and kind of having big warehouses where the picking happens outside of the network stores. And we are tracking that very closely, but we are much less bullish on this model because what we have found is that for these models to work, you need a lot of density of orders. And that means that these kind of large warehouses end up being very far away from customer, and all of the costs that you save in picking, you end up losing in cost of delivery because you have to drive to a customer that’s much further away. We are also seeing that customers actually value speed, and that’s why I have emphasized speed so much in my letter because we know it drives conversion. We know it drives growth.
And so if you’re very far away from the customer, you are not able to deliver with the speed that we can in doing that from 85,000 stores. So, all-in-all, I would say that we strongly believe that we have the winning model here, the model that’s most efficient, most responsive to customer needs. And we are continuing to develop all of the technology that retailers need to embrace that model in all of the ways that they find to be good for their business.
Michael Morton: Thank you so much.
Operator: Thank you. Our next question comes from the line of Andrew Boone of JMP Securities. Your line is open.
Andrew Boone: Thanks so much for taking my questions. Can you talk a little bit about the product road map for off-platform advertising? Where are you guys today and what can this look like over the next couple of years? And then Fidji, a bigger picture question on Health more broadly. Can you talk about the potential for HSA and what that could mean for the business? Thanks so much.
Fidji Simo: Thanks Andrew. So, on off-platform marketplace advertising, step one was taking our entire advertising infrastructure and making it available to retailers on their owned and operated properties through Carrot Ads. And that’s a growing part of the business. We are finding more and more retailers be interested in leveraging this technology because that allows them to stand up a retail media network literally overnight and get a new incremental revenue line. So, that was step one. I would say step two is taking all of the vast set of data that we have on customer behavior and leveraging the data to make advertising on other platforms more efficient and that’s where we did the partnership with The Trade Desk. We recently announced a partnership with Google, a partnership with Roku where we can leverage this incredibly valuable data that we have to make advertising on other platforms more efficient.
And that’s something that we want to continue to develop with more platforms and scale with more advertisers. And then I would say the last step is actually expanding our advertising platform beyond online and into the store. And that’s why we announced recently add on Caper Cart. And obviously, it’s very early. We’re just starting to roll out a lot of Caper Cart. But we think that if we are the advertising platform that allows an advertiser to reach customers both online and in-store and have an advertising product in store that knows exactly what’s on your cost right now, which aisle you’re on, what you purchased in the past and your loyalty data, we can build a really incredible advertising platform. Again, still early, but that’s really the future that we’re driving towards.
Then on your question on Health and HSA/FSA. We launched it in Q4 as you know. It’s still very early but it’s very much kind of reproducing the mental model of EBT SNAP, where if you look at SNAP, SNAP has about $130 billion of funds loaded into SNAP every year. FSA/HSA $65 billion, of which $3 billion of those get forfeited by Americans at the end of the year because some of it is use it or lose it. And we think we can be a really good platform to remind them that they can use these dollars the same way we do with SNAP by leveraging that FSA/HSA code. Again, very early but something that we wanted to plan this season because we have seen such success with SNAP that we want to see if we can reproduce that with FSA/HSA.
Andrew Boone:
Operator: Thank you. Our next question comes from the line of Mark Kelley of Stifel. Your line is open.
Mark Kelley: Great. Thank you very much. Nick, a quick one for you just on take rates. As you think about the full year, obviously, you’ve got a couple of dynamics where you’re lapping that big Q1 of last year that you had where you’re kind of towards the high end or at the high end of your long-term range? And then just thinking through the RIF and taking those savings and thinking about marketing and incentives and things like that. I guess what’s a good starting point for us to think about take rates for our model? And then just a quick one for Fidji. You just brought up ads on Caper Cart. I’m just curious if you think, like are those budgets that are likely to come from, are those still retail media budgets? Are those like digital out-of-home budgets that you would might be able to capture? I guess, how should we think about that? And does that open up bigger opportunities outside of what you just mentioned in the last answer? Thank you.
Nick Giovanni: Thanks for the question. So, just on take rates, I think you’re referring to transaction revenue and ad and other revenue as a percent of GTV. And we do not guide to those specifically for the full year. What we did say is that we do expect adjusted EBITDA to be up not only in dollar terms but as a percentage of GTV year-over-year. And one of the reasons we said that is because we want to be able to make investments, whether it’s in the contra revenue line or the OpEx line to drive most efficiently, especially having just overhauled our incentive systems and having seen good results in deploying more incentives. When we decide to spend $1 in incentives and spend $1 in performance marketing, that makes it look like our transaction revenue is lower, but it makes it look like our sales and marketing expense is also lower.
And so net-net, it’s driving value and it’s just hitting in different places in the P&L. It’s important to also just remember that the efficiencies that we’re delivering, and we did call out that we continue to gain efficiencies in Shopper Pay, also benefits transaction revenue. But then we have a decision about where to invest that and how to invest that to continue to grow the business. So, we’re not providing overall guidance. We’re not changing our long-term targets with respect to transaction revenue reaching 6.5% to 7.5% of GTV or ads and other revenue reaching 4% to 5% of GTV. And we continue to believe that we can demonstrate real EBITDA leverage throughout the year.
Fidji Simo: And on your question on ads on Caper, again, important to realize that it’s very early and ads on Caper are a little bit of a different beast because they are really combining the best of online advertising, which is measurability, targetability but then doing that in an in-store environment. And so like any new innovation in advertising, we are probably going to start to see different advertisers take from different budgets to fund this. Again very early, but what we’re seeing is that advertisers are very excited about the combination of these capabilities. You can imagine going through the store and you’re dropping cookies into your cart and Dreyer’s Ice Cream can advertise to you that the ice cream can be found in Aisle 5 and then you can do an ice cream sandwich with the cookies you already have in your cart.
And based on your loyalty data, we might already know that you like chocolate ice cream, so we can direct you to that. It’s a really incredible experience that we can unlock, but it’s still very early. And the other important thing to remember is that we’re doing this hand-in-hand with retailers. We are sharing revenue on ads on Caper. We are doing it as a way to also create a new advertising revenue stream for them, and we’ll want to continue to partner with them closely on making sure that these budgets are incremental to what they’re seeing.
Mark Kelley: Perfect. Thank you very much.
Operator: Thank you. Our next question comes from the line of Deepak Mathivanan of Wolfe Research. Your line is open.
Deepak Mathivanan: Can I ask a couple on the cohort trends? Can you provide additional color on the rate of decline of the COVID cohort pre-2021 now with us maybe earlier in 2023? At what point can we expect the drag from COVID cohorts to be less and potentially reach some sort of like a stabilization level? And then secondly, can you also elaborate on the trends that you’re seeing in 2022 and 2023 cohorts in terms of the growth rate so we can think about contribution from the newer cohorts? Thank you so much.
