Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
Building a D2C Growth Platform | Swagat Sarangi @ Smytten
238 Plays3 years ago
In this episode of Founder Thesis, Akshay Datt speaks with Swagat Sarangi, Founder of Smytten, one of India’s leading product discovery and trial platform that is revolutionising the way consumers interact with brands.
An alumnus of Indian School of Business, Swagat worked with Google as the Head of Business Marketing before starting his entrepreneurial journey. His extensive 17 years of experience in varied industries helped him understand the pain points between brand-consumer interactions. Founded in 2015, Smytten facilitates and shapes the way new brands discern their consumer landscape.
Tune in to this episode to hear Swagat speak about Smytten’s interesting tech-enabled business model that is disrupting the online product discovery space!
What you must not miss!
- ROPO model
- Use of AI for customer acquisition
- Powerful B2B flywheel of the business model
-----
If you want to be a podcaster, then check out Zencastr.com which is also the generous sponsor for this episode. Zencastr is like a Shopify for podcasters taking care of all your needs from recording to editing to hosting.
Show your love for this show by using this link to sign up: https://zen.ai/founderthesis
Recommended
Transcript
Introduction and Zencastr Shoutout
00:00:00
Speaker
Before we start today's episode, I want to give a quick shout out to Zencaster, which is a podcaster's best friend. Trust me when I tell you this, Zencaster is like a Shopify for podcasters. It's all you need to get up and running as a podcaster. And the best thing about Zencaster is that you get so much stuff for free. If you are planning to check out the platform, then please show your support for the founder thesis podcast by using this link, zen.ai slash founder thesis.
00:00:27
Speaker
That's zen.ai slash founder thesis.
Smitten's Disruptive Market Research Model
00:00:32
Speaker
Hi, I'm Swagit. I am the co-founder of Smitten.
00:00:45
Speaker
If you speak to any of the top marketers in the country, they will tell you that marketing is more about data than creativity. In the good old days, they would get access to data through market research agencies that would do in-person interviews or telephonic surveys and it would take months before a brand owner would get insights about his brand's performance. Smitten is a startup that is disrupting this decades-old industry along with many other industries.
00:01:11
Speaker
Smitten was founded by Swagata Sarangi, an ex-Googler who helped lead Google's early efforts at building its business in India. Smitten's incredible business model starts with acquiring customers for driving product trials and then leveraging the customer base to run a B2C e-commerce business and a B2B brand marketing business. Here's Swagata telling Akshay about his journey of building Smitten.
00:01:36
Speaker
This idea was kind of brewing up in my mind. So while I was like a big believer of digital, I realized that while digital will democratize access, digital will not actually solve for the adoption issues, which a lot of brands are going to face, or at least still not solve the paradox of choice for a lot of consumers, rather it's going to make them even more confused. And that's why people will move away from offline transactions.
00:02:06
Speaker
That is going to be this whole void of having still a physical touch and feel of the product and trying it yourself. And then if we look at them in the supply side, there is one side of the equation is your traditional brand houses. The other interesting thing that was coming up is the whole new H-brats. Yeah, I mean, D2C is a term now, but when we started, we said, okay, hey, there are a lot of these new brands that will come in.
00:02:36
Speaker
Because two or three things were happening, right? Your distribution was being democratized through all these marketplaces and all. Your access and reach was being democratized through the Googles and Facebooks of the world through digital. So India is kind of starting up. And while these brands will have an access and distribution, these brands will not have big marketing dollars to be building the brand.
Challenges in Digital Marketing and Smitten's Consumer-Centric Approach
00:03:03
Speaker
And the only thing they will lead is our product innovation.
00:03:07
Speaker
And how do you convince a customer on a product innovation without giving the customer an experience? So one side of the supply was like the traditional brands who always kept saying, we want a meaningful engagement. The new age brands who we knew that it's going to be extremely difficult for them to acquire customers without giving a real product experience. The demand side again, when you realize that, okay, there is a set of consumer who is constantly living it for more.
00:03:36
Speaker
This is the new age, millennial, new age shoppers who we talk about, right? The digital natives. So the digital natives, they are kind of, and we used to have a term called research online purchase offline. So which was AROPO. So I mean, suddenly you saw a kind of explosion of research that is happening online for a particular product or a category or whatever it is.
00:04:01
Speaker
And then we said, okay, there is so much of a research happening and there is so much of a confusion also that's being created. And the consumer is constantly looking for validation, they're constantly looking for new things and all that. So how can we kind of give them kind of a quick, I mean, click button kind of outlet where they don't have to go to a physical retail or kind of look for things what they want.
00:04:28
Speaker
I mean, would consumers pay for it or would brands pay for it or like, what would be the mechanics of this? See, we actually kind of started with a thought that brands should pay for it, right? Because sampling traditionally has been always free in the consumer's mind. The openness of getting your product is always with the brand. You produce, you type with channel, you just give me a wallet over a product too.
00:04:57
Speaker
But very soon we realized that there is a reason why the sampling also hasn't scaled. Because at the end of the day when there is a sampling happening,
00:05:08
Speaker
There is a value being created on both sides, right? So the brand is able to use, let's say, first experience or a first trial of their product with an relevant audience and hope that there is a strong intent created. And for a user, the user is looking for a product for certain concern or a certain solution.
00:05:28
Speaker
And the user thinks that, okay, this product could be for me and hence getting a smaller trial back and all. That means either you are getting a validation, your decision-making process is becoming easier, or you are avoiding a bias regret, which like you would have impulsively gone and bought something and you saved your money, let's say. And so we say that, okay, why the load always has been shared by the brands?
00:05:54
Speaker
We need to figure a way out. And then the moment that load, all the load goes to the brand, that there is this whole investment gap and that ROI and all those kinds of discussions coming.
00:06:06
Speaker
The other thing which I also think, I mean, when brand pays for sampling at point of sale, they're getting access to a high-intent audience, like an audience which intends to spend money on that shop. But in this kind of a setup, there is a chance that
00:06:26
Speaker
Look, you know, Indians like free. So that is exactly the other reason why we did not have it for free, right? And if you see, we are globally the first platform who has taken a very full-based approach to product sampling. It's always a push-based strategy for this thing. So here the consumer chooses what they want to.
00:06:51
Speaker
Right? So if you see at the entry point itself, the brands get a very leaned on consumer, who's already probably has gone past the awareness and consideration stage in the purchase funnel. Because we are showing thousands of products which are available for trial to a consumer, right? But the consumer at any point in time can only choose six products.
00:07:15
Speaker
So, if I'm choosing six products out of so many ones, first of all, I'm filtering out and I'm going for something which I'm finding interesting, or I'm in need of. So, that is the way, that is the way, kind of a natural selection that happens.
00:07:36
Speaker
for brands in terms of their product sampling. So the second thing is for brands, we also, I mean, thanks to digital, you have a lot of data about the users, their preferences, their understanding and stuff and all that. So we actually kind of gave a very micro-targeting option.
00:07:54
Speaker
to the brands where they can speak to a very relevant audience. I mean, there is a whole consumer science, there is a machine learning, there's AI that kind of kicks in, which kind of makes the targeting for the finer and finer for the brands so that they're talking to their most probable consumers.
Evolution of Smitten's Business Model
00:08:10
Speaker
Okay. Talk to me about the journey of figuring out product market fate and figuring out how to add value to both sides, supply and demand. What did you start with initially and how did those assumptions fair? What kind of experiments worked out? What didn't work out? We actually launched the platform in two phases. We launched a beta version somewhere around end of 2015.
00:08:40
Speaker
The consumer version was launched in February 2016. So we did about two months of code testing, just to kind of understand again, I mean, very close bunch of users, like 4,000, 5,000 users.
00:08:54
Speaker
That time we were giving like three sample products. So you just sign up and once you saw there was a delay because we would actually go and approve a user and disapprove a user. And there was like a small algorithm and that time with whatever public information that was available about the user. Based on that, we will create a persona and we'll accept a rejected membership request.
00:09:18
Speaker
And once they're accepted, it's free order. So you could just come and select any three products in your field. And three a month or just three? Three at any time. So you could select three and you could kind of check out. And then after like a one week, you have to kind of manually fill up the feedback for those three products. And then you would be allowed to try three more.
00:09:42
Speaker
Okay. Got it. Got it. Okay. Okay. So the thing was that you have to manually give feedback on your trial experience. We realized that if you do this, the platform would be probably a lot more utilized for NPDs and product research and all that, where brands need a very detailed feedback, very good for consumer engagement.
00:10:06
Speaker
But it will never be used in terms of a larger discovery and kind of a go-to-market channel for values. And for that, then we had to kind of make it a lot simpler and a lot more attractive for the brands itself. Then we said, okay, so let's just take off all the costs from the brand overhead. Let the brand not pay anything for every valid sampling.
00:10:35
Speaker
The brand would still give you the samples free, right? The cost of product is still being borne by the brand, but they're not bearing any other cost in terms of a promotional cost to access the discovery, feedback, etc. The cost of product in the overall scheme of things, the cost of product is less than 10%. If you look at your overall sampling cost, which is the 90% is your distribution cost.
00:11:03
Speaker
and the reach cost. So that we kind of took away completely from the bans. And what changed on the consumer side then? Like you made it free on the brand side. What about the consumer side? The consumer started charging. So in fact, we started charging 99 rupees to consumers for the three.
00:11:22
Speaker
So initially, we kind of give three and we used to charge 99 rupees. Did they also have to fill a feedback for or yeah, they still had to fill a feedback. Okay, so to get before they can access the next three, they have to fill out a feedback for that. Yeah.
00:11:38
Speaker
And it's up to them. They can do as many sets of three as they want. They can do four in a month or one in a month. But they are paying for it so automatically that there's no misuse. I think demand has never been a problem. Why is demand not a problem? I would imagine
00:11:56
Speaker
It would be as much of a problem as any need to see brand because you're still selling products to consumers. We're not selling products, right? So if you see today, as I was telling you, the consumer is living in FOMO all the time. They want to try, but the biggest barrier for them is your entry point for a new product trial or all is hype.
00:12:20
Speaker
Like if I have to try a deep sea brand shampoo, let's say a coffee shampoo, you are asking me to pay you 700 bucks. What if, what if I don't like the coffee shampoo? What if suddenly I start hair fall? So, so I want to try it, but there is a barrier. So we took over that barrier.
00:12:40
Speaker
was there a significant cost saving like you're saying 700 rupees for a shampoo and here for 99 rupees they get three products would it include that full size 700 rupees or let me correct you the consumer is not even paying 99 rupee because you immediately we used to immediately credit that 99 rupees back as a cash back oh right so and you automatically get that discount of 99 rupees when you are buying something
00:13:10
Speaker
And like buying, where? Like buying on Amazon? So they have a smitten wallet and they do that sampling and they try stuff for 99 rupees and that 99 rupees also then goes into their smitten wallet, which then on the smitten like list of products, they can then spend it on actual. Yeah. Okay.
00:13:37
Speaker
But the sampling packs were smaller, like traditionally you have like sampling packs would be 20 to 25, good enough for three to four years. So yeah, essentially it's free then. Okay. That free is still being maintained as a value prop. However, misuse is being prevented because they need to like shell money out of their bank account into the smitten wallet. Yes.
00:14:01
Speaker
So you're solving for intent then? Yeah. Got it. Amazing. Interesting. What kind of traction did you see? You had 4,000-5,000 people in that beta version. What kind of numbers did you see growing year-on-year? Give me some idea of the growth from the consumer side.
00:14:25
Speaker
See, again, I mean, I would say that 1.5 years is a bit restrictive for us from a supply side of things. And we probably will go about having half a million users in the first 20 to 18 months. And so these are users who have made at least one paid transaction.
00:14:44
Speaker
Yeah, for us, trial is the transaction, right? So there are two types of trial. Once we had about 100, 200 brands on the platform who had their portfolio on Saturday, things became a lot easier for us on the acquisition side of it. There was a lot of skepticism in the market, especially with the brand owners on this model. And I think they had some fast, bad experiences.
00:15:13
Speaker
with few other models where they will promise something was in teleboard. So like other startups who tried something. So it's like, for example, I mean, I'd say a lot of subscription boxes, which came, came into existence and which would go and promise a lot of things.
00:15:30
Speaker
that as a post-purchase journey or as a rural awareness metrics and all that. And they were like, not valuable. We call ourselves at the end of the day, more of a consumer science company. We are an out and out, pre-purchase consumer science company. If you have to kind of give us a definition or a domain, I would say we would probably fall there.
00:15:51
Speaker
I'll come to breaking this down. This is an interesting term. You use pre-purchase consumer science. But like 2017, you were at, say, half a million. So how is that grown? What is it today? We're approaching about 10 million users on the platform. Now, let's come to the consumer science aspect of this.
00:16:13
Speaker
So you are no longer asking people to fill out a feedback form or that is still mandatory. So that is not mandatory. That is optional now. So it's not mandatory. But yeah, I mean, we do get millions of feedback every single month.
00:16:29
Speaker
based on what we want to test out, what brands want. So it would be just a simple NPS kind of a thing, saying, okay, rate me and tell me, would you recommend this product? Or it would be like, let's say product feature study saying that, hey, how did you find the fragrance of this product? Rate on a scale of 1 to 10. Or it could be a detailed question where you want to test out certain hypothesis that will be shared and stuff and all that. So it's completely customizable.
00:17:00
Speaker
But that's your call or the brand's call? The brand may want to ask 40 questions. We would say that, okay, what is the core hypothesis you are trying to test out, right? You don't even need 40 questions. You may just need two questions.
00:17:14
Speaker
Maybe there's some gamification there or something like that to encourage people to give feedback. So we have a loyalty program in which we can keep rewarding people who engage constantly, give their feedback. We reward people more when they actually give a lot more constructive and subjective feedback.
Consumer Feedback and Brand Insights
00:17:33
Speaker
So all those things happen.
00:17:35
Speaker
Is this feedback social? Can other people on the platform see my feedback? Can they follow me for my feedback? Can they trust my feedback? No, right now, we again published our feedback, but we kind of hide your identity.
00:17:50
Speaker
So because again, I mean, we put user privacy at the forefront of it. So that is an option where the user would select that, okay, they want to publish their feedback in public. And in that case, we could kind of put a picture and kind of name and stuff and all that. Otherwise we even mask the surname and everything, and we just put the name.
00:18:12
Speaker
But I'm again talking as a user. I love leaving feedback on Google Maps about places I go to. And Google Maps has gamified it very well with giving you badges and stuff like that. And they keep prompting you that 2,000 people found your photos useful or 50,000 people have seen all the photos you uploaded and 10 people are following you or stuff. So why don't you build that in?
00:18:38
Speaker
It's a community program that we have. So we have reviewers and we have the thing who have signed up to that. So if you sign up to that community program, we will make it public. We'll give you a batch and all those things are there. But as a general user and listen until you have signed up to our community program, we will protect your parents.
00:19:00
Speaker
Okay. So now you're saying that you are essentially a consumer science company. Is there another way to monetize the consumer science insights that you are getting here? Because you're not directly charging the brands anything, right? All the samplings. Today we kind of monetize on a lot of services we offer to the brands.
00:19:22
Speaker
So, I mean, it could be a new market entry, it could be a product discovery. And of course, I mean, as a standard marketplace, we make margin when sales and everything kind of happens. So, go back to the, remember, 2000, let's say, 2008, 9, 10 days. Any modern trade outlet you would walk in, you pick up a shampoo, you will have a small conditioner attached to it.
00:19:45
Speaker
Yeah, yeah, yeah, yeah, yeah. And I mean, I remember in HUL, because I was in modern trade, like it was a mandate that not even a single shampoo has to go to shank without a conditioner being kitted. Why? Yeah, and the industry did that consistently for almost like half a decade.
00:20:06
Speaker
And the reason that industry did that at such a scale and such consistently is because your consumer behavior was not to buy a conditioner along with the shampoo. Today, you don't even think twice. You're buying a shampoo, you're buying a conditioner, right? That time, the conditioner wasn't there in the consumer basket. Now, after half a decade of this kind of inducing trial, it became a behavior. Now, you're kind of buying the conditioner.
00:20:34
Speaker
Imagine a product like, let's say, hair mask, which is a preconditioning, right? So you apply it before you shampoo, and then you apply it kind of conditioner. And hand mask is a category like every new edge brand, if you today see, you can go and search for hair mask. Every single brand will have their hair mask. But the hair mask as an adoption will take again a few years before the consumer actually kind of makes it a part of their regular basket.
00:21:03
Speaker
Like you would know at a micro level, like these 10,000 customers are going to be a good segment to pitch this product, right?
00:21:16
Speaker
Would that data help a brand to show them ads across other platforms also, like those 10,000 customers would see relevant ads on Facebook also, or is just that on smitten, they would see relevant product? Yeah, we are always the brand first approach. And anything that helps the brands in reducing their consumer engagement faster consumer acquisition cost, we will facilitate that further.
00:21:43
Speaker
Right. So if it is about kind of using these data signals to optimize their marketing campaigns across other digital platforms, if so, it will be available to them. So in what form? At what level? Like would it be that at the micro level that I say that it's a digital this thing, right? Like for example, today what you are typing in WhatsApp.
00:22:10
Speaker
Right? So it's all intrepid. So somewhere, the system is making a lot of sense out of it because the system state gives you recommendations, suggestions, and all those things. But it is not kind of actual data read by anyone. So similarly, we will pass on the digital footprint of you, where we will be able to retarget you as a digital persona, but they would not know exactly that you are accented or whatever, whoever it is.
00:22:41
Speaker
It's called data hashing. You can ask your data and you can feed it onto any of these platforms. If you like to hear stories of founders, then we have tons of great stories from entrepreneurs who have built billion dollar businesses. Just search for the founder thesis podcast on any audio streaming app like Spotify, Ghana, Apple podcasts, and subscribe to the show.
00:23:11
Speaker
So I want to understand, is Mitten a B2C company or a B2B company? What do you see it as primarily? I mean, are you into the business of e-commerce and therefore competing with Amazon Flipkart, Miso, etc? Or are you into the business of being like a Shopify which helps businesses to grow?
00:23:31
Speaker
Okay. First of all, I don't understand this obsession to bucketize a business, right? Whether a B2C or a B2B. Let me give you an example of YouTube, right? What would you call YouTube? Is it a B2C company or is it a B2B company? So we are probably taking one of the very few players who are taking a stab at the full funnel.
00:23:56
Speaker
So we are building a full stack solution across the consumer funnel, starting from the discovery phase, the standard awareness, consideration, intent, purchase, advocacy, that kind of the traditional marketing model. And there is a supply side in it, there is a demand side in it. We have solutions for both sides.
00:24:17
Speaker
So what is your earnings split between these two lines of businesses?
Targeting Top Shoppers vs. Amazon and Flipkart
00:24:21
Speaker
I would say split would be because consumer again it's the much wider segment to tap into, right? It seems a bit audacious to go up against Flipkart and Amazon considering how much capital they have raised and
00:24:37
Speaker
how much investment they have done in enabling next-day delivery of making everything work. What's your strategy here? Are you going to be a head-on competitor building everything in terms of that whole capability of next-day delivery?
00:25:00
Speaker
the whole full-stack e-commerce ability of Amazon, which might need a lot of burn? What's going to be your strategy? First of all, I don't think we are competing with Amazon or Flipkart, right? Rather, we are hoping that this trend builds up dollars and creates the market for us.
00:25:20
Speaker
Because at the end of the day, if in India, if you see if the online consumption or penetration has to increase, I mean, we barely in a retail market of $900 billion, e-commerce is barely $40-50 billion. We haven't even penetrated 5% of it. And this 5%, if you compare any developed economy and all, this 5% has to go to 20-25%.
00:25:45
Speaker
So there is a headroom of at least a five X growth from here in terms of where the overall online commerce or transaction market could go into. And so that itself is you're saying today, if it is 40 billion, it will become 220 billion, right? So there's extra $200 billion of consumption that will come into the online space. And that $200 billion of consumption is not going to happen without billions of dollars of investment again.
00:26:11
Speaker
And we are hoping Amazon Flipkart to make that investment so that once the consumers are habituated and their basket is gradually moved into more of an online stuff and all that, that is the time when we go and we help them to shop online and make their decision online much better.
00:26:32
Speaker
But why do you say you're not directly competing with them for a certain set of categories you would be, right? Like, say, food, beauty, you know, these kind of categories, maybe not electronics, but everything else which is sampleable. You would be directly competing with Amazon Flipper, even like, say, a big bazaar.
00:26:52
Speaker
See, again, as the market increases, right, there are going to be a lot of vertical play. And those vertical plays stand alone themselves are going to be decently sized. So today, even like, okay, let's take an example of
00:27:09
Speaker
i want to go back home or an i can i talk about it like if you see from a volume perspective from any perspective amazon is probably i don't know i mean i am just estimating they would be talking more than a ten billion dollar ten billion dollar in it. Which is almost like let's say seventy five eighty thousand dollars in that platform doing thousand dollars is that a competition for amazon.
00:27:36
Speaker
Yes, sir. Yes, sir. No, right. I mean, from 80,000 crore to 1,000 crore, yes, in that particular vertical, that 1,000 crore could be a lion's share. So, yeah, I mean, these kind of models are going to evolve in India. So, you said your margins are pretty healthy. I think it's because you don't need to do discounting because
00:27:56
Speaker
a customer is when somebody has done sampling, then they have that money in their wallet, which they would want to spend. So they are like, you know, that stickiness is pretty high with customers. You don't need to attract them by doing discount. Yeah. Yeah. And again, as I said, I mean, the categories we deal with, a consumer is not making a decision just purely based on the discount. You would not buy a night cream paying 2000 rupees.
00:28:22
Speaker
just because there is an additional 20% discount. I mean, you would want to really know that whether this thing really works for me or not. And whether it is authentic or not, there are a lot of test issues, again, a lot of count repeating happening in this thing. But you have to be very clear that whether you want 500 million users as your base or you are happy with 50 million users as your base. And there is a business model for everyone. Like Amazon today will be optimizing for 500 million.
00:28:50
Speaker
There are businesses who are optimizing, like if you look at Fred and Fred is optimizing for 15 million, pretty million credit card users in the pantry. It's a four billion dollar valuation company. I mean, just by the sheer volume of the TG they have is just 20 million people. So what are you optimizing for?
00:29:13
Speaker
100 million top shoppers in the country. So as I said, I mean, who have the disposable income and who was behavior, shopping behavior is dramatically changing and moving from offline to online. And the basket is exponentially expanding. And currently, what is the demographics like age or tier one, tier two, tier three, or stuff like that? So we primarily focus on people within the age group of 21 to 35.
00:29:42
Speaker
So that's our core audience segment. And the reason also being that these are the people who are discovering, who are exploring, who still haven't figured out what really works for them or what they want. And then from a demographics perspective, again, I mean, 70% of them would be female or 30, 40% would be male. 50, 60% would be metro tier one, and then another 30, 40% would be tier two. So beyond.
00:30:12
Speaker
And as we grow, of course, the mix is going to change. And I think the tier 2 is going to play a very significant role, not only for us, I think, for a lot of platforms, because that's where the next wave of growth is coming in.
00:30:28
Speaker
We have a lot of organic traction because if you go today, go and see almost like every single day, there are about hundreds of people making video about us and posting on their Instagram, social handles and stuff and all that. So there is a lot of organic traction. And then of course, like everyone else, we do the beta advertising and all that.
00:30:48
Speaker
On the paid side, what do you spend on? Is it just the ads or are you also doing stuff like paying influencers and the other formats besides just regular campaigns?
00:31:05
Speaker
See, again, it's a full stack approach to starting from digital to influencers to kind of a bit of an auto strategy, O2O, which is offline to online. So we also do a lot of activations, like small kiosk activations, experience zones across smaller towns, share to towns, where we believe that probably the
00:31:30
Speaker
Digital penetration is gonna take a lot longer in terms of building the trust and all that. So it's a mix. I mean, I would say we have a lot of co-marketing partners we work with on multiple things. So we have partnered with banks, we have partnered with Ponsiad services, so that we do a lot of cool campaigns for their customer base. Yeah.
00:31:53
Speaker
Got it. Okay. Like for a bank when they give credit card rewards and like you are entitled to these rewards when you get a card. So one of them would be like maybe some hundred rupees on smitten or 500 rupees. It's not that we are, we are again very focused on the consumer experience, right? We would not kind of do just a opening and all that. So there are partners where when they onboard a new consumer, we actually co-brand our box and we send out a welcome kit.
00:32:20
Speaker
Um, so with a, with a, with a kind of a smitten signup automatically done for them. Um, so there are, there are again API integrations and integrations. So we are almost kind of integrating with every new bag in the country.
00:32:36
Speaker
This is pretty cool. The automatic gift hamper to users of certain like these alliance partners. And when these users of alliance partners get the hamper, do they also get that balance in their account or they just get a hamper with like co-branded by Smitten?
00:32:56
Speaker
No, so they get a hamper with the complete sign-on process integrated, right? So they can just scan the code and just come onto Smitten. And then for the first experience, we would have, I mean, defined a very different path for them, saying that, hey, we welcome you. So there's a lot of product play that happens where we kind of differentiate. Sorry. They get a physical hamper with some products. Yeah, yeah.
00:33:26
Speaker
Okay. Yeah. And how would you decide who gets what? It depends on the partner, right? We understand the demographics, what kind of a consumer they are acquiring, and then we'll keep changing the option. And there are many times brands also kind of want to reach out to a certain TGA.
00:33:45
Speaker
So we will go and kind of talk to brands saying, Hey, I mean, this is the partner. These are the kinds of users that are onboarding. So we like to talk to them. So we sometimes also co-design with the brand partners, just to give them an access to all certain audience based on who the commodity partner is. Okay. Okay. Okay. Okay.
00:34:06
Speaker
And sorry, I interrupted you, you're telling me the flow for such customers who are like a line of customers. So again, when they come in, so they automatically are onboarded and they have a different journey where we kind of do onboarding and welcome messaging to them to giving them specific offers. Like for example, if you are an ex-bank customer and you are coming in through onboarding and all that, you will have a complete differentiated offering built in for you. Okay.
00:34:33
Speaker
And the offline activation you do, is that again, consumers sign up and they get a physical hamper on the spot? It's like an app replica, right? We'll do an experience center. You come in, you can sign in automatically. There is a tab automatically. If you enter your phone number, we just sign you up.
00:34:55
Speaker
And then you can enter this thing, you can pick up your products and by the time you're out and you can just show the products and then you scan the QR and then we credit the cash back into the
Customized Brand Offerings and Data Utilization
00:35:05
Speaker
app. So let's talk about the B2B side of the business now, like help me understand what are the various offerings in that that you offer to brand. Are these like productized offerings or are these like custom built service based kind of offerings?
00:35:20
Speaker
No, so again, I mean, it's prioritized, but yet it is customized because what we realized is every brand's need or problem statement is very different. I mean, there are many times brands look for very specific solutions. Like for example, like if a brand is looking at, let's say, a price discovery or a feature discovery.
00:35:39
Speaker
So those are the types when we customize the solution, we layer it and sometimes we also facilitate a lot of direct consumer interaction where our team would actually kind of do interviews, direct consumer interviews and recorded lines with the consumer consent and then those are kind of decide forward.
00:35:59
Speaker
So we many times also used NLP's and all that just to understand feedback at scale, which is provided to grants in a kind of a much better or understandable way. So yeah, so quite a few solutions. Okay. So like helping a brand understand awareness of their brand, how does this happen?
00:36:25
Speaker
See, it's a lot of brands do brand track, right? I mean, this is a very standard practice. So I don't know. I mean, it's always like every month, they will do a brand track understanding whether their affinity has gone up or gone down. I guess competition. So what our system does is basically you will define your TG saying that, hey, my code TG is, let's say,
00:36:50
Speaker
18 to 22 with certain characteristics like the typical psychographic view. And now once you have defined that, the system automatically kind of identifies the pool of the customers where they want to show this as a survey response.
00:37:07
Speaker
Typically, when we show it, the consumer is incentivized a little bit. We also have in one loyalty program called Smitten Bucks. Every time a consumer interacts or engages on the app, we actually give them certain Smitten Bucks. You could use the Smitten Bucks. There's a massive reward catalog that our partner vouchers and all that, which you can redeem this for.
00:37:33
Speaker
And the consumers kind of fill it up, which we kind of straight it back to the brand. And the whole audience recruitment, segmentation, and everything is done by the machine through the learning of our consumers. So there is not much of a human intervention that actually happens. What about preparing the questionnaires?
00:37:59
Speaker
that one brand makes the question. So we have a team, so with a couple of researchers and all that who have worked with Nielsen and Barbies of the world. And they will help the brand kind of prepare a decent and a questionnaire. And many times brand would come back for a simple problem. They would come back with a questionnaire of 50 questions.
00:38:22
Speaker
And then we would go back and we say that, okay, if you ask the consumer 50 questions, you're never going to get the true response, right? I mean, after 10 questions, the consumer just does A, B, C, A, B, C, because that is an incentive ultimately. I mean, it's like you've already invested time and you're like, now step up on that. I want that reward. And the next 30 questions, you just do it for the heck of doing it.
00:38:46
Speaker
So, essentially, what traditional offline market research agencies like IMRB Nielsen are doing, you are able to do that much cheaper with people who have the smitten app on their phone because you're able to directly incentivize them with smitten bucks and
00:39:04
Speaker
Therefore, the need for having a very large workforce who does calling and administering surveys and stuff like that is completely eliminated. Yeah. See, one is, of course, because the nature of the users on the platform is that they try and they give it feedback, right? So we asking them feedback is nothing off track. That is anywhere the consumer kind of does.
00:39:30
Speaker
And second is, it's a lot of technology and machine-driven interventions rather than manual ones, right? Even for reading feedbacks and all that at that scale. So, I mean, we would kind of use a bit of an NLP rather than somebody going and reading through this
Rapid Feedback Service for Brands
00:39:49
Speaker
thing. And so we just launched a very interesting service where, I mean, it's on my LinkedIn also, it's 72-hour research.
00:39:58
Speaker
So we actually are able to deliver responses from 10,000 plus respondents to you with their trying the product and giving you the feedback within 72 hours. And for our traditional market research player to do this, like even for a thousand respondents, they would take minimum three months.
00:40:22
Speaker
So how do you do it in 72 hours? It must be taking half that time just to get the product to the customer, right? It's an express delivery, right? So the next day, so it's an overnight delivery. So we'll get the product the next day. And then the consumer has 48 hours to kind of try and kind of respond. And yeah,
00:40:48
Speaker
And we will keep reminding the user because it will be live in their seats for 48 hours more. So we'll send a reminder saying, hey, it's live for another 24 hours or 48 hours. You just stand up, fill it up.
00:41:01
Speaker
We have done concept testing for a lot of brands, right? Like when they're expanding their product portfolio or when they're launching a kind of a new brand, like maybe something in the adjacent category. Like for example, let's say a brand who's into skincare is kind of looking into getting into active skincare and they want to understand a few things.
00:41:23
Speaker
So concept testing, category expansion. So these are, I mean, see, the typical four, five kinds of research has happened. I mean, it's, again, a very evolved consumer science. So all those kinds of standard researchers are available to you with, again, very relevant and bias-free audience recruitment process.
00:41:51
Speaker
Is this completely off-the-shelf self-service? Say I don't have a business yet, but I'm thinking, okay, I think this product would sell. Can someone like me also come on to Smitten and run a survey just to test my idea?
00:42:08
Speaker
No, you have to be a partner with us, of course, because you'll have to have access to the portal and everything. So this is like a subscription product on the business side, which is not just brands, but even agencies could subscribe to this.
00:42:23
Speaker
What other services do you provide to brands, like if we can talk through the details of them? There is a lot of re-engagement that we offer to the brands. For example, we track their NPS from their feedback closely. Let's say if they want to do a deep dive,
00:42:42
Speaker
into a low-end base score they're getting from a certain cohort or a certain kind of an audience. They could actually kind of do that because we host consumers who also deal with multiple categories and do purchases through trials for everything. So we are able to at least give brands a very relevant insight in terms of how that consumer cohort is behaving. To what extent is this
00:43:07
Speaker
pure product, no human intervention and to what extent is it customized? This sounds like hard to have pure product which is completely system generated because even deciding whom to compare against and drawing out these analytics looks like something you should need human intelligence to do and then give a brand a meaningful report.
00:43:32
Speaker
See, again, as I said, a lot of standard things are machine driven. And if you are looking for anything specific, of course, there is a human intervention. And a lot of cases, as I said, every brand's problem is different. Every brand's ask is different.
00:43:49
Speaker
So we will layer, of course, the human intelligence on top of it. So a brand can probably pick and choose cohort options and comparative brand options and then generate a report from it on their own. Like I guess it could be something like that. Like you can choose parameters to make it. That would be our first restrictions, right? Because we cannot give you a data against your direct competition.
00:44:12
Speaker
That will not be right. So everything is essentially aggregated at a category level or at a subcategory level. For example, you could see that, okay, let's say Telangana as a state, my affinity is kind of going down over the last few months. That is a downward trajectory compared in the category I'm operating.
00:44:37
Speaker
But it will not tell you that, OK, who is taking the share, who is kind of which brand is essentially leading.
00:44:44
Speaker
Okay. Okay. So you also told me that you help companies to get more bang off from their buck in terms of their marketing spend on other platforms. Like if they're spending on a Facebook campaign, then you are able to help them do better targeting and get like drive more efficiency for those kinds of campaigns. So can you talk to me on the mechanics of that? Like how you do that?
00:45:10
Speaker
Like as I was telling you the other day, let's say a brand is launching a body oil. The body oil penetration as a category in the whole beauty space is less than 2%. So now if you advertise on any other platform, so basically you have a built-in 98% bill fridge in your media plan.
00:45:31
Speaker
But we can give you access to that 2% because we know that this X number of people have actually tried. They may not be consumers of what you are today because it is probably few months, few years away. But the whole inherent nature because the trial is risk-free, a consumer starts trying things not before even these things of buying the product.
00:45:55
Speaker
And that is the phase where we know and we are able to give brands access to a very relevant cohort where they are actually targeting them on the intent, not on the top of the funnel of saying, okay, I will blast with impressions for having my top of mind awareness. And then I will blast another 200 impressions where my consideration uplift will happen. And then I will finally come to intent.
00:46:20
Speaker
So that's where we kind of bring in value for the brands in terms of doing this micro-targeting, micro-segmentations. And also because, again, a lot of users try the products of the brand, in terms of a re-engagement perspective, it is a lot more efficient for them.
00:46:42
Speaker
to reach out that segment and we can help the brands reach out to that segment not only on our platform, but across other platforms. So we will essentially give them this data access to reach out to the users on let's say Facebook or Google or an open web. So at least they know that they're talking to an audience site which has an identity.
00:47:08
Speaker
It's again, nothing new. I mean, a lot of these programmatic platforms and all that, if you see, they take the data from multiple data sources and they kind of ultimately give you the API to
00:47:21
Speaker
push the data into the whole lab networks, including Google, Facebook, and other networks. Give me an example of a programmatic ad platform, like a name I may have heard. Think of some other thing. Yeah, any DSP, any demand-side platform, that's what they do. I mean, it could be any more we think of a platform like, let's say, MiQ.
00:47:42
Speaker
So MIQ. So it's a MIQ. Yeah. So it's again, I think, I think they're headquartered in London.
00:47:53
Speaker
So if you see their whole data is basically they have tie-ups with a lot of telcos and a lot of retailers and stuff and all that and they collect all that data and then they kind of try to kind of layer the data before one advertiser pushes their ad into a Google or Facebook. So basically they will make your audience more relevant on a Google or Facebook compared to what targeting a Google or Facebook is giving.
00:48:20
Speaker
Facebook doesn't have shopping intent data. So a platform like MIQ is able to help you spend your money better by giving you some hints about intent based on data which they have collected across multiple other sources about that same profile. For example, they can say that if you want to target top spenders, then Facebook wouldn't know who's a top spender, but MIQ might know who's a top spender. Yes.
00:48:48
Speaker
See, every DSP, essentially that's the strength of a DSP, right? So they will try to kind of collect. What's the full form of DSP again? Demand side platform. How would they be earning? Like some percentage of your ad spend goes to them. Yeah.
00:49:07
Speaker
So, and that like is like say if you're spending $100, so you will pay the demand side platform $100 plus sex percentage or Facebook will get $100 minus sex percentage. How does it work? So, they will have a platform fee ready.
00:49:26
Speaker
So either the platform fee can be based on a number of impressions you serve, or it could be a number of transactions you drive. So there are different models these guys work under.
00:49:40
Speaker
Okay. So like this is something which would be in your roadmap as well to create a DSP because again, as you said, the USB or DSPs is that they have more intelligent information about customer intent. And that is something which again, as per 10, you already have that.
00:49:59
Speaker
So we would not create a DSP because DSP is essentially an ad business, right? So which kind of taps into larger ad networks and all that. Essentially for a D2C brand, they don't need to use a DSP then. Like if they are working with Smitten, then Smitten is their DSP. Or you're saying that Smitten will be one of the DSPs that they will work with.
00:50:23
Speaker
See the DSP will give you gender consumer data. We can give you your consumer data. So basically because those users essentially have tried your product, have given feedback about your product. So it's not against some random triangulation. So essentially you would like to be in that place where
00:50:49
Speaker
for certain, let's say, brands which are targeting more premium than mass market consumers, you would want them to ideally look at you as their only DSP. That would be the long-term objective. They don't need to go anywhere else.
00:51:08
Speaker
Yeah, I mean, yes and no, as I said, I mean, I don't, we don't say that, okay, you won't value your ad expenses to us and all that. But we would say that, okay, I mean, if you are driving a specific objective and probably, so again, I mean, there are two objectives for the brand side. So one, they want to reach out to a much wider audience. Or kind of, they would make a lot of spend with them on kind of build a graph.
00:51:37
Speaker
or they want to do a lot of things because they want to kind of re-engage or keep the user loyal. So we're saying, well, we'll have use case for you in certain areas and you decide what you kind of want to work. It seems to me that your B2B site is really powerful and
00:51:59
Speaker
The B2C is helping that flywheel to deliver much better margins on the B2B side, I would say. Because all of these are industries in themselves. DSP is an industry in itself that you're disrupting. Market research is an industry in itself that you're disrupting. I mean, you're actually disrupting multiple industries on the B2B side. That sounds like the real
00:52:26
Speaker
reason why a VC would be excited to invest in Smith, for example, rather than as a competitor to Flipkart Amazon and operating in that retail e-commerce space. I mean, that seems like, yes, you need to do that because it gives you data, but that's not really where the value creation is happening. That's what it seems to me.
00:52:49
Speaker
We are a full-stack solution. Of course, there are a lot of interesting solutions for the brands. But again, I mean, from a consumer perspective also, we are here to solve a real problem for the consumers, which is the problem for adoption and discovery.
00:53:05
Speaker
If you like the Found A Thesis podcast, then do check out our other shows on subjects like marketing, technology, career advice, books and drama. Visit the podium.in for a complete list of all our shows.
00:53:26
Speaker
Before we end the episode, I want to share a bit about my journey as a podcaster. I started podcasting in 2020 and in the last two years, I've had the opportunity to interview more than 250 founders who are shaping India's future across sectors.
00:53:42
Speaker
If you also want to speak to the best minds in your field and build an enviable network, then you must consider becoming a podcaster. And the first step to becoming a podcaster starts with Zencaster, which takes care of all the nuts and bolts of podcasting, from remote recording to editing to distribution and finally monetization.
00:54:03
Speaker
If you are planning to check out the platform, then please show your support for the founder thesis podcast by using this link zen.ai founder thesis. That's zen.ai founder thesis.