How Recall.it grew to $3M revenue and 22.5K customers in 2026.

Nathan Latka (00:01) Hey folks, my guest today is Sankri Naya. She's the co-founder and COO of a company called Recall, one of the fastest growing AI knowledge platforms, management platforms, helping users turn information overload into lasting knowledge. Before this business, she spent seven years helping scale Uber during its hypergrowth era and later worked at Cloud Kitchens. We have a we have a Travis, we have a Travis original in the house today, bringing world-class operational experience to the AI startup world. Sankri, are you ready to take us the top? Sankari Nair (00:23) You do. I am ready. Thank you, Nathan. Nathan Latka (00:31) You bet you spent a lot of time with Travis at Uber and CloudKit. What makes him so special? Something you've seen that you feel like no one else, you know, gives him credit for. Sankari Nair (00:39) I think two things. Like one, he really is a visionary. Like he's had grand ideas. And two, he's so committed to seeing them through. ⁓ very hands-on. From the very beginning, Uber was very transparent if you were inside. ⁓ we had an all-hands meeting, all questions were answered. There was transparency around the metrics, and I saw that, if not more, at Cloud Kitchen when it came to being in the weeds, understanding what's going on. So I think the combination of being a visionary. And then still being able to execute and get into the weeds ⁓ is really ⁓ quite the combo. So so that's what I would say. Nathan Latka (01:16) Now you're doing your own visionary thing. The AI space is very hot. I think all of you listening right now are guilty of what I'm about to share. We take our favorite podcast episodes, maybe we download them on YouTube and then forget about them. Or we take a really long voice note in Apple notes when we're waiting to board a plane because we want to remember important conversation. Or there's that book we read about how to build a great system. And we're like, man, how do I get a PDF of that so I can use it in my future knowledge? And then we lose all this information. Sancry, how are you fixing this problem? Tell us what recall does. Sankari Nair (01:19) Yes. Mm-hmm. So recall is a self-organizing personal AI knowledge base. So you can save all of that content you consume: your YouTube videos, your podcasts, your PDFs, even your own notes. ⁓ Recall organizes it for you, categorizes it for you, and then as you can see here, feeds it into a chat. So it actually becomes the context for your conversation. If you think about it right now, Nathan, you head to Chat GPT for an answer. That context is the internet or maybe the history of your past conversations. ⁓ Maybe it's a document you uploaded. With the recall, we're prioritizing your knowledge. We're bringing your knowledge to the forefront. And so we're giving you answers from your context. ⁓ It also does a web search. So if you're missing something, it will go out and search it for you. ⁓ And overall, it's really giving you the peace of mind that the content you care about is in one place, organized for you, ready for you to interact with. And becomes part of the conversation, so that is recall. Nathan Latka (02:47) People say you sort of become the pe you know the five people you hang out with the most. In the age of AI, you sort of become whatever content you're curating and that you look at every day. So this is like very, very important. My question to you is: I will sometimes save a Paul Graham speech and then some other investor, their speech, and the speeches directly contradict. So when I ask recall a question, how do you know which one to give me the answer from? Sankari Nair (02:54) Mm-hmm. Mm-hmm. Mm-hmm. Mm-hmm. Mm-hmm. Mm-hmm. This is such a great question, and this is where in recall we have a mix of models. So we'll prioritize the model based on your question, or you can choose the model. Now, with some of the latest models out there, it will ensure that it's citing the source. So when you ask your question, it will say, This came from Paul Graham, this came from another podcast, this is where they agree, and this is where they contradict. So we serve the answer. And we cite the source. So when you think about perplexity, ⁓ where it's all about citing where those answers come from, we do the exact same thing, but with your own knowledge. You know, we'll even find the exact moment in his podcast or his blog where something was mentioned. So you can even go and reference that and even maybe have some delight by reading that full paragraph or listening to that little snippet. ⁓ so that's the way we ensure that. mm, the knowledge is grounded and citable, but you can also go back to that original source. Nathan Latka (04:10) My audience is listening to this going, hmm, I wonder if this is working or not. Are people actually gonna use this? I don't wanna bury the lead. Are you able to share a user number? Are are people using this thing? Sankari Nair (04:19) So I'll share a couple numbers and I know Nathan you like to do some back of the envelope calculations, so I'll let you do that. ⁓ but we've had over ⁓ 750,000 signups to recall. we have a very healthy user base. we've grown around ⁓ four X in the last year. and we have some a mix of both, ⁓ a mix of both. Nathan Latka (04:26) Ha ha ha. That's revenue growth or user growth? Okay. Sankari Nair (04:49) And we also have some very ambitious goals for this year. So we have a very if if you're interested, we have a Discord community which you can join and just be a part of the conversation. And it's just really fun listening to users, bug bashing and actually developing the product alongside our community. So having a lot of fun ⁓ doing this. Nathan Latka (05:11) So seven hundred and fifty thousand doll signups is obviously not a small number. Give us the backstory here. What day did you launch and how did you get your first thousand signups? Sankari Nair (05:20) Yo, it it it is rough. It is sometimes a journey and it is not just this beautiful I launched and people went crazy. So the story actually started with a Hacker News post. And so I'm the COO and our CEO had started recall. He he actually had a version which was primarily for storing Wikipedia articles. So it wasn't as rich as context as it is now. He posted it on Hacker News. ⁓ the post went something like a place to a store all the shit you want to remember. The post goes viral. And this was sort of like his last resort. He wasn't sure, you know, is there any traction? Is there anything out there? Post goes viral. Within eight hours, he's on the call with an investor who gave him his first check. And that was really where the story began. ⁓ it wasn't that big. I think it was around 100K. nothing crazy. Really nothing crazy. Just like, hey, I can quit my I can quit my side hustle. I can go all all in on this. This is a real this is a real thing. That is Nathan Latka (06:04) How big was that check? From this from this post. Sankari Nair (06:19) Yes, this is the post. So November 2022. ⁓ This was the post that started it. And you know, the thing about the story, it is magical. When you post something out there, it resonates and it could literally turn into a check. but that doesn't always happen like that. So there's a bit of luck in here. This ⁓ this was a lucky starting point for recall. But it wasn't all, you know, it wasn't all highs from there. From there, the product was iterated on, ⁓ expanded upon. And where things really started taking off is with Product Hunt. So that's actually really how we got the first few users. We had a Product Hunt launch. I'm a big product hunt fan. So I think if you're in B2C, I'd highly recommend that being something to play around with. ⁓ Launched on Product Hunt, got the first few users. Things were still a little bit flat even then. And then we had a big launch in June. So the June launch, if you scroll down. Nathan Latka (07:12) Hold on, hold on. I'm trying to get visuals to follow along with the story here. Was it the June 10th launch, 2024, a thousand upvotes? Sankari Nair (07:15) yeah. Yes. Yes. This was a really big milestone for us. this led to a lot of growth in twenty twenty four. ⁓ product of the Nathan Latka (07:24) Can you quantify that? Like how many signups do you think you got if you had a back of the napkin it? Sankari Nair (07:28) We had around 27x growth in 2024. ⁓ in terms of numbers of users who signed up. That's right. ⁓ and this this really helped. ⁓ and this was a bit of a bet of a feature. So basically in recall, you're saving content, you're interacting with content. And the idea with this particular launch, it was about resurfacing content through a spaced repetition quiz. So you might be thinking as like, Nathan Latka (07:32) Okay, and then number of users signed up basically. Sankari Nair (07:55) Quiz, you know, why do I, you know, why do I want to quiz? Well, the truth something like that, I mean, it looks a lot better today. Looking at these images is giving me anxiety. we've come a long way from these images. ⁓ but basically the concept is we have a forgetting curve. And if you don't reinforce information at a specific interval, you will forget it. In fact, the brain is designed for you to forget it. ⁓ but by resurfacing it, you can offset that forgetting curve. Nathan Latka (07:57) Like this. Ha ha. Sankari Nair (08:25) And so in recall, when you save your content, you can generate a quiz and we'll resurface that content to you. ⁓ so if you keep getting something correct, the less you'll see it. If you keep getting something incorrect, we'll keep resurfacing to it to you until you start getting it right. So this was actually a really important launch for us and and really helped with with those first few customers. Nathan Latka (08:48) Okay, so th it sounds like your first ten thousand signups was a combination of Hacker News plus Product Hunt. Is that accurate? Sankari Nair (08:54) That is completely accurate. And the other one I want to mention is Reddit. When Google started indexing Reddit, and then obviously the AI chatbots come in, they're also indexing Reddit as a source. That was a really big moment for us. In fact, about 30% of our growth in 2024 and 2025 came from Reddit. There were a lot of conversations in the community happening about recall. there were posts from Paul, the founder, saying, Hey, I'm the founder, this is what I'm launching. ⁓ specifically there was focus around some keywords, like summarization was really trending around that time. ⁓ and that's when things really started to take off. Nathan Latka (09:35) How much of it is just your brand name? It's called recall. Sankari Nair (09:38) That's the problem. I think if you go back one, you would have seen a big post around Microsoft Recall. I think it's like the fifth post. So it was tricky. So we launched Recall, and then Microsoft Recall had actually come out. ⁓ different products, but playing in sort of the memory, knowledge, retention space. And there was so much backlash with Microsoft Recall and there was confusion, you know, we don't want recall, it's got this bad reputation. So that's actually been tricky. Nathan Latka (09:43) I saw that's why I asked. ⁓ Sankari Nair (10:06) since then ⁓ we bought a new domain. ⁓ we have brand Google brand search ads. That's something I highly recommend if you're facing the type of conflict we are with is naming issues. ⁓ and so that was a tricky one. The the naming was a little tricky. Nathan Latka (10:20) Can I ask how much you guys are investing in in paid spend on those brand keywords today? I mean, are we talking like ten grand a month like, you know, a hundred grand a month? okay. Sankari Nair (10:26) No, no, no, no, nowhere near that amount. Honestly, it's closer to around a thousand dollars a month. It's not a lot. for specifically anyone can do this. No, anyone can do this. And I highly recommend doing a brand search because the worst thing is if a consumer's looking for you and they can't find you, that's frustrating. And that could have been a quick purchase for you. So highly recommend an affordable brand search ad. Nathan Latka (10:32) So any guys, anyone can do this. That's why I asked, right? Anybody can do this. How much of your growth is coming from the desktop website versus the very smart integrations you've done with the app stores like Google Play? Sankari Nair (10:57) the majority comes from desktop. ⁓ the mobile app has almost half the conversion of our web app. And the main reason is on desktop we have a really powerful browser extension. ⁓ so our browser extension, Chrome. Yeah, it's Chrome, it's Firefox. ⁓ the community has really been asking for Safari, so that'll come soon. ⁓ and for this one we've been having ⁓ this just is I love the extension. I use it all day. Nathan Latka (11:11) Chrome. Mm-hmm. Sankari Nair (11:26) we've got great feedback on it. I can't watch a YouTube video or a blog without clicking the extension. So ⁓ this is why the conversion is just so much better. In fact, a user who installs the extension is twice as likely to convert into a paying customer. Nathan Latka (11:40) Interesting. So what i I mean look, that's a big debate right now, right? Is like what is the new in the age of AI, what's the good sort of free-to-paid conversion rate? Are you comfortable sharing what that actually is for you guys today and how you're trying to obviously drive it up as much as you can? Sankari Nair (11:53) Yeah. Yeah, I can because I think benchmarks are tricky. ⁓ we were almost stagnating at around three percent. At the moment, we're at five percent. And whenever we have a big launch, it really skyrockets depending on the channel, between eight to even thirty percent we've seen on some channels. So five percent is the stable rate for now, but our goal is to get it to eight percent. ⁓ and I actually think there's a lot of room on the table to to optimize. the way we get for there that you know we haven't even fully focused on optimizing our funnel and so I think there's a lot we can do for sign up to paid. Nathan Latka (12:27) Well obviously the big the the most friction on that is what's your obviously price point. So this is probably a good moment for you to go deep on how you guys set this this pricing and are there edits that you guys are testing right now because you feel like it's not optimized yet. Sankari Nair (12:38) Yes, this is a new price, and this just launched in April. So before this, we only had two pricing plans. That was at $7 a month bold yearly and ten dollars a month bold ⁓ monthly. ⁓ because of the AI costs, we had this trade-off. We wanted users to have a better experience in recall, so we've upgraded our models. They are more expensive, which means our cogs go up. Nathan Latka (13:01) Mm-hmm. Sankari Nair (13:04) And so we did have to do a slight increase. So now we're at $12 a month for plus. This is for lifelong knowledge management. It's everything you need. And then we we did just have these users that were pushing the limit of what was possible in recall. They really wanted bulk actions, they really wanted the frontier models, and that's why we launched Max. Pricing is tricky. We did a lot of research on the price points, ⁓ competitive research, surveys to our consumers. It's still early days. It's just been about a month since we've launched this. And so, ⁓ yeah, we have room to optimize this. And honestly, the mindset is I would love everyone to have access to recall in a high quality way that they're getting great answers that they can trust. We want it to be affordable and to bring it to as many people as possible. ⁓ it's not just about having these really expensive fancy tiers. for fun the fun of it. So we spent a lot of time looking at models, costs, and optimizations to try and make it as affordable as we could based on the best experience we could give. Nathan Latka (14:06) And so Sancry, is this twelve dollar per month paid monthly plan? Is that an increase or a decrease of where the pricing for the same plan was, you know, six months ago? Okay. Sankari Nair (14:13) It's a s it's a slight increase. So it was ten dollars and now it's twelve. Yeah. Nathan Latka (14:18) So is it fair to say then your legacy like paying customer base since this change just happened, they're all paying around that $10 per month mark somewhere in there? Sankari Nair (14:25) Yeah, that's right. And we did grandfather our pricing. so we did want to make sure that for another year anyone who was on the previous pricing ⁓ got to keep that. Nathan Latka (14:34) Yeah, that's otherwise you have a revolt on Reddit and product and they're all they did they turn against you. Sankari Nair (14:39) Exactly. And I also get it. We also just love our community and I just I I want to peop people to be excited about all the features we included in this big two point launch and not be so thrown by the price. And so yeah, we it's the right thing to do, yeah. Nathan Latka (14:51) Yeah. That makes sense. And can I take the 750,000 historical signups you have times that conversion rate? You know, the worst one, let's do three percent. You said the base is now higher, but let's do the worst is three percent. I don't want to put too much pressure on you here, but can we say you have more than twenty thousand paying customers at this point? Okay, okay. Do you want to be more specific? Sankari Nair (15:02) Mm-hmm, mm-hmm. Mm-hmm. We can say that, yes, we can say that. Directionally we can say that. I won't be. I've got strong instructions on what can and cannot be shared before joining this podcast. And I knew there'd be some calculations, but I also know you're great at your math, so definitely in the right direction. Nathan Latka (15:23) Why'd you why'd you say, you know, I love that you come on because my audience is learning? Why did your team get comfortable saying yes to have you come on? Because you knew I was gonna ask these questions, obviously. Sankari Nair (15:31) it was a great opportunity for people to learn and I really wanted to tell an honest real story because I have had a journey and if I can help anyone and just bring a real experience that it's not always roses but it's so rewarding, ⁓ I wanted to share that. So I was just excited to tell a real story and if it could help someone, help them in their day in any way, happy to do that. Nathan Latka (15:55) Give me more of the backstory here. Was there ever a moment where you guys considered shutting the business down? Sankari Nair (15:59) that was actually in 2023. so after that first check came in 2022, there wasn't a lot of traction in that year. And so Paul he had our CTO Ego on board. I was helping from the sidelines, but I hadn't fully committed. And we were like, is this gonna work? We don't actually know. And we got into Jason Calicanus' launch program. So that was a pivotal moment for us. we got into that program. Things actually started turning around for recall, especially when we launched that browser extension. ⁓ And we had we're part of the accelerator and then we had that big launch in June, which you just shared on product hunt. And then we closed a pre seed loun round a precede round with Jason. ⁓ so so that was a tricky spot. And I don't think it was ever shut down. It was more like, ⁓ shit, I need to pay some bills. I gotta find some extra income. But I believe in this thing and I'm gonna keep going. But it was really a blessing to be able to go all in. I can't imagine doing something else. like so so much joy in here. Nathan Latka (16:59) So the funding history then would have been the 2022 hundred K check after the Hacker News post and then a pre-seed round after the accelerator in 2024, is that right? Okay. And what was the size of the pre-seed round? Okay. And looking back now, was that the right amount? Was it too conservative? Did you like the valuation of the cap, or would you wish you negotiated that harder? Sankari Nair (17:05) Mm-hmm. That's right. Yeah, and that was it. one point six. I think we raised enough. I'm quite I my background's in analytics. So backgrounds in engineering. I've been in analytics for the last 10 years. And so we're taking a very experimental, intentional approach to growth. Look, if you have a lot of capital, you can just burn it a lot faster on like crazy experiments and feel good about it, or just burn paid ads and be comfortable with it. ⁓ but I actually think based on the amount we raised, we have a great team. ⁓ we're actually profitable, break-even profitable. ⁓ and so when we do spend it's on a month where we're like, Hey, we wanna take a bet. We wanna spend and burn this month. So yeah, I think I'm really comfortable and I thought it was an incredible experience being part of the accelerator. ⁓ I'm still close with the team. Nathan Latka (18:05) What was w what was the biggest bet you took over the past twelve months measured by how much dollars you spent on the bet? Sankari Nair (18:12) we spent a shitload on two I don't think I can say shit. I'm sorry, maybe cut that one. I'll go again. I'll go again. We spent okay. We spent a lot on ⁓ YouTube influences and paid ads. And YouTube influences, dedicated YouTube video influences have converted incredibly well. I expanded into more integrations. Nathan Latka (18:15) I love that. I wasn't ex I wasn't expecting you No, that's gonna make it so good. Sankari Nair (18:40) so that's just when someone gives a quick mention of you, maybe in the pre-roll or the mid-roll or in roll, and that did not convert well. And our paid ads we've turned off. So the bet was like going all in on YouTube. the learning was integrations don't convert well for us. Maybe it's for brand exposure and that's fine, but it's not gonna give you dollar immediately. And paid ads, it's a tricky game and you have to be prepared to burn. So I think we wanna optimize our fun a little bit more, make sure mobile's working a lot better. Have healthier rates on mobile and then revisit paid ads. Nathan Latka (19:13) For someone that wants to do the same YouTube strategy you use, but in their own niche for their own product they're selling, can you maybe give the breakdown here of like Andy, right? I see he's got your code here, so you're tracking his traffic. It's 10, look, it's not a million views, but you don't need a million views, right? There's 10,000 views. So break down the example. Tell people how they could find their own Andy Stapleton. Sankari Nair (19:19) Mm-hmm. Mm, mm, exactly. Yeah, so first I mean think about your or if you're on YouTube, what problem is your audience facing where that channel's gonna convert for them? So that's like one thing. The second is do they have a trusted audience? Does this audience trust that influencer? That's the most important thing. And then the third is also just relevance because here someone who's watching Andy, they're interested in AI, they're interested in knowledge, they wanna they're going deep into research and they happen to be on YouTube and recall works. really well as a YouTube video summarizer or to save your YouTube videos. So that's how we cohorted it and found it. Then I tend to look at views. So views is one of the most important metrics. If I looked at any of his other other videos, I want to look at views, I want to look at comments, and I want to look at likes. You have to be careful because there's actually a lot of influencers where they're just buying views and then when you go in there's no real comments. There's no real likes and then you would have spent your money on nothing. So based on this, I Nathan Latka (20:31) So you're looking here, you're like, wow, okay, not only does he have 11,000 views, but he's also got 51 comments. He didn't buy 11,000 views, it's real engagement. Sankari Nair (20:33) Yes. Yes. And I'm reading the comments. So I will also go through and read the comments and be like, are these real comments? You can very quickly get a feel for like flame, flame, flame, great video, great video. Like it's not a real person. So I'll check that. And then I know it's legit and it's a good fit. And then you need to test a couple and you end up having a formula where you're gonna be able to predict roughly what your ROI is gonna be on the investment of the influencer. So I just have a big calculation where I look at likely clicks, likely sign up to pay likely paying conversion. And then I do a quick calculation to see is this worth the investment. Nathan Latka (21:10) And so what do you pay Andy? Is it like a flat fee per month or is it commission based only on sales through his link? Sankari Nair (21:16) It's once off ⁓ and it's flat regardless of performance. that's one model that you can have. So it is a bit of a gamble. Another model is some ⁓ some creators we work with, it's a mix where you can pay flat and you can also pay affiliate commission. So depending what they actually bring in. So you can really negotiate this. But this dedicated YouTube video channel has been a game changer for us. So highly recommend if you think it can work for you, like do the hard work of finding the right creators. The hard thing is getting a hold of them. Very hard to actually get a hold of the creators. ⁓ like we ramped up really slow, but now we're in a great space with great relationships with a lot of the creators or their agencies. Nathan Latka (21:55) So what I mean, just give me a sense, obviously, don't share confidential information, but for somebody like Andy, I mean, are you paying 5K flat there, 50K flat fee? Like what does that look like? Sankari Nair (22:04) I would say the range on a creator like Andy could range between five to fifteen K. ⁓ it really depends on the channel, the video, ⁓ whether it's an integration or something dedicated. You can definitely get something around five K. Maybe with smaller channels, you can even start playing around with about a 2k budget, but hard to go below there. you might not be getting the necessary quality that you need. Nathan Latka (22:30) Okay, so like today, we're recording this in June of twenty twenty six. If I opened up your books, your PL, I would see paid ads is zero, you shut it off, but your influencer marketing is gonna be your okay, just a hundred one one grand per month for brand search. But your biggest marketing line item there is going to be ten, twenty, thirty thousand dollars paid out to five, ten, fifteen YouTube influencers. Sankari Nair (22:34) Mm-hmm. Mm-hmm. Mm-hmm. One one br one brand search ad live. Yeah, in fact we did fifty YouTube influences last month. ⁓ and that's where actually I'm saying we spent a lot, but we also had learnt some hard lessons. Nathan Latka (23:02) So what is I mean, fifty K times five K pop, I that's two hundred and fifty grand, right? Am I doing my math right? Yeah. Sankari Nair (23:06) Yeah. No, that's right, that is right, and that is ⁓ roughly what we spent. Nathan Latka (23:11) my god, does that make you nervous? I mean you've only raised one point seven total. I don't I I don't know how if you're sitting on a bunch of profits in the bank, but Yeah. Sankari Nair (23:15) no, because we we made a lot of it back. ⁓ so th that's actually why but the thing is I got nervous at the start. I'm s I'm South African, you know, why I grew up not spending th this type of money, I name it to my mom, she's gonna she gonna lose her mind, you know. Like it's hard to even get comfortable spending these ⁓ these types of numbers. But it took us a long p a long time to get to this point to feel comfortable spending this amount of money on influencers and we really started small one at a time, one a month, three a month, like We really are meticulous and we only do it when we actually feel confident that there's gonna be an ROI here and this isn't just a big bet and a big burn. Nathan Latka (23:51) That makes sense. Okay, well, ⁓ talk to me as we wrap up here. It's now June. Well we just finished obviously last month, but if we look at where most of your signups are coming up from today, what's like the monthly cadence like? Are you guys signing up a hundred thousand new users last month or 10K? What's it look like? Sankari Nair (24:05) Nathan, it actually was a tough month with two things. ⁓ one, we changed our domain, so that's still warming up. ⁓ and that's a warning to founders to just make sure like there's a warm-up time when you change a domain, and a lot can actually go wrong. Make sure your metadata translates your schema, it translates. ⁓ and when it comes the the second thing is the world is changing when it comes to direct traffic, especially with AI overviews. so with these two combinations, it's hard to give a stable number. But we have a very ambitious growth goal, so I would I will say that. Nathan Latka (24:39) What is your goal for twenty twenty six? Is it a user goal, a revenue goal? What do you hope you hit by the end of the year? Sankari Nair (24:44) It's a 10X revenue goal. and that's our goal for this year. Year over year that's our goal. ⁓ so there's a a lot we need to to do, a lot of homework, but we're excited and I honestly feel great about the product. and you know, based on our users and our community, ⁓ I'm confident. but yeah, it's it's about having a big goal, working hard for it and having some resilience when maybe don't reach what you hope for, but keep keep coming back and keep fighting. So Nathan Latka (24:47) Year over year. We're five months through twenty twenty six. Are you on track to hit that ten X goal? Sankari Nair (25:17) Not quite, but I will say that the trajectory looks good. We're definitely not quite where we want to be. There's some big things that will be rolling out this year. ⁓ so ⁓ more to do to get us on the exact track to feel confident to answer that question. Nathan Latka (25:31) And your team prepared you well for this, so I'm not even gonna try and be creative in getting your revenue number out of you. Can you share anything about revenue? Sankari Nair (25:39) Mm. Hard to say. I feel like I gave enough benchmarks. So I I feel like there's enough I've shared in terms of some back of envelope calculations you've done. but yeah, I I will leave it at that with revenue. Nathan Latka (25:54) Okay, let me do the calculation because you know I'm gonna do it and then you can sort of give a reaction, right? 750,000 signups at a poor conversion rate of three percent would be twenty more than twenty-two thousand five hundred paying customers. And using your old cheaper legacy pricing, your ARPU's higher now I imagine, but that would be ten bucks per user, that would be you're north of two hundred and twenty-five thousand dollars a month using that math. If I just leave it at that, is that a fair statement? Sankari Nair (26:17) I'll say there's ranges in each of your calculations, so I think you could have buffers both up and down, ⁓ but I don't think your math is completely wrong. Nathan Latka (26:26) Is that code for you saying plus or minus ten percent? I'm in the right ballpark. Sankari Nair (26:30) I'll I and I I'm gonna I'm gonna stop at that one. Too too far. That's my limit now. Too far. Nathan Latka (26:33) Fair enough. I I'll respect that because I appreciate your time coming on and you taught us a ton about how you're thinking about paid ads and everything else. Last question before we wrap up. Any new products you're really excited about that you want to sort of leak here on the show and get people excited to go check you out. Sankari Nair (26:49) Yeah, absolutely. So we I mean, we had a massive launch of Requal two point O and the amazing thing here, I don't know if anyone was part of that trend with like LLM Wiki, Andre Carpathi did a whole big tweet about it. It was about bringing your knowledge into your chat bot. And so I think with Recall's Agentic Chat, the ability to chat with your knowledge and the internet is insane. I use it all day...