Why Basket Refuses Advertiser Revenue for User Trust

0:00This show is brought to you by the Global Talent Co, a marketing leader's best friend in these times of budget cuts and efficient growth. We help marketing leaders find, hire, vet, and manage amazing marketing talent for 50 to 70 % less than their U.S. and European counterparts. To book a free consultation, visit globaltalent.co. Hey, everybody. Thank you for listening. My name is Andres Figueira, and I will be your host for the day. And this show is brought to you by The Global Talent Company, a company that is connecting top remote professionals from Argentina, South Africa, and all over the world with leading tech companies worldwide.

0:38So if your team is looking to hire world-class South American or South African talent, or from anywhere really, just head over to theglobaltalent.co and see how we can help. And also here with me, I'm speaking with Lex Deek, the co-founder and CEO at Basket, which is an AI shopping agent with which they are reimagining online shopping by automating the hard parts and making it truly personalized. So Lex, how are you doing? It's a pleasure to have you here. Hey Andrus, yeah, I'm really well, thanks. Yeah, glad to be here. You find us at a very busy time where we are releasing a bunch of new features and educating the market and doing a ton of stuff.

1:18So yeah, really exciting times. Oh, that is great. And that is so fun. Congratulations, by the way. There's nothing better than to be expanding and to be getting bigger with what you guys are offering. And in case anybody who's listening to us isn't very sure about what basket is or they missed the intro, would you be able to give us more context about what is it that you guys do? Yeah, we started life as a smart wishlisting tool. If anything, you're thinking of buying, you can save to one place. and once you put it there, like a universal e-commerce basket, cart, whatever you want to call it.

1:56So everything's kind of away from random screenshots, from pasting links into your notes app, from just storing things in the static repository. It's now dynamic and all in one place, really easy to save anything to basket. It just uses the share feature for the most part from wherever you are in socials, in browser. So that's the first part. That's where we started off. And then over the past two years, we've added more and more functionality, more and more value for users. So we'll track prices and let you know when items that you've added go on sale. We'll let you know about stock levels. We'll automatically apply coupons for anyone that's used honey and that raft of products that'll be familiar.

2:44We'll find the best cashback opportunities. We'll do automatic price comparison. You can share your collections or lists with friends and family, which is great for gifting and also has application for creators and influencers. Yeah, a ton of things that you can do. We're now moving into this space since we're in the era of AI. It would be remiss of me not to mention, but we're really leaning into that because we're so well positioned to become this, we call it ambient action, this idea that it doesn't require an input from the user. It's happening behind the scenes, letting you know when the friend's birthday is coming up that you might need to buy gifts for.

3:29If you're going on vacation, are there any things that you might need to buy? Reminding you where you are and your relationship with consumers and being real, like the prophesied vision of a proper shopping assistant companion. That's the future that we're building into right now. Wow, that's crazy. And Lex, what is like the people's first reaction when you tell them that you're building an AI shopping agent? Are they a bit confused about the idea? Are they excited about it? Like how is the perceptiveness on behalf of the audience? Depends on the audience, right? Of course, a sophisticated tech audience, you know, AI is moving at such a pace that what last week was at the vanguard, this week is passé.

4:16So, you know, they're fine. It's very palatable for a tech-informed audience to get their head around. Consumers, you know, they are lagging behind. They don't necessarily understand the terminology because a lot of it is not very consumer-friendly. Ambient, a shopping agent. Let me just say basket, my basket. Yeah, exactly. In layman's terms, a tool to help people to be really in control of how they shop and to save money and to save time and to feel organized with everything. That resonates really well. People get that. They get that shopping has become increasingly disconnected and fragmented.

4:59And there's a lot of work involved in the pursuit of kind of meaningful shopping. And if you want to be practicing sustainable shopping, or if you're really price conscious, or if you like to do your research, or if you're a real kind of coupon, discount type person, or you're the sort of person that likes to talk with friends and get referrals into things. that's all separate and we're looking to bring it all together in one really cohesive experience. Wow okay okay and could you tell us a bit more about the features that you guys are going to be rolling out and what your vision for Basket is further more than just being that AI companion at the moment because I do feel like people create agents but one specifically designed for shopping sounds like pretty interesting.

5:47Yeah so the focus of agentic shopping at the moments is mainly around checkout actually you're seeing okay doing there's various efforts there to make the checkout more seamless our belief is that checkout is actually already pretty seamless if you're an apple user for example an iphone user you know apple pay is pretty good right maybe i can reduce one click from that process but kind of that so our approach is much more holistic it is around the research part, essentially a bit more jargon, unfortunately, but collapsing the funnel. So from the moment of you realizing, hey, I'm going on a hiking trip and I'm really keen for a new backpack to the actual purchase, everything that is in between that journey, if you can imagine that taking place from the research, the understanding, the watching of reviews, the aggregation of Reddit and subreddits about that product or that product suite, the price comparisons, sites, the tools that you use to automatically apply coupons, the card that you use to pay on because you get the best cash back, where you store the receipts, how you exercise any warranties, the whole relationship with that product.

7:01It's incredibly complex and it's nonlinear. It's just, it's a kind of matrix of things as well. Yes, it's true. Like the consumer spine just like goes back and forth. It is right. And especially with our attention spans, we've seen all the time, right? You know, there's a part of our mission, which is about conscious consumption. So, you know, I do it quite a lot. I'll be on Instagram, for example. I'll see some useless kitchen gadget or sweatshirt with a funky logo on or something that I'm like, yeah. And until now, you're faced with quite a binary decision. Either buy it now, because I'm probably going to lose track.

7:39Like, it's very transient. If I don't buy it now, I'll never get back to it. Or you don't, right? And if you're in the right state of mind, you'll buy it and then it comes and it's really bad quality or it's a terrible size and you're just like, oh, God. I thought I'd do this. I'm just going to contribute towards landfill. The ability to just really seamlessly save those things, to just consider it a little bit. And look, we don't want to take any of the fun out of impulse purchasing because there are moments when it's great. You might buy something in the heat of the moment and it turns out to be a fantastic buy.

8:10That's great. But I just think it's helpful in an age of overstimulation, instant reaction to just have a tool that helps you take a pause for thought and make a more informed buying decision. Of course. Okay. And in terms of how Baskit works, Lex, from how I'm hearing it, I feel it's more of something that's happening behind the scenes rather than how consumers actually just, I don't know, nowadays they have to go into OpenAI and then use ChatGPT because they want it. Baskit is something that you integrate into your experience and it just gathers information? No, it is very much a D2C interface.

8:49It's an app. It's an iOS app with a Safari extension. That's our flagship product. It's also Android app. It's a Chrome extension on desktop. It's a Firefox extension. The flagship, the product we're proudest of that's most fully featured is the iOS app with the Safari extension for mobile toggled on. Well the way that it typically works is you just need to wherever you are, if you're, let's just say I'm on Amazon for example in the native shopping app I see the bookshelf that I'm weighing up I just need to find the share icon in the same way as I would share anything typically across communications apps so take what's an example if I want to share that to my friend in WhatsApp I just tap share and then I choose WhatsApp from my share sheet that pops up and that's how that link is transferred to them via WhatsApp it's the same process for basket it just you choose the basket app as being the destination to share to.

9:49And that works for any product link, whether it's on an e-commerce store, whether it's from socials, wherever it might be. So that's the principle. That's the way that most people add things to their baskets. In an even easier way, if you have the Safari extension on iOS, when you're on a product page, it doesn't work in native apps, obviously. But if you're in Safari and you're looking at a Nike store, for example, every time you're on a product page, a little basket icon appears in the bottom left. Quite subtle, but that just allows you with one tap to save it to basket. So it doesn't interrupt your funnel at all.

10:27It's really seamless. And that's how it works on the desktop browser as well. So that's the main way that people add from other places. but then we're working on discovery within basket which will be similar to the experience that you'll get from chat gpt or whatever your chosen llm is which is where we're putting that experience inside basket so if i need a bit more advice or i need a steer or i want to ask more detailed questions what's the best rucksack over 35 liters that's waterproof under 100 bucks then i'll have a chat GPT-esque experience within Basket, but it'll be much better actually because it's built for shopping and integrated with a lot of our data and all of the context around who you are as well as an individual.

11:13So yeah, really powerful. Okay, that is so cool. So then the user is the one that decides what to feed the AI when you add it to Basket with the icon, right? Because I imagine that's how you can also avoid the ai recommendations from getting a bit creepy if somebody goes to look for a product that they don't want the ai to recognize that it could be kind of weird like you're showing your mom hey look at what i have and it says a wish list of something i don't know but is that how you guys i know what you want to say there but yeah it's really important it's relevant okay i think this is the other thing now when in a world of so much noise it's really important that things are really relevant.

11:53But also, you know, a lot of us are open to surprise, right? And don't really want to feed somebody's bias. You know, if all that I had saved to my basket was red t-shirts and black sneakers, you know, maybe I don't, you know, maybe I'm open to blue t-shirts and white sneakers, but it's a delicate balance and it's constant kind of calibration of the algorithms there and asking people for feedback. I mean, you can only assume so much at a certain point. You have to own it and say, you know, we think this is what you're into, but listen, please tell us if this sucks because we want to make it good.

12:27And I think a lot of tech companies have done that very subtly and they've kind of tried to hide that away because so many of their business models are served by ads and it's important that those algorithms run free and you find a little more menu where it's like, hey, this ad was or wasn't relevant to me. It's like, own it. So look, this is how tech works, right? This is how algorithms work. If you want really good recommendations and you don't want to complain about this product, you need to give it a little something. We can't just magic wand out of nowhere, understand who you are and your deep preferences.

13:01And, you know, there's a really interesting data science challenge there. And the industry as it stands has largely been structured upon a foundation of inferred purchase intent i.e andreas has been served an advert which you know for red sneakers and he left his computer to make a cup of coffee so the ad served 15 times therefore he must love red sneakers of course of course and it's weak it's thin so yeah there's a lot of improvement okay now i totally get that and i believe you guys are constantly analyzing customer data Have you seen any specific shopping behavior from consumers that surprises you that you say, hey, I didn't know people from this region would be buying this so much or something like that?

13:50We're only available in the UK at the moment. Okay. We are expanding internationally really soon. So that's exciting. So caveat, this is a focused locale. Yeah, we see, you know, people are really responsive to trends and popular products from social media, particularly TikTok and Instagram, particularly a Gen Z female audience, very responsive to what celebrities are wearing. There's no great surprise here, but very strong correlation between those things. People tend to buy towards the end of the month in fast fashion, health and beauty, because it's payday and it's blown up on these things.

14:34More expensive items considered for longer before purchase so the consideration cycle is significantly long particularly for higher-end electronics white goods luxury fashion actually tends to be quite impulse turns out yeah our female nail shopping bits are quite different females will tend to add a lot more things for consideration whereas men will add a few things which have obviously been through a filter in their mind before and I just do it's all for safekeeping. Yeah, it's fascinating insight. And one thing I find really interesting is, again, it's classic behavioral psychology paradox, but people, what people say is not what people do.

15:20The classic example is sustainability. If you ask people about, you know, how heavily they value sustainable shopping and buying from B Corps or whatever. They'll say, obviously, that it really matters to them, but then you can see what they're actually buying. It's Timu and it's Shein. I was about to mention they'll go and spend their whole paycheck in Shein or in Timu. Yeah, that's really, really interesting. It's an endlessly fascinating and deep space, though. It is so interesting that you guys are able to have those types of insights into the different demographics and then who buys what when you mentioned that guys consideration list is is a bit shorter than the usual woman's one i totally it's like yeah i need five white shirts some black shoes and a cap that's it if it comes i'm happy and women seem to have the ability to hold more in their kind of frontal cortex where it's just well i'm evaluating 12 different you know outfits or whatever else and mixing and matching and that they seem to be able to dynamically sort much better in that way which is really interesting And yeah, like the way that the ad industry is practically grouped people is so flawed.

16:34You know, just because you're a 14 year old girl, it doesn't mean that you love X, Y and Z brands and the color pink and blah, blah, blah. Right. And the way it's been working up until now is I'm a vendor, I'm a merchant, I'm an advertiser. I have a product. I believe that my customer is that group. Therefore, I will target that refined group who are all very similar, right? You choose your ad targeting preferences on whatever ad platform. And oh, yeah, I'm going after 14-year-old girls who like pink and who like Lulu. The reality is a lot of products, increasingly people are like individual. And it could be the case that a 14-year-old girl likes golf and actually has a penchant for cheese.

17:17and that just so happens to match a 45 year old male who likes golf and likes cheese right that now with traditional advertising you couldn't target those two same people in the same breath whereas you know i think in the future you'll be able to really dig down into people's kind of preferences and their interests and what you buy if you look through your purchase history is very similar in how exposed that makes you feel to looking at your browsing history. You know, if I look at the last 20 of your purchases, including, you know, groceries and other things as well, excluding boring things like utilities and blah, blah, blah.

17:58If I looked at what clothes you were buying, you know, what content you were consuming, what subscriptions that you had, all of this kind of stuff, that is a pretty unique picture, pretty revealing picture of who you are, which is fascinating. to. No, it really is. And seeing it from the perspective that you're painting it, it really does ring a bell because it is something as comparable as the browser history. It's like what you buy, you don't necessarily tell anyone. Then you then have these products. You could be advocate for sustainable fashion, but then yeah, all the clothes you have are from Shein.

18:31And it's like, nobody actually knows that. So you really get an insight into who people are. I just started to think about my last purchases and I'm like, I think I'm just a regular college kit buying a Nintendo beer. I'm like, oh my God. And like you mentioned at one point, cause I really am interested cause I'm hearing what basket is. And I'm just like, this could work so well for my girlfriend. She wants something new every five minutes, but if she would be able to get an AI to help her to like filter all that, it would be so cool. But the point to which I'm going is you guys are going to expand worldwide.

19:04So I could be able to download basket here in Argentina, for example. Yeah. Give us two or three months only before the end of the year and yeah some of the localized features may be restricted or maybe hidden but yeah absolutely like it's built for scale and shopping habits the behavioral psychology behind shopping habits is largely global i mean there are idiosyncrasies in each territory but broadly speaking everyone buys clothes and cares what they look like everyone needs to put white goods and furniture in their home. Everyone watches content on devices and electronics and likes to wear jewelry and goes on holiday.

19:42So yeah, these things are global needs. It's another interesting bit in the, I expect us to find in the local idiosyncrasies in how people shop. I imagine certain countries will be more considered in their shopping and other countries will be more frivolous. You both splurge a bit more. yeah you think so right and yeah that'll be really interesting to see how that data compares yes like i'm hearing you talk about the data and i'm like i could eat popcorn watching all these insights of the different data and different countries because it really is just so interesting and actually more than interesting it's really valuable does basket in any way have partnerships with other businesses or have you been pitched the idea of i don't know a business being partnered with baskets and suggested their products instead of others?

20:32Or is that a way that you guys want to steer away from? Yeah, we don't want to do that. It's really important that we're impartial. And we aren't selling any data at the moment at all or doing anything other than really just using that to provide a better user experience. We're very product and user centric. Can't help it just feel really, really strongly that it's important to build something that just awesome for consumers, you know, that's joyful, that's useful, that's relevant, that's not noisy. And it's also simple, hard, hard to do all these things at the same time. So yeah, we recognize there's a lot of commercial value in what we're doing, but we need to be very careful about how we unlock that.

21:17And that's one for early next year. No, of course. And I totally get it because more so with the product that is as personal as your own AI shopping assistant, for example, it's like you don't want it pitching you products that are advertised and not based on your likes or whatever you prefer because it takes away the trust that the application can generate in the user. And if you lose the trust with AI and such a personal app, then it's not going anywhere because people are very fickle. They can like a lot one day and then the next it's just like, nah, this doesn't work. It's all BS or whatever.

21:52So I get that. And I respect it as well, because in this day and age, it's hard to remain true to the vision and just like have the consumer in the forefront of the idea. When so many businesses, I imagine, would be dying to get their hands in this pie of just knowing that data. It's crazy. And Lex, do you have any specific go to market strategies that you want to apply when you guys are ready to expand worldwide or in whichever countries you have? Or like, how do you implement that in the UK? What's worked really well in the UK has been a combination of approaches, obviously, using nano micro influences and then creating UGC, which we've boosted has been really effective.

22:36Probably the lowest cost of acquisition channel for pure paid. Obviously, there's quite a lot goes into the management of it and the costs of maintaining that. But that's been good. We've done a ton of paid. amount of time so we've done a fair bit of paid across the traditional channels which is fine it's a bit dull it's okay for you know i guess lighting the fire and sometimes pouring a bit of extra petrol when you need but um pretty stayed pr's been an interesting one if you get the right people on board that can be great partnerships to some extent it's been good we have leveraged our own users to get a good chunk of our new users so okay basket's inherently quite shareable because particularly if you're creating lists for birthday christmas you know bumitz for whatever right then you're sharing basket with others and that acts as a really useful acquisition channel for us so that's been good influencers and creators using basket to share their own lists of products as well, which is like a micro-shop type interface, which is quite an interesting one.

23:41That's the stuff that we like the most, obviously, because it's the lowest cost and the most earnest. Of course. I think we'll do a similar in the New Territories. And just, I think the mantra is really just to test very quickly what works and what doesn't across all different channels, and then just to be able to adapt, to move and respond really, really quickly. We expect to learn, but I think we've got a decent blueprint. I think, it's only my personal feeling, but I think the UK is probably fairly representative of certainly the US and perceptions to consumption there. And how we can, and shopping there.

24:20South America, I don't know. You tell me, what would be your gut? I feel that South America, it does have a lot of like e-commerce and shopping impulses are very strong in these countries that are doing well. So it's not like if you were to put basket in Venezuela, which is where I'm originally from, I believe it would be a hit just because people don't have the acquisition power to even go online and think about buying something. But in countries like Argentina, Chile, Mexico, Brazil, even Brazil is a hell of a market. I think there, there really is an opportunity for basket to become something.

24:57If you get the application to just be in the language of those countries in Spanish and Portuguese or whatever, I think people could really like it because I haven't heard of anything like that here still. There's this really big enterprise, which is called Mercado Libre, which is almost the biggest e-commerce. It's like the Amazon of South America, just because they're in Brazil, Argentina, Chile, Mexico, everywhere. So if, I don't know, if Basket could be integrated with that, you let me know, guys, because that'll be amazing. Yeah.

25:32And Lex, as now as I've heard you speaking even more, that I hear the we in your sentences and how you're planning to expand. What's your team looking like? Like, how big are you guys? Are you looking for more people that help you out? Or since it's an AI assistant and it's mainly done, I believe, with AI, do you not require that much of a workforce? We are 12 at the moment and we're eight of those being engineers. We will grow the team as we expand internationally. I could see there being another eight hires over the course of the next year, probably on that side and no aspiration to build a huge engineering team.

26:11I don't think we need to. I think we also benefit from, and yeah, you guys will be very, very conscious of this, I'm sure. But if you are already a large organization, retrofitting AI capabilities in your engineering product team is quite tricky. And you run the risk of getting rid of tons of people needing to hire them back. And we've seen that quite a lot. If you're really small and lean and you grew up on using these tools to be more efficient and increase your output exponentially, then it's a lot easier to just like to keep doing that without being, all right, we don't need to go from eight to a hundred, like you would have done back in the day.

26:52If we went from eight to 16 and with the compounding use of AI, that's effectively probably comparable to going from eight to 50 five years ago. But a lot easier to do if you are lean and mean and you haven't got the legacy of a massive kind of structure and an HR base to consider around you. So yeah, We'll take it slowly. We won't be hiring for the rest of this year, but early into the new year, we expect to. And hire slow. The cost of a bad hire, the cost, not just financial, but on the culture, on momentum, on these thousand, is real. And we've done that a few times, and you're just like, damn.

27:39I like that. Yeah. So taking time to get to know people, and more and more, actually, it's less about the technical skill, more about the attitude and the level of curiosity. So we live in a world where no question needs to be unanswered. Like if you are curious enough and if you are self-motivated enough to be using the tools to satisfy your curiosity, then, you know, what was it, you know, junior becomes a heavyweight in a year, right? If they want to, if they care enough, if they committed enough, curious enough. I do worry for senior engineers and devs who are really stuck in their ways. I think they'll become antiquated really quickly.

28:23Yes. I totally agree with what you're mentioning because it is something that I'm seeing and especially doing this podcast with so many different businesses and stuff. If companies got too big before this AI wave, now they're trying to hop on the wave, but it's also kind of, I think of it, they're like in this moral ambiguity of if I want AI, I need to fire my workforce and it's then downsizing to then be better with AI. And if you guys are getting it right since the beginning and just staying lean, leveraging AI and just growing when you need to, I think that will add a lot more value, as you mentioned, in the future, because a junior that leverages AI properly and is curious and does stuff in a year, they'll be a senior and they could even achieve even more than that.