
Loading summary
A
In the rapidly evolving world of retail media, few platforms are as uniquely positioned as DG Media Network. With over 20,000 stores, Dollar General serves as a lifeline for over 90 million shoppers across the US at a time when shoppers are looking to save money. Digi Media Network Media built better. Hey, gang. It's Monday, November 21st. Rachel and listeners, welcome to behind the Numbers New Market Video PODC made possible by DG Media Network. I'm Marcus and joining me for today's conversation, we have our retail briefings analyst living in New York is Rachel Wolf.
B
Hey there. Thanks for having me.
A
Absolutely. Thank you for being here. Today's fact. Where did the term stealing someone's thunder come from? So, as the BBC notes, in 1709, a man named John Dennis invented a machine that made a noise just like a thunderclap. It worked really well. But Mr. Dennis play wasn't that great. And soon after its opening night at the Theatre Royal in Drury Lane, which is the oldest theatre in London's West End, his play was replaced by a production of Macbeth. When John Dennis went back to the theatre to see this new production, he was outraged to discover that this thunder machine was being used in Macbeth and no one had asked his permission. The story goes that he st up and shouted, they will not let my play run, but they will steal my thunder.
B
That is very interesting. I always thought it was like Greek mythology or something where there's some story of somebody stealing Zeus's thunderbolt or something. But this is much more interesting than that.
A
Nothing to do with the sky. Yeah, apparently. I don't know why he was building a thunderclap machine. That's a lot of effort.
B
Yes.
A
I mean, no wonder his play failed machine.
B
And it probably used it once.
A
Exactly, exactly. Oh, Mr. Dennis. Anyway, today's real topic, the three biggest questions surrounding Amazon at the moment. Okay, so every quarter we will be looking at how the big tech players are getting on and asking what are the three biggest questions surrounding said company, in this case, Amazon. Right now we're going to discuss my ranked list. And then Rachel would tell us what she would change or add, maybe we've missed something. By we, I mean me. Okay, so I think, Rachel, the number one story at the moment for Amazon is their corporate layoffs, which made a lot of headlines. As Alina Selyuk of NPR was explaining, Amazon is laying off 14,000 corporate workers as it spends big on AI. They did give some reasons, well, a few different reasons as to why. But I'll start with this question. Why does Amazon need to cut workers when they're doing so well, do you think?
B
I mean, that's sort of the crux of it, right? And 14,000 is just the total number of reductions that they're going to have in their workforce. I think some reports have the total number of layoffs at 30,000, but Amazon has said they're going to hire more in certain places, so. So the company, as you said, they say it's not financially driven, it's not AI driven, but they have talked a lot in the past about how AI is going to make the company more efficient and reduce the need for all of these management layers and just enable faster innovation. So I think part of it is a bureaucracy thing. Amazon has talked a lot about how they just have too much bureaucracy, too many people getting in the way of doing stuff, and I think that's certainly one reason for the layoff. But I do think the AI piece is definitely an interesting one.
A
Yeah, they said their reasons for doing this was part of its right sizing. Amazon ballooned its workforce during the pandemic, as a lot of companies did. And then the other part is investors aren't happy enough with the growth of their cloud business. AWS is growing very, very fast, but not fast enough compared to Microsoft and others. But the AI piece and losing jobs whilst you're doing so well I think is very interesting because. And I think Amazon might feel like it's a, or might be a bit of a microcosm for the jobs market. One reason might be because they're the second largest private sector employer in the world and so there's somewhat of a bellwether. The market goes or the job market goes as Amazon goes, so to speak. But at first it does look like they are spending a lot on AI and then cutting jobs over here to kind of save money. But when I looked into it a bit further, it seems like the headlines have maybe spooked folks, but the AI related job losses haven't actually hit yet. Ernie Tedeschi, former chief economist for the White House Council of Economic Advisor, says I'm very skeptical of a extracting signal from headline layoff notices. And as we were mentioning, Amazon CEO Andy Jassy saying that the layoffs weren't financially driven or about AI. However, back in the summer he did write to employees saying, essentially, we will cut some jobs to add others in the next few years. We expect this will reduce our total corporate workforce as we gain efficiencies from using AI. And Amazon has been cutting jobs back to 2022, laid off 10,000 people that year. So they've been doing it for a while. But it does feel more like the headlines are scaring people into thinking this is happening right now, when actually companies are kind of maybe planning that this will change things in the future.
B
Yeah, I think that's definitely true that AI is not the job stealer that people think it is at the present moment. But I do also think that in some ways Amazon's claims that AI had nothing to do with it is kind of disingenuous because they did say in the letter talking about why they needed to make changes, that AI is changing the way the organization operates. And I think the way that a lot of retailers are approaching this is maybe instead of talking about, okay, AI is going to replace people, then they're framing it as AI will enable the current employee workforce to do more. So essentially keeping the workforce stable, not hiring more people, but also not firing them.
A
Yeah. Yes. When people. Especially because people leave on their own accord, it's not the only way that people stop working at jobs. They leave. And sometimes when they leave, companies are asking, can we do parts of their job with AI? And it is hard to really understand what's going on because to your point, in terms of how much is AI really affecting jobs at the moment, there was a new study from economists at the University, sorry, Yale University Budget Lab and the Brookings Institute saying that the doomsayers are wrong and that there is as yet very little evidence that AI is putting people out of work. The other part of the problem here is that as Emily Peck of Axios notes, the government shut down. And so the Bureau of Labour Statistics isn't putting out job reports. And it's kind of difficult to see the full picture of what's happening. It's possible AI is driving layoffs in the past month or two, but it's. There's a list from Challenger Grand Christmas and they look at the top three leading causes for layoffs this year. They don't put AI in their top three federal government job cuts. Doge cited in nearly 300,000 layoffs through September employees. Number two was employers waiting out uncertainty over tariffs and inflation, cited in 200,000 cuts. And then store and plant closures from typical business churn was in third place. So it's easy to point the finger at AI, but there's a lot of going on.
B
And I think the other piece that you have to talk about when you talk about Amazon is their automation push in their factories or not their factories and their fulfillment centers. So There was a report in the New York Times that said they plan to replace 500,000 jobs with robots at some point in the future. I mean, I think it's probably in the next few years. It's not immediate, but that would also be a significant reduction obviously of Amazon's workforce that is also AI related. Maybe it's not direct, it's, maybe it's not Genai related. But that is also something to think about when looking at Amazon and their plans for the future.
A
Yeah, you do kind of have to pass out what is being done right now with AI and how it's changing jobs and what that could do AI could do to jobs down the road. There's this survey from retail executives by amperity suggesting just one in five employers expect to reduce headcount due to AI in the next 12 months. The share who actually planned to add employees due to AI was actually slightly higher. And then this chart here as well that folks can see on the screen if you're watching data from the World Economic Forum shows that more companies are upskilling or hiring because of AI versus downsizing. It does feel like it's easy to make the correlation between, okay, this company is taking money away from here, these people to spend on AI. But just when you look at a lot of the research and a lot of the data, it doesn't necessarily suggest that even just the overall unemployment number was a little bit below 4% a few years ago. Now it's a little bit above 4%. So that hasn't moved too much. But it does. I mean, there's a lot, there's a lot of cognitive dissonance, I think, for folks, Rachel, when you look at the company that just made $180 billion in Q3, up 13%, its fastest growth in seven quarters, and it made $21 billion in net profit in Q3, up nearly 40% as it's cutting 14,000 jobs, 4% of its workforce. It's hard for folks to wrap their head around why.
B
Yeah, absolutely.
A
And if Amazon's not doing well, what does that say for the rest of us? So that's what I think is the number one story here because it also kind of plays into the broader jobs market picture. I think that that matters for a lot of folks all the time. So the second thing I think should be on the three biggest questions for Amazon list is their new Help Me Decide feature. So our senior analyst, senior retail analyst Zach Stambor writing that Amazon is rolling out a new AI powered button called Help Me Decide that appears at the top of a product detail page when the shopper has been browsing comparable products without making a purchase. The tool analyzes browsing activity searches, purchase history and preferences to recommend the right products. Presenting shoppers with recs that include an explanation of why it's a great fit, insights from customer reviews and how it aligns with past purchases and preferences. The question here, I think Rachel, is, is Amazon's new Help me decide feature a significant stepping stone towards agentic AI? You know, kind of, I don't want you AI to decide on your own yet at least, but get me as close to the decision as you can.
B
Yeah, I think Amazon is experimenting with all of these, you know, little AI enhancements to the customer experience they have. I think Rufus right now is really the main one that people interact with where it gives you product information, it will summarize reviews and you can ask it questions. And something like 250 million people have used Rufus this year. Yeah, It's a huge 140% increase in monthly active users, 210% increase in interactions. And I think the most crucial part for Amazon is that, and this is all internal data but, but customers who use Rufus are 60% more likely to complete a purchase. And so you tack on products or features like Help me decide and one can only imagine that these tools are just going to be even more effective at nudging customers to check out whatever they put in their carts. So I think it is very much. I don't know that it's necessarily a push toward agentic commerce necessarily, but certainly one to improve the experience of customers shopping on Amazon and to keep them shopping on Amazon.
A
Yeah, I had two takeaways from this. One is it's a new kind of comparison offering. Kelly Tycho of Axios was pointing out that once you press the Help me decide button, the AI surfaces one main recommendation along with an upgrade recommendation or a budget recommendation. These two upgrade and budget alternatives. And it reminded me of kind of liking it to airlines. When you're booking a ticket, you have a certain ticket type, but they also show you how much it would be to buy a slightly more expensive ticket. Economy to business to first class and retailers. That's not always the case. A lot of the time they're comparing side by side items. I thought was interesting that how this could move shoppers up and down in terms of their price range or in terms of what they're going to buy versus left to right.
B
Yeah, I think that's a really interesting feature especially now when people are very conscious of how they're spending money and, you know, maybe some people are willing to splurge and some people want the budget option. So I think certainly giving customers more options is always going to be better than just directing them to one.
A
Yeah. The second takeaway was from Ms. Tycho as well, from her piece basically questioning how are these recommendations decided. She was warning that the feature could raise questions about how Amazon's AI decides which products can get spotlighted and whether paid placements, brand partnerships, or its own retail lines could ultimately influence those AI generated best picks.
B
Right. And there's also the question of what does this mean for Amazon's ad business when most of the money that it makes comes from search placements, you know, through a traditional search on its website.
A
Right, yep.
B
And I think that's a question that they have to answer sometime soon.
A
Absolutely. So that's my second pick. Amazon's new help me decide feature number three. I went with Amazon's AI smart glasses for workers. So Amazon's unveiled prototype AI smart glasses for its delivery drivers, writes BBC's Lilly Jamali. She explains that Amelia. The Amelia glasses, as they're called, include a camera and built in display and pairs with a button in a waistcoat. The glasses help drivers find the right packages in their trucks, give them turn by turn directions to the right address, and take photos of successful deliveries instead of needing to use a phone. So my question here, Rachel, was is this the Trojan horse for smart glasses? Is this the way we get used to people wearing them out in the wild?
B
Maybe. I mean, I would say that I think for the majority of the consumers, the Amazon smart glasses are not, you know, necessary. They don't necessarily have the features that they're looking for at this moment in time. I would say, yeah, I think it's a, it's an interesting piece of kit. I think the way that I look at them is more through the lens, no pun intended, of, of what Amazon is doing overall with fulfillment. Right. I mean, with these smart glasses, the aim is very much to make it easier for these drivers, speed them up, get them to the door faster, help them find the package faster. And I mean, throughout this year they've been rolling out all of these innovations in their fulfillment centers with these last mile delivery capabilities to just get stuff faster. So I think it's very much in that vein. I mean, going back to, you know, the automation push and fulfillment centers, I think it all fits in very nicely with that.
A
Yeah, yeah. What are the efficiency gains here? Amazon's VP of Transport Beryl Tomei, I think that's how you pronounce it, estimates the glasses could create up to 30 minutes, half an hour in efficiencies per 8 to 10 hour shift by minimizing repetitive tasks. So quite significant. Also it helps, you know, people find the packages in the trucks, get them to the places, but also if they drop it at the wrong place with precision location targeting, they can, it can basically warn them, hey, you've put this at the wrong house, which could be very beneficial as well. It seemed, Rachel, that another one of my takeaways here was safety first, privacy second. So safety first. I say that because Amazon said the smart glasses can detect when you're, when they're in a moving vehicle and it shuts them off automatically, which I thought was pretty cool, and also supports prescription lenses, which is going to be an important part of this. And then primacy second. Shaun Hollister of the Verge noting that Amazon's blog post doesn't discuss any possible ethical concerns that workers or customers might have as these glasses monitor the last mile of the delivery process. How big of a concern is that?
B
I think it's a valid concern. I mean, one thing that I was thinking of was that you do get a tremendous amount of data from these drivers going from door to door, essentially. And that can all be used obviously to help Amazon build out their geospatial strategy for fulfillment and all those things. But it is a lot of data that people may not be comfortable giving up. And I do question how much of this mapping, again, not to bring it too much back to automation, but again, the more that you regiment these things, the more that you hand it over to AI to make those decisions. I feel like the easier it is to eventually hand it over to say, a robot delivery thing. I don't want to say person machine to handle those tasks.
A
Cyborg. They are helping people find the packages, helping people get to the locations, helping people drop it off. It does seem like at some point it'll be very easy to pluck the human element out of this and replace it with whatever a cyborg robot drone replacement looks like. But to what? Something you said earlier, it's a kind of interesting piece of kit. I do think it is a bit of a stepping stone into getting people comfortable using these, because I think people wearing these out in the world, it kind of freaks people out a little bit. Is the camera always on? You know, what are they, who are they recording? Whereas for a worker, it's kind of behind a pigment, kind of hidden behind a brand, so to speak, or a company. And it's, it's a tool, you know, it's something that people are using to help do their jobs faster. So I don't, I think if, maybe if the postal service, whoever like started using these, people have less of a problem with it and then maybe people do get more comfortable with it because it's kind of almost the reverse of a ring doorbell. There's this camera that is staring out into the world at all times without your permission. And this is the glasses that are going to be kind of seeing the reverse angle of that experience on the last mile delivery portion. And I think it is a baby step towards broader adoption potentially.
B
Yeah, that's an interesting point. And I think, you know how many Amazon workers are going to be using these? Presumably a pretty sizable amount and they might see the benefits in other areas of their lives as well.
A
Exactly, yeah.
B
And you know, boost adoption that way. So I think that's definitely interesting piece. I would also just wonder how many people would know that these drivers are wearing smart glasses.
A
Great point. Yeah. Because they don't look that different really. They look like similar. Ish to what do you call it, racquetball? I think we call it squash glasses. You know, they're a little bit different.
B
Yeah.
A
But not where you would notice it. And yeah, I think maybe people used to know their, their mailman, you know, 30, 40 years ago, but I don't think many people are staring at their Amazon driver that much. So yeah, it's a good point. All right, that's my top three. So recap. I've got a question about Amazon's corporate layoffs. The second one is about Amazon's new Help me decide feature. And finally, Amazon's AI Smart glasses for workers. Are there any bigger questions or should these be ranked differently in terms of significance?
B
So I would maybe reframe question two a little bit. Still thinking about AI. But I think one of the most interesting stories about Amazon over the past few months has been how it is approaching agentic commerce. I would say more specifically with regard to third parties. So they just sued Perplexity for allowing its agent to make purchases on Amazon's website after Amazon changed its terms of service to explicitly prohibit that type of behavior. It's shut off Google and OpenAI from doing the same. And I think there is a big contrast to be made with how Amazon is approaching agentic shopping versus how say Walmart is. I mean, Walmart has a partnership with ChatGPT, literally to make agentic commerce possible on Its, you know, to allow people to purchase from its site through instant checkout. Yes. So I think it's a kind of an almost an existential. I don't want to know. I'm not going to go that far. It's not an existential crisis for Amazon, but it is a serious problem for them because, you know, going back to the question of ad revenues, as I said, most of Amazon's ad revenues come from search. And if more people are using ChatGPT or perplexity to do the product research on their behalves, then that sort of takes Amazon out of the equation a little bit more and they lose ad revenue, they lose customer data. So I think it really is a big question for the company how they're going to respond to this threat.
A
Yeah, I think that's fair. I think a reframing absolutely works.
B
Do we want to talk about Grocery?
A
Always grocery. So the reason I'm not including this list and there's a ton of stories about what they're doing, what they're doing with Whole Foods, how the future of Amazon Fresh, what they're going to do with Grocery in the whole, is because I thought there's so much here, actually we could move it into the Retail show. So, Suzy, host of the Retail show, you'll be covering that in a future episode. I haven't told you yet, but you're finding out through the public airwaves. But yes, I think there's definitely a conversation to be had there. What jumped out to you about that?
B
I mean, it's pretty much the same as previous quarters, but Amazon Fresh I think is kind of on the decline. I would say it's on its way out.
A
Sounds like it. Yeah. Especially if the uk, I believe they're packing up shop in the uk. Yeah.
B
No more Amazon Fresh for you UK listeners. But I think in general, what Amazon is trying to do with Grocery is interesting. It's trying to get people to order more, more groceries online, which is not necessarily a bad strategy. If you look at how Instacart and Doordash and Uber eats and all these companies are growing pretty quickly, but it's still a challenge. And because they don't have the physical footprint that Walmart has, I think it's a logistical challenge that they haven't yet figured out how to solve.
A
Yep. Yeah, I think that's a good one. Their grocery business still. Yeah. Physical store sales, still small. It's making about $6 billion a quarter, but it's only about 3% of their business. So it's not nothing, but they still haven't figured out the right direction to take and then double down on the investment there. The other two stories quickly that I was thinking about including in the list. One was Amazon stretching Black Friday and Cyber Monday into a 12 days of deals starting November 20th running through December 1st. I thought that was interesting because we've always talked about the kind of stretching out, the elongating of shopping holidays and how long would it just be the Cyber five and when would it become more than that. So I thought that was interesting. And then also second story, Amazon paying two and a half billion dollars to settle an FTC lawsuit after the agency accused the retail giant of knowingly duping users into its $15 a month prime program and making cancellation unreasonably difficult. But not big enough for the top three, I think. Anyway, that's my list and Rachel pretty much agrees. So happy to have you. Which is rare because that is true. She's normally very combative. Rachel, thank you so much for being on today's show.
B
Thanks for having me. This is a lot of fun.
A
Thank you to the whole production crew and to everyone for listening to behind the Numbers, the Market video podcast made possible by DG Media Network. Subscribe Follow as always, rating and review if you can and we will see you on Monday. We do have a special edition episode episode coming out tomorrow, so check that one out on Saturday, a Saturday special. Otherwise we will see you on Monday. Happiest of weekends.
Podcast: Behind the Numbers: An EMARKETER Podcast
Date: November 21, 2025
Host: Marcus Johnson
Guest: Rachel Wolff, Retail Briefings Analyst
This episode explores "The Big 3" questions currently shaping Amazon's future, focusing on workforce changes amidst AI investment, innovation in customer decision-making tools, and the advent of AI-enabled devices for delivery workers. Marcus and Rachel analyze these issues through the lens of retail, technology, and labor, considering both the immediate impact and long-term implications for Amazon and the larger market.
Is AI really the cause behind massive layoffs, or is the situation more complex?
Is Amazon’s new AI-powered recommendation tool a step toward ‘agentic’ commerce?
Are Amazon's AI-enabled glasses the gateway to widespread smart wearable adoption?
The episode balances industry analysis with practical retailer implications, maintaining a conversational, candid tone throughout. Marcus and Rachel bring skepticism toward headline-driven narratives around AI and job loss, stress the importance of transparency in AI-driven features, and offer nuance on both the technical leaps and public skepticism inherent in Amazon’s AI strategy.
Highly recommended for anyone looking to understand Amazon’s current inflection points—and where retail innovation might go next.