Latest / Talk Commerce / Revolutionizing Digital Advertising Through Shirofune Automation with Mitsunaga Kikuchi and Rocco Baldasarre at eTail Palm Springs
Transcript
- Speaker 3 (00:00.27) Hey now. Welcome to this special e-tail episode of Talk Commerce. Today I have Rako Baldassare and Mitsu Kikuchi from Chirofune. Mitsu and Rako, Mitsu why don't you go ahead and do an introduction first for yourself. Tell us your day to day role and one of your passions in life. Okay, so I'm Mitsuki Kikuchi, I'm founder and CEO. I'm also an engineer. So I created the algorithm for Shofune to optimize campaigns to get better results. And why I found Shofune is, know, human experts always do the things that AI cannot do. So I realized that. you know, system like that acts like an extra human, that would be best for saving time and getting better performance. So I started coding to replicate what the extra human do. And now we can do the tasks that AI cannot automate. So that's our uniqueness and that's my place where I can put my, you know. passion into it. Speaker 1 (01:23.566) That's great. And rocko. Nice to meet you first of all, thank you so much for having me here. I'm Rocco Bollosari, I'm the head of business development and passion, like I many passions. would say soccer is the one that I spend a lot of time. Football. Yeah, I hear they think about American football if I say football, but yeah, the real football is the one we play with the feet. Yeah, yeah. But yeah, no, thank you so much for having us here today. I was in 2005 when they still had the dream team. we won't talk about soccer today or running. tell us a little bit about, and I love the fact that you're in tech, that you were in tech, and that's my background too. I was in tech and did coding and then I founded the company. So anyways, tell us a little bit about the beginning and the solution that you saw. You kind of gave us a preview of it, but tell us about... Dream me. Speaker 1 (02:16.29) how that solution now has evolved into Shirofunen. So I began my career at a digital advertising agency. so the Azure Manager worked very hard to optimize campaigns. And so our inspiration is that if we can create a system that acts like a human, so I started coding to replicate that. kind of activity. Yeah, one of the things that I think is very important to underline here is that a system that works like a human. Shilofuno is 15 years old now. We are not an AI company. We are a human company. We call our technology human intelligence because we try to replicate the work of a person rather than an AI take crazy decisions for you. Can you give us an example of how that would work for an actual retailer or merchant that's using the system? Speaker 2 (03:24.622) But of course, so essentially let's assume you are having campaigns across five platforms and there is a lot of data being processed. What the person would do is he would download the data, read the data, interpret and make decisions. What our system does is exactly that, but the decision is going to be brought to you in a yes or no format. The data is telling us that this is a potential opportunity for you or a problem for you. we suggest to implement this change. Do you approve this change or not? So that's something that an example of the many things we do where we replicate the work you would spend countless hours doing, but we do it in a much faster fashion so you can focus on strategy, can focus on growth. What are some of the examples of, you talked about getting some ROAS and getting some advantages over what you have, you have some good examples of what that would look like in terms of increasing your spending. Yes, essentially to give you a good example, we actually calculate how much time people save using Shirofune. The lowest we have seen is 40%, the highest 70%. And that's a pretty big saving. In terms of ROAS, we see a lift within the first few months of 50-60%. Tell us about your ideal client for you who uses the system and what is the typical use case for your ideal client? Speaker 2 (04:55.182) It's somebody that has a large enough budget to not have time to optimize it essentially. Where they believe that there is a lot of missed opportunities in the account from a budgeting, from a bidding perspective, essentially not having the time or never being able to have the time to work on that. Those kind of customers, they're going to see value from day one because they understand that they are leaving money on the table. And of course it needs to be somebody that is willing to adopt technologies, needs to be an adopter. Otherwise you're going to end up with people that think it's a magic bullet where magically you're going to get double the results. No, it's a gradual process, like a person would do that, but they need to recognize the fact that they are missing those opportunities in the first place right now. Mitsui that you've done this for you started this 15 years ago you must have seen a huge amount of different transitions from different technologies and now AI is such a buzzword yeah how you have adopted across all those different technology shifts yeah That's actually a very good question. So we are growing with technology of course. So humans use technology to get better performance. So we are actually similar to humans. So we are thinking what technology we can leverage for our algorithm. So, technology grows, as technology grows, our algorithm can use many technologies, like AI or machine learning thing, we can handle human component and AI component together, so we can maximize the performance and our capability. So, we grow with technology. Speaker 1 (06:52.994) We're here at eTail and the buzzword today is agenic. Agenic commerce, agenic marketing. How have you embraced some of that new idea of this agents that are going to go out and buy and sell things for us autonomously without us even doing anything? Well, that's a very interesting topic because first of all, it's going to take probably some time for full deployment. But that doesn't mean that they don't know what I like and save me time to go buy a product, right? What I think though we need to remember is that no matter how they go about finding what to buy, they're probably going to be using some sort of algorithm to predict our behavior during those purchases, funnels. So that means that you still go through the optimizations as if you're selling to a person. You're just saving time by not letting the person shop. But the concept of having a good product that answers the five W's, the one H, is still gonna be there. So essentially the base, the core, I think is not gonna change. It's gonna change the way we interact with things. And so tell us about how your product works within the funnel. They come in and it's going to optimize the type of product they're going to show the client or the type of clients that are going to see different products. Tell us about that. The way our product works is, let's say you're running 100 campaigns. We are essentially going to shape-shift the priority of these campaigns through budget, bids and modifiers, and we're going to identify pockets of opportunities in each campaign, times of the day where you have good results but you're not spending enough, and we make those changes for you. And of course, if a product is performing better, we try to push that product more. So essentially, it's an efficiency-based game that we're playing. Speaker 2 (08:47.278) and we are making sure that the goals that you have are met the most efficient way possible. A very good example is somebody that might be selling a product for $100 and another product for $50. The typical AI would be like, I want to sell more than $100 because I have, let's say, profits on that. But what if the guy that buys the $50 product buys it seven times throughout the year? We are able to look at that data and be like, you shouldn't spend money on the $100 product because by the end of the year you get $300 back. On the $50 you're going to get $450 back. That's where your growth is. All of that is automated for us. And we are able to essentially grow the bottom line of the businesses by using their own data that they're not using right now. So this the business started in Japan did it so you've now gone into different markets Has it been a challenge coming to the US market? Is there anything that you've seen that's a harder and are you in the European market already? I don't think there is a huge difference because the people are doing the same thing like managing ads, Google ads, ads. But slightly different is in Japan the people are more detailed focusing so they want super detailed numbers or they want to know why the performance gets better or worse. In America it seems to not to care too much about the details. So we just provide the performance or time savings. So that's slightly different, but basically it's the same, I would say. Speaker 1 (10:29.25) Do you see that the managers value the time savings over just cost savings? Like sometimes managers don't equate time with cost often? We make sure they understand that because particularly with agencies, when we work with them, we give them a timetable and we ask them to write down how long it takes to perform tasks. Then with that, we go to the hiring manager or whoever is their boss, who is like, how much does an hour cost for you? And we say, this is how much it's costing you to optimize this account every month. Then we give the same time sheet at the end of our trial. And usually you see that 70 % cut in time. And we go there, by the way, you're paying us one dollar, we saving you 12. That's a deal, like they always sign a contract at that point because they are seeing the number. So here at eTail, outside of AI and outside of agenic commerce or whatever the buzzword is today, what are you seeing that's exciting that we're going to be looking at coming into the future for technology? For technology I would say attribution is like more and more companies are waking up to the fact that it's not what the platforms are reporting, it's about way more than that. Speaker 2 (11:44.366) For example, a lot of companies, they optimize by thinking, I'm getting a sale from an email. Email marketing is working well. And now they're starting to think, how did I get that email in the system? So is it really email making the sale or was it the Facebook campaign that I'm not spending much because I don't think it's performing? And that's very exciting that they're realizing that. How do you think it's gonna be a challenge with some of the cookie stuff going away that people are, that it's gonna be harder to attribute some of those, that funnel or where that original email came from? Because it's maybe, like you said, it may go through four or five steps. It may go from TikTok to Facebook to your email list. Third party data, so attribution company help with first party data, that's why it's so important and they are able to trace that for you. It may be anonymized format but I don't need to know if you bought something. I need to know overall what happens to a good percentage of my customers. That's good enough to make decisions. How about the spends outside of the traditional media like I think Google and Amazon are traditional being if it's still a thing but like Walmart is now a thing and Target and some of the other platform marketplaces are you seeing customers moving some spend into different areas to test it? I think Mitsu is probably a good topic in regards to why we are integrating with DSP, Trade Desk, 360 because they're doing that. if they're doing that, they also want to make sure they can optimize it. And everybody's trying that. And the reason why they're trying that, they are understanding that it's much better to put your eggs in multiple baskets because you never know if there is a platform that can give you more immediate gratification. Speaker 2 (13:32.99) Part of it is the inflation. Costs have gone up on the traditional media. Some companies are like, there anything out there that doesn't cost me as much to get a similar result? Yeah, so if you look at how much the cost to spend on something new, the newer marketplaces, testing those marketplaces to see how well they work is maybe one of the places that you work well in. Yeah, essentially one of the things that we in Japan, for example, we partner with Bing a lot. If we're seeing a platform performing better, we take money away from another platform and put it there, automatically. So for them it's a much safer investment because they're like, want to test it and if it does well, I need the technology to balance my budget distribution. So that's something that we thrive a lot in. Is there too many places that you can try to spend ads on? And then think of like TikTok shop and all the different social medias. X has ads spends. Do you recommend trying a lot of them and just testing it? And that would be something that you would be able to help with by testing, hey, maybe in a certain vertical. X works really well for you and maybe it's cheap now because nobody's on it but nobody's on it but maybe just your vertical is on it. Is that something that you would... Speaker 2 (14:53.036) Absolutely, and we always tell our customers if you're smart, you test at least one platform a month. And we can help them simplify that, but it's like you always want to test some things because otherwise you're going go in a stall and you're limiting the opportunity for growth because what if the X in that vertical works? Yeah, so just from a usability standpoint, do you see that some ad networks are going up? is Google coming down and others are kind of taking their place now? Is Amazon taking some of that business, or is it still pretty well distributed? It is very well distributed. We are seeing a lot of people try more TikTok shops. What is interesting though, what I think is going to be something that might change in the future. is that a lot of Amazon only sellers, they're trying to think about going directly to the website because of all the fees associated with it. So I'm seeing a lot of that, but it's slow because they don't want to lose those sales they're getting on Amazon. But in the last few years, that's something that I've seen more and more of. So we have a few minutes left. Mitsu, I have to ask you, are you still doing some coding? Do you see that as like a stress relief to do a little coding? Speaker 1 (16:36.012) That's great. Alright, so as we do close out the podcast they give you a chance to do a shameless plug about anything you want which just means you can promote anything you want shamelessly. So what would you guys like to promote today? What I would like to more than promote mention is I want to talk about why people trust us. whenever somebody wants to use our platform there is a free trial. Usually two months depending on the company can get up to three. The free trial usually with the three months you get a good understanding whether you like us or not. That's a very big risk for us. but it's not the only risk we take. The second risk we take is that we don't have contracts with us. It's a month-to-month agreement. Month-to-month agreement means that every month we need to win your business over and over again. So whenever somebody offers something that crazy, three months free and month-to-month, you might be asking yourself what their retention rate is like and what they deliver because that's big offer. And that offer has been there for 15 years. So I let the public judge if we are a good company by looking at the way we offer our service. And that's usually something that we tell our customers. Look at what they're trying to let you sign and you will understand how good the service is. And we have a 96 % retention rate today. That speaks highly of what Mitsu has built over the years. I didn't quote a single line of code, by the way. That's awesome. Both of you, Mitsui Kikuchi and Rako Balasari, thank you so much for being here. It's such a great pleasure. Speaker 2 (18:16.142) Thank you