How Amazon and Walmart can be disturbed by chatt and confusion

Now he is on a different e-commerce bet for the Ai age after launching and launching an on-demand car service business for the past decade. His new company, refibuy (an abbreviation of research, find, buy), aims to create software that helps consumer brands and retailers to navigate in a new online world, in which AI shopping officers make purchases as often as human customers.

Assets Recently hit Wingo to discuss how generative KI and AI agents interfere with online purchases and how e-commerce power plants can be influenced at this moment. It is not surprising that Wingo, as someone who tries to build a big new business in the room, is perhaps more bullish than many about how soon and very new AI shopping tools can interfere with existing e-commerce giants and old retailers. The interview was processed for length and clarity.

Assets: What did you see in e-commerce that made you come back in?

Scott Wingo: The LLM companies (such as chatt or confusion) receive the attention of consumers so that they receive consumer distribution. And if you get this distribution of consumers, do you have to find out how I have to monetize them? It became pretty obvious to me – Especially when confusion launched your first (shopping function)) -The this would be the next big wave of e-commerce, and I couldn’t miss that. I feel like my team and I have a unique view as experts from topics, and I see how e-commerce has been done so far, the good and the bad.

What is the “bad” to refer to?

If you look at e-commerce in the USA, the growth level has really slowed down, right? Depending on which numbers you look at and what you count, we have committed ourselves between 15% and 20% of the total sale of retail from e-commerce, and that was a rather steady number.

But if you look at other countries, this number feels wrong. They look at Europe and this number is much higher. And Asia is far from the charts. Why is that? And I think that’s because our user experiences are really followed or stagnated. You are a big one Amazon Pendant like me, but something happened where you completely stopped being innovative in retail and just started with cash.

How do you see that it is affected in this new world because you mentioned Amazon?

So you have the retail business, a (third-party seller) marketplace and advertisements.

The display layer in my head gives the consumer experience no added value. So imagine a world in which someone (a AI experience) builds on Amazon and nobody uses the Amazon frontend -they start with a AI chat or an agent. Amazon is now only a backend, so that people do not see these product indicators at Amazon. Basically, there are basically 60 billion US dollars of the profit margin. This is a problem No. 1.

Problem No. 2 is now only the backend for another frontend (such as chatt or confusion), and at some point you have to pay a tax (to the AI ​​company) because you can be switched off. So Amazon hates paying a tax. You are the tax collector, right? So there will be a fight for who the tax collector is. So that will be interesting.

Then I think that the third thing is when I am chatted or (Openai CEO) Sam Altman, what if I only calculated sellers 5% for listing their products (compared to 12% to 15% for many categories at Amazon)? “You can list your products at Amazon, but we would also love it if you are listed with us. And since we are cheaper, would you offer a lower price?” A brand like Dyson (could) not accept this deal, but the small business seller may accept this offer.

So there is a way that you could really attack Amazon by getting rid of the advertising business quickly, which would be pretty disadvantageous. And then I think that they could chew in a large part of the market business. And these are the two – AWS aside – profit centers of the company.

When you talk about someone who is based on Amazon, talk about AI search engines or shopping officers who are the starting point for more customers for online purchases. How will you or I search for products by KI shopping officer than shop today?

So it will be multimedia – it could be voice, typed or whatever you want. I think we will be interested in the voice because it will be really good.

You will have this conversation with what your personal assistant will be and everything will do for you. So you will have a conversation about it about your calendar, and you will tell you: “Remember that you got this dinner this evening and Bob will sit on your left, and he is the CTO of this company and remembers that you appear this story about his company.” Basically, it will be an advanced assistant for you.

And then it becomes: “By the way, can I have a short conversation with you about the things you need to fill up?” And you will say: “Sure.” It’s like: “Well, I got the toilet paper in that day, the milk that day, do you agree?” And you will say: “Sure.” Or you will just trust him so much that it simply ordered. This is how the basics will happen. It will only be a short conversation or you will simply put it on autopilot.

And who is at the back there?

Maybe it’s Amazon. Maybe that’s that Walmart. The agent or assistant decides for you. They give him parameters, possibly for value or convenience. So the value would be: “I don’t care where the milk comes from as long as it is below X Dollars a gallon and it is 2%. “

Then it gets really strange, because now the agents could do small auctions and talk to each other and say: “I have this great filling, who wants it?”

Buy something new on the convenience page and it is a considered purchase -say you want to spend $ 150 for a pickleball paddle. And it will know the right questions you should ask and suggest the attributes you have to find. And maybe they say: “I need it in two days and I have to pay with (the purchase now, paying later).”

How much can Amazon’s new AI functions like his Rufus AI shopping assistant help the fight?

They are good effort, but I think chatt is getting better. Part of it is that Rufus cannot be so good that you do not need advertising (Amazon product search). So that Rufus can be dramatically improved, it almost has to replace the existing viewfinder experience on Amazon. And that is a section that Amazon will not cross because the advertising revenue of $ 60 billion would kill that are essentially pure profit margin.

Do you get a meaning from the conversations you know from Amazon or Walmart due to the size of the threat, at least as you see it?

I have no conversations with Amazon, but Walmart people said very publicly that we will pursue a very open approach to the purchasing agents And welcome. In a way, this is clever because they are a challenger for Amazon, and I don’t think they have anything to lose. This is an opportunity for you to suspend the field and get Amazon. They are okay if they are not the front end for all of these transactions, since they also have the shops and many of their online orders are picked up in the shop. At least they will have an interaction of consumers at this level.

What are your top or two messages to retailers or brands about how your processes have to change or what you do for different than now?

So the honest answer is, we believe that this is only the beginning and it will be bad and you have to accept these operating systems, and your data is not ready for it.

There are a few data pools that we have in the retail world, and they are not really ready for agents. So there is no standard for product catalogs. I lived that at Channeladvisor – everyone would send us their product catalog and they were a hot mess.

Select a category and everyone has different attributes (properties or details that describe a piece of goods) for each product and use different terminology. Let us assume that a gold standard for a GoPro 10 had 200 attributes. Each retailer can only have 20 of them. If you are Amazon and Walmart, you may have 50 of these properties.

However, the agent engines need more content and context. The average product search can contain four terms, but AI queries or chats is 15 to 20 words. In order to bridge these things, we would like to help single dealers to look at their data, to evaluate and say them, here they are fine, and they are not prepared here, they are prepared here and they are super unprepared here. And then we have the skills to help.

How quickly will all this happen in your view?

I thought about this idea and talked to people in the fourth of 2024, but in the last 30 days it has become somehow interesting if really large retailers of several billion dollars “I get pressure from all sides”. So there is pressure at the board level. There is a lot of it. But what you really see is that the distribution of incoming traffic changes very quickly and very quickly. In May it started in e-commerce and the people are behind the scenes, on fire and freak out.