October 20, 2025 • 34 mins
The source provides a detailed examination of Starbucks' new strategy under CEO Brian Niccol, focusing heavily on the implementation of artificial intelligence to combat recent sales slumps and operational issues. Specifically, the episode introduces the "Green Dot" system, an AI-powered digital assistant designed to help baristas with complex recipes and troubleshooting to reduce errors and wait times, emphasizing that this technology is meant to augment, not replace, human employees. The article outlines Niccol’s background in successful retail turnarounds at Chipotle and Taco Bell, contrasting his efficiency-driven approach with ongoing labor tensions and union skepticism regarding the AI rollout. Ultimately, the source explores the implications of this AI pivot for both customer experience through personalized ordering and the broader industry trend of integrating sophisticated technology into the hospitality sector.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:00):
Welcome back to the deep dive. Today, we're strapping in
for a look at the well, the pretty high stakes
world of global retail, and we're zeroing in on a
company that, let's face it, basically defined coffee culture for decades, Starbucks.
Speaker 2 (00:14):
Right, that whole third place idea, you know, the spot
between home.
Speaker 1 (00:18):
And work, exactly, that cozy hub. But well, lately, if
you've been in one, maybe you've noticed, and our sources
agree that magic it's been let's say, significantly diluted.
Speaker 2 (00:31):
And that dilution it's not just a feeling, it translates
directly into well, financial strength. Yeah, we're looking at a
really critical moment for Starbucks, a moment of trying to
reinvent their operations.
Speaker 1 (00:40):
The pressure must be immense, Oh absolutely.
Speaker 2 (00:42):
I mean, look at the numbers, same source. Sales dipped
and dipped hard, a painful seven percent in the fiscal
year ending September twenty twenty four. That's not just a
slow quarter. That's bordering on a crisis of identity, you know,
and execution.
Speaker 1 (00:55):
Okay, So enter the crusader, right, a new CEO, Brian Nicol.
He's been brought in specifically because he's got this reputation,
this track record of resurrecting major brands.
Speaker 2 (01:04):
And his playbook is very clear. Yeah, data driven, efficiency
focused discipline.
Speaker 1 (01:11):
And the core strategy we're really diving into today is
what he calls his Ai Barista, this green dot system.
He unveiled it back at the Dreamforce conference October sixteenth,
twenty twenty five.
Speaker 2 (01:22):
Right, So our mission here is to really understand this
tech because Ai Barista, I mean, it makes you think
of robots making your lante.
Speaker 1 (01:30):
Tually, like the Jetson's or something.
Speaker 2 (01:31):
Yeah, exactly. But what this deep dive shows is something, well,
something far more subtle and maybe frankly, far more impactful
in the near term.
Speaker 1 (01:40):
So the big question is, is this green dot tech
the first step towards, you know, automating the human barista
right out of a job, or is.
Speaker 2 (01:48):
It a smarter, more sophisticated evolution. Yeah, something that actually
solves the real pain points, the long lines, the messed
up drinks, and maybe allows that human element to actually
come back to reclaim the sort of artistry of making coffe.
Speaker 1 (02:00):
That's the core tension we're tracking then, automation versus augmentation.
So let's start with the context. We need to get
a handle on just how serious the situation was that
Nicole walked into. He took over September ninth, twenty twenty four,
replacing Laxman Nearasimhan.
Speaker 2 (02:15):
And that's seven percent dip in same store sales. I
mean for a company the size of Starbucks, that's catastrophic.
It signals a massive, massive operational failure somewhere.
Speaker 1 (02:25):
And it wasn't just numbers on a spreadsheet, right, it
was the experience exactly.
Speaker 2 (02:29):
It was experiential. You the customer, were dealing with these
just frustratingly long lines. Yeah, but maybe even more damaging
inconsistent drinks.
Speaker 1 (02:38):
Ah the worst. You pay what seven bucks for a complicated.
Speaker 2 (02:41):
Drink and it's different every single time. The menu complexity
just completely outstripped their ability to deliver consistently that ritual,
that reliable morning coffee. It just dies.
Speaker 1 (02:52):
And that whole third place vibe that Howard Schultz built
the empire on, yeah.
Speaker 2 (02:56):
Just crumbled under the chaos. Hard to feel cozy when
it's frantic and your drink is wrong.
Speaker 1 (03:00):
So Nicole comes in with this very public mission. He
calls it the back to Starbucks crusade, and he's not
being shy about AI is he not at all?
Speaker 2 (03:08):
He told Yahoo Finance quote, He's all in on AI
and the goal pretty ambitious, becoming the world's greatest customer
service company. Again.
Speaker 1 (03:19):
Wow, Okay, so that's basically declaring war on inefficiency.
Speaker 2 (03:22):
It absolutely is, and it's really vital we clarify this
up front. The AI beuris isn't a robot arm.
Speaker 1 (03:28):
Pouring milk right, not yet anyway, not yet.
Speaker 2 (03:30):
It's a real time digital assistant. Think of it like
an advanced layer of intelligence designed to guide and optimize
what the human beriast is doing, making sure every step,
you know, grinding the beans, steaming the milk, is perfect consistent.
Speaker 1 (03:43):
And to really get the scale of this tech bet,
we absolutely have to look at the guy placing the bet.
Brian Nickel, the architect. His playbook is all about resurrection
through data, right.
Speaker 2 (03:53):
His resume is incredibly specific on this point. He's known
for taking these big established brands that have kind of
lost their way, stagnated exactly, and driving them back to profitability.
And he does it largely by focusing on digital efficiency
and crucially simplifying the in store experience.
Speaker 1 (04:09):
His time at Taco Bell over a decade there. That
seems like a precursor. He didn't just like climb the ladder.
Speaker 2 (04:15):
No, he fundamentally shifted how people interacted with Taco Bell.
He pushed the breakfast menus, which was huge for them,
and critically he rolled out app based ordering that became
the standard across the entire fast food industry later on.
Speaker 1 (04:28):
And that app wasn't just about making it easier to
order at Chilupa, was it.
Speaker 2 (04:32):
Not at all? It was about data capture, that whole
digital pivot. It's credited with boosting Taco Bell's sales by
a huge twenty percent. He proved really early on that
you streamline ordering and gather customer data, you unlock serious growth.
Speaker 1 (04:47):
But the real high wire act, the thing that probably
got in the Starbucks job, was turning around Chipotle.
Speaker 2 (04:52):
Oh absolutely, that was his masterpiece. Arguably. He took over
his CEO in twenty eighteen. And remember Chipotle was reeling
major Ecali health scares. The brand was just stagnant. People
were scared to eat.
Speaker 1 (05:03):
There, and the results under his watch phenomenal, just phenomenal.
Speaker 2 (05:07):
Under Nickel, Chipotle essentially doubled its revenue, doubled to nine
point eight billion dollars by twenty twenty four. He didn't
just stabilize it. He injected massive growth, mainly through modernization.
Speaker 1 (05:18):
And we know from the sources that AI was really
the backbone of that Chipotle turnaround, digital kiosks, loyalty programs,
and that really advanced AI for inventory forecasting.
Speaker 2 (05:29):
Yes, that inventory AI at Chipotle is key. It wasn't
just some fancy tech gimmick. It was a fundamental cost saver.
It cut food waste by fifteen percent.
Speaker 1 (05:39):
Fifteen percent with fresh ingredients like they use. That must
be huge.
Speaker 2 (05:42):
It's enormous. A fifteen percent reduction in spoilage. They go
straight to the bottom line hundreds of millions of dollars.
It just confirms Nichol's pattern. AI isn't hype for him.
It's a serious, quantifiable operational tool.
Speaker 1 (05:54):
Okay. So this history, this sort of Midas Touch reputation,
it explains the frankly staggering commitment the Starbucks board made
to get him.
Speaker 2 (06:02):
It really speaks volumes about how desperate they were operationally speaking.
Nicholas compensation package in twenty twenty four it was nearly
ninety seven point eight million dollars, Yeah, including a ten
million dollars signing bonus, and then a jaw dropping seventy
five million dollars in performance based equity. A figure like
that just signals absolute faith that he can do the
Chipotle thing all over again, but on an even bigger scale.
Speaker 1 (06:25):
And Wall Street bought it immediately, didn't they. The stock jumped.
Speaker 2 (06:28):
Twenty four point five percent instantly, just on the announcement
he was appointed. Wall Street definite believes in the power
of the data driven executive.
Speaker 1 (06:36):
So he's already running experiments, these starting five programs.
Speaker 2 (06:39):
Yeah, testing innovations internally, but this Green Dot AI this
is the big one, the bold centerpiece of this massive
financial bet, this whole strategic shift.
Speaker 1 (06:49):
Okay, but we have to be impartial here. We need
to talk about the controversies that seem to follow this
efficiency first playbook, especially when it comes to labor, right.
Speaker 2 (06:58):
And that history is important context for what happening at
Starbucks now at Chipotle. His time as CEO included that
really high profile closure of a store in Augusta, Maine,
back in twenty twenty two, and that store was in
the process of unionizing.
Speaker 1 (07:13):
And the NLRB, the National Labor Relations Board, they ruled
on that didn't.
Speaker 2 (07:17):
They they did. They ruled it was illegal union busting,
so that hangs in the background.
Speaker 1 (07:20):
And now he walks straight into Starbucks where labor tensions
are already I mean at a fever pitch. How many
stores have unionized.
Speaker 2 (07:28):
Over four hundred since twenty twenty one. It's a significant movement,
and the source material really highlights the let's say, stark
optics of the friction. You've got reports of Nickel commuting
via private jet from his home in Newport Beach to Seattle.
Speaker 1 (07:44):
While baristas are pushing for fair contracts.
Speaker 2 (07:46):
Exactly pushing for better wages of better conditions. It creates
this immediate, pretty dramatic clash of cultures, priorities, and.
Speaker 1 (07:54):
The union's reaction to the AI push not positive.
Speaker 2 (07:57):
Union spokesperson Jasmine Leli was quick to jump on it.
She basically called the whole AI initiative a distraction, a
way to divert attention from quote declining sales and brand perception.
Speaker 1 (08:08):
Her argument being that fancy tech doesn't matter if the
workers aren't valued.
Speaker 2 (08:11):
Pretty much, that operational efficiency is kind of irrelevant if
the workforce feels exploited or isn't fairly paid. So Nichol's
challenge is not just technical, it's absolutely a cultural battle
for the soul of Starbucks.
Speaker 1 (08:24):
Okay, let's pivot to the technology itself. Then let's look
under the hood. If this is the big solution, what
exactly is the green dot Assist? How does it actually
work when a barista is, you know, slammed during the
morning rush.
Speaker 2 (08:35):
Well, the green Dot Assist it's actually surprisingly simple on
the front end how it looks, but incredibly complex behind
the scenes. It's basically an iPad beamed AI interface and
it just appears, like the name suggests, as a small
green dot on the barista screen.
Speaker 1 (08:52):
Okay, a green dot.
Speaker 2 (08:53):
And then what when things get hectic or maybe they
get a really complex order, the barista just taps the
dot and it summons up contextual, real time help right
there on the screen.
Speaker 1 (09:03):
So it's like having i don't know, an instant expert,
a perfect recipe book and troubleshooter rolled into one right
when you need it.
Speaker 2 (09:09):
Exactly that tailored for that exact moment. Think about it.
Say it's eight thirty am peak rush. A new hire,
maybe it's their first week, gets hit with like the
dreaded venty iced white MoMA.
Speaker 1 (09:21):
Oh yeah, those modifiers right.
Speaker 2 (09:23):
Venty iced white mocha, but with almond milk, extra pumps
of sugar free vanilla, and oh, just a light amount
of camel drizzle. The barista's head is spinning.
Speaker 1 (09:33):
Okay, so they tap the green dot.
Speaker 2 (09:35):
They tap the dot and boom. The screen instantly shows
the precise number of syrup pumps needed for that size
in those mods, the exact volume of almond milk, and
crucially it flags the allergen info because of the almond
milk swap.
Speaker 1 (09:48):
Wow, that is huge, especially for those super customized drinks
that are basically the norm.
Speaker 2 (09:54):
Now.
Speaker 1 (09:54):
It takes the mental load off right, the calculation under pressure,
that's where mistakes happen.
Speaker 2 (09:59):
That's exactly where the happen. It doesn't just do recipes.
It helps troubleshoot equipment too. Say the steam wand pressure
suddenly feels low. Tap the dot. It might suggest exactly
what to check first. It's like having a dynamic, always
available expert shoulder surfing.
Speaker 1 (10:12):
With you and the brain behind This is massive, you said.
It's powered by Microsoft's Azure Open Ai. They announce that
partnership back in June twenty twenty five. What does that
tell us about the data it's using.
Speaker 2 (10:23):
It tells us this system has access to truly enormous
data sets. It's pulling from every single recipe variation. Starbucks
is ever offered every equipment manual for every machine in
the store. And this is key, the historical record of
every single order air the company has.
Speaker 1 (10:42):
Ever logged, every mistake ever made.
Speaker 2 (10:44):
Essentially, yeah, it uses that huge pile of data to
predict where mistakes are most likely to occur in a
given situation, and it tries to step in before they happen.
Speaker 1 (10:53):
So, just to be clear, it's not like chat GPT.
It's not generative AI writing palms about coffee.
Speaker 2 (10:58):
No, No, definitely not. It's a highlight targeted, highly optimized
knowledge retrieval system purely focused on operational consistency and efficiency.
Speaker 1 (11:06):
Got it, and the pilot programs the tests as they ran,
they're already showing some pretty impressive quantifiable results stuff Nickel
can point too.
Speaker 2 (11:13):
Absolutely, the numbers speak directly to his data driven approach.
Speaker 1 (11:16):
Okay, let's look at those numbers. The training time reduction
seems like a big one.
Speaker 2 (11:20):
It is green dot cut the time it takes to
train a new hire from like weeks down to just.
Speaker 1 (11:25):
Days weeks to days.
Speaker 2 (11:27):
Yeah, think about the impact of that. Instead of spending
maybe two weeks just trying to memorize hundreds of recipes
and troubleshooting steps, the new Beerrista has this instant digital
cheat sheet, gets some up to speed productively making drinks
way faster.
Speaker 1 (11:42):
And the financial impact the waste production that sounds enormous too.
Speaker 2 (11:46):
It cut drink remake rates by twelve percent in the
test stores. Now you factor in the cost of wasted ingredients,
the barista's time remaking it, the customers lost time. The
company estimates that errors and waste costs roughly fifty thousand
dollars a year per store.
Speaker 1 (12:02):
Fifty thousand per store. Wow.
Speaker 2 (12:05):
Yeah, so a twelve percent cut in that figure across
thousands of stores nationwide. That is a massive financial win.
And it's better customer service right, you get your drink
right the first time. It's a direct margin improvement.
Speaker 1 (12:15):
And the rollout is moving fast. It's not just a
test anymore.
Speaker 2 (12:18):
No, this is happening. It's rolling out across North America. Now.
Over two thousand US stores had it by October twenty
twenty five, and the plan is a global expansion in
twenty twenty six. This has basically become the core operating
system for the stores.
Speaker 1 (12:31):
And Nicol sees it going beyond just making drinks correctly.
Predictive maintenance.
Speaker 2 (12:35):
Yeah, that's his longer term vision. He wants the system
to evolve so it can analyze subtle signs from the
equipment itself. Like maybe it monitors the specific vibration pattern
of a coffee grinder over time.
Speaker 1 (12:46):
And flags it before it breaks down.
Speaker 2 (12:48):
Exactly, alert the manager, Hey, grinder three's vibration signatures, looking
if he might need service soon before he just completely
fails midshift.
Speaker 1 (12:56):
Preventing that downtime is huge. If the espresso A machine
goes down or the main grinder fails, the store is
instantly crippled, isn't it.
Speaker 2 (13:06):
It happens more often than you'd think. The sources say
equipment failure downtime affects about ten percent of shifts. So
if the AI can predict that a steam waand needs
preemptive service based on its past performance data, it avoids
that failure. It maintains that consistency the customer expects.
Speaker 1 (13:22):
And there are other AI tools already in play too, right,
like the inventory AI you mentioned from Chipotle.
Speaker 2 (13:27):
Right, that's a direct import from his Chipotle playbook. It's
already in five thousand stores. It uses computer vision basically
Cameron's mounted in the stockroom to automatically count inventory.
Speaker 1 (13:36):
So no more manual counting with clipboards.
Speaker 2 (13:39):
Pretty much. The AI literally recognizes the shapes and volumes
of syrup bottles, milk cartons, cup stacks. It calculates inventory
levels in real time without a human needing to scan
every barcode.
Speaker 1 (13:52):
That's seriously sophisticated. What's the impact. What does that actually
do for the store.
Speaker 2 (13:56):
Well, the big claim is it cuts the physical time
baristas and manager spend back in the stock room counting
stuff by thirty percent.
Speaker 1 (14:03):
Thirty percent less time counting boxes.
Speaker 2 (14:05):
Yeah, and that time theoretically gets redeployed out front, making drinks,
maybe talking to customers. And crucially, it helps ensure the
staples you know, like the infamous pumpkin spice, are always
ordered on time and kept in stock. Prevents those frustrating
sorry we're out moments.
Speaker 1 (14:22):
Okay, but this level of automation, this minute tracking of inventory,
even the help from green Dot, it immediately brings up
the other side of the coin surveillance, Yeah, employee monitoring.
We have to talk about neurospot.
Speaker 2 (14:33):
Yeah, this is where that tension between efficiency and the
human element really gets sharp.
Speaker 1 (14:38):
Ye.
Speaker 2 (14:38):
Back in June twenty twenty five, there's this viral post
on x the platform formerly known as Twitter. It highlighted
this existing AI system Starbucks was using called Neurospot, and
Neurospot apparently tracks barista productivity and something they called dwell
times twelve times.
Speaker 1 (14:54):
That sounds orwellian, like how long you stand in one place?
Speaker 2 (14:57):
Exactly how long an employee spends in one spot? Is
it too long? Talking to a customer? Too long? Resocking?
It immediately sparked these big brother fears. People worried that
green dot, even if helpful, could be weaponized use for scrutiny,
not support.
Speaker 1 (15:12):
So what's Starbucks's official line on this? How do they
separate green dot from Neurospot.
Speaker 2 (15:17):
Their official response is that green dot is opt in,
meaning the barista chooses to tap the dot for help,
and that the data collected is anonymized. They insist it's
focused purely on system support overall performance improvements, not on
tracking or punishing individual baristas. The stated goal is support.
Speaker 1 (15:34):
Not scrutiny, but the shadow of neurospot is still there.
It leads straight to that question from the ex user
at Rainmaker in nineteen seventy three, efficiency or exploitation. Where's
the line?
Speaker 2 (15:44):
It's the tightrope nick All has to walk, isn't it
net The data is being collected, no doubt about that.
Whether it's ultimately used to guide the barista or to
judge the barista, that defines whether this whole strategy gets
embraced or rejected by the actual workforce on the ground.
Speaker 1 (15:59):
Okay, this this brings us squarely to the human equation.
If the tech is driving this relentless efficiency, how is
Starbucks convincing it's what two hundred and forty thousand baristas
globally that this AI is their friend, not their replacement.
Speaker 2 (16:14):
Well, Nicole's public messaging is relentlessly focused on augmentation, not replacement.
He keeps saying he wants more real baristas back in stores,
more baristas, not fere That's the claim, and it's significant
because it lines up with this other strategic shift they made,
pulling back from those super efficient but maybe soulless mobile
only store formats they were testing. They actually halted those
(16:35):
in July twenty twenty five, so they're consciously saying they
want to pivot back towards that human centric third place idea.
Speaker 1 (16:43):
So the pitches let the AI handle the annoying complex stuff,
the drudgery, as the sources put.
Speaker 2 (16:49):
It right, the remembering of obscure recipes, the troubleshooting.
Speaker 1 (16:52):
And that frees up the human barista to actually focus
on the customer interaction, the craft, the welcoming vibe.
Speaker 2 (17:00):
That is absolutely the promise, and look at the potential
benefits for the workforce if it works as advertised. Starbucks
has this notoriously high annual turnover eate. It's like one
hundred and fifty percent.
Speaker 1 (17:11):
Wow, one hundred and fifty percent. That means they replace
basically their entire staff every eight months or so pretty much.
Speaker 2 (17:17):
It's incredibly high. So green Dot could offer new hires
and remember many are teenagers, juggling school gives them instant expertise.
It reduces that initial feeling of being completely overwhelmed, which
is often what leads to burnout and people quitting quickly.
Speaker 1 (17:31):
And stabilizing the workforce, reducing that churn. That's crucial for
the customer experience. Too, isn't it high turnover means you
rarely see a familiar face, kills that neighborhood third place
feeling exactly.
Speaker 2 (17:41):
So if green doc can actually reduce that one hundred
and fifty percent turnover by making the job less stressful,
more manageable, it fosters a more experienced, stable team, and
it helps veterans too. There's this anonymous brist to post
on X that really summed it up. They said something like,
finally tech that doesn't make us feel stupid for forgetting.
Speaker 1 (17:59):
A pump count ah. Interesting, So it's like error proofing
for everyone, right.
Speaker 2 (18:03):
It allows the experienced priests to stop worrying about tiny
details and concentrate on real personalization like remembering your regular
word extra hot, no foam, oat milk, that kind of thing,
the stuff that makes you feel recognized.
Speaker 1 (18:16):
Okay, that's the potential upside for the employees. But the
union opposition is still strong. They seem to view this
whole AI push through a very different lens, one of
distraction and ongoing conflict.
Speaker 2 (18:28):
They are deeply skeptical of the motives. Absolutely. Starbucks Workers
United even put out this satirical video around October seventeenth,
twenty twenty five, basically portraying Nichol as some kind of
robot overlord. It just highlights the profound mistrust that exists.
Speaker 1 (18:42):
And their spokesperson, Jasmine Leylai, her main argument is that
this tech focus is just a smoke screen to hide
the real issue.
Speaker 2 (18:51):
Yes, her argument is that it's a calculated move to
distract everyone from the core issue of wage stagnation.
Speaker 1 (18:56):
Okay, so that's the heart of the labor fight. Then, wages,
that's the crux of it.
Speaker 2 (19:00):
Baristas are currently earning somewhere between fifteen and eighteen dollars
an hour typically, and this is while facing what's seven
percent inflation. The unions are demanding significantly more that, pushing
for twenty dollars plus per hour minimum and some form
of profit sharing.
Speaker 1 (19:14):
Their argument being if the company can afford nearly one
hundred million dollars for the CEO and millions more for
AI systems, they.
Speaker 2 (19:21):
Can afford a living wage for the people actually making
and serving the coffee. That's the core contention.
Speaker 1 (19:26):
So how is management responding to that specific wage critique.
Are they saying the AI savings will pay for raises or.
Speaker 2 (19:34):
Nickel's trying to frame it differently. He's trying to show
that the efficiencies gained through AI are actually enabling growth
and allowing for higher wages, rather than being about cutting
staff to save costs. He announced plans to hire ten
thousand more baristas by twenty twenty six, ten thousand more yeah,
claiming these new jobs are essentially funded by the AI
efficiencies making the stores more productive. And importantly, he's planning
(19:56):
to raise the minimum wage floor to eighteen dollars an
hour across the board. Eighteen dollars eighteen, So it's a
strategic move right. It directly counters the union narrative that
AI automatically means job losses, but it still falls short
of that twenty dollars plus demand the union is making.
Speaker 1 (20:12):
And that ethical tightrope round monitoring it doesn't go away.
Even if green dot is opt in. If other systems
like neurospot are tracking dwell times or productivity, how do
you guarantee that data isn't used, maybe even unintentionally, to
pressure buristas or penalize them for taking that extra moment
to chat with the customer, the very thing Nichol says
he wants more of.
Speaker 2 (20:32):
That is the absolute key question for whether this whole
thing works long term. Starbucks says they use diverse data sets,
human oversight, all sorts of things to try and mitigate
bias risk in the AI. But let's be real, as
long as performance is being measured somehow, even if it's
aggregated or anonymized, the system creates a metric, and metrics
inevitably get used to justify decisions.
Speaker 1 (20:54):
Right, what gets mattered gets managed.
Speaker 2 (20:56):
Exactly, So the success of this entire turnaround it really
rests on whether the barista community, the actual people on
the floor yeah, truly come to believe that green Dot
is an ally helping them do their job better, or
if it feels like a sophisticated warden constantly monitoring their
every move.
Speaker 1 (21:13):
Okay, let's shift focus from the internal friction. Now, let's
talk about the customer. Why should the what is it
one hundred million people who visit Starbucks every week actually
care about this AI revolution behind the counter? How does
it change my experience?
Speaker 2 (21:26):
Well, it starts with making ordering smoother, more frictionless, and
this is where Nicol dropped what the source is called
his boldest tease, an app that actually predicts.
Speaker 1 (21:36):
Your order, predicts my order, Yes, Starbucks.
Speaker 2 (21:38):
Already has what thirty million active users in its loyalty app.
That's a ton of data. This new system uses advanced
natural language processing NLP to understand your past orders, the
time of day, maybe even the weather.
Speaker 1 (21:50):
Okay, so the app opens up and it might already
suggest my usual VENTI iced coffee to pumps vanilla splash
a cream exactly.
Speaker 2 (21:57):
It anticipates what you likely want. It moves that little
bit of friction, that decision fatigue makes the whole process
feel seamless.
Speaker 1 (22:04):
And that seamlessness they expect that to drive loyalty significantly.
Speaker 2 (22:08):
They're predicting it could boost loyalty program uptake by another
twenty percent. The whole digital overhaul is about shrinking the
gap between you thinking I want coffee and you actually
holding the cup. And to make this happen, they're rolling
out updated point of sale systems, you know, the screens
and registers the barist has used to all forty thousand
stores globally by the second quarter of twenty twenty six,
(22:30):
forty thousand stores.
Speaker 1 (22:31):
Wow. And these new POS systems they also use AI
for dynamic pricing and upsells. Is that right?
Speaker 2 (22:38):
Yes? And this is where the personalization gets really sophisticated.
The AI looks at what's in your basket, what you're
buying right now, plus your past history, and it generates
these really relevant, hopefully subtle up cells.
Speaker 1 (22:50):
So instead of just asking everyone want a pastry.
Speaker 2 (22:52):
With that right, it might analyze your history and see
you often by baked goods on Tuesday mornings, So the
screen might the barista or maybe even the app suggests
add a blueberry scone for just one dollar. It's tailored,
designed to gently increase that average ticket size without feeling
a knowing or generic.
Speaker 1 (23:11):
Okay, it makes sense now. Shifting to speed. You mentioned
the drive through is the big bottleneck, right, It accounts
for half of all US sales. What's the AI doing
to tackle those long lines.
Speaker 2 (23:23):
They're implementing something called smart Q. Think of it like
an intelligent air traffic control system, but for coffee orders.
Speaker 1 (23:29):
Smart Q Okay, how does that work?
Speaker 2 (23:31):
Instead of just a simple first order in first order
made system, smart Q dynamically routs orders. If a really
complex freppuccino comes in, it might rout it to a
station that's less busy or better equipped for blending simpler
orders like a black coffee get routed to the absolute
fastest path. It maximizes the throughput of the whole system.
Speaker 1 (23:50):
So it's not just about individual barista speed, it's about
managing the flow of all the orders together.
Speaker 2 (23:56):
Precisely, it's traffic management for drinks, and the goal is
to minimize the time cars spend just sitting there idling
in the drive through lane, which is obviously better for
customer satisfaction, but it also helps with their environmental goals
cutting down on vehicle emissions, especially in busy urban areas.
Speaker 1 (24:14):
And the scale of this personalization using all that transaction data,
it's kind of mind boggling. How many transactions did you
say they analyzed daily?
Speaker 2 (24:24):
Seventy five million daily. That's the fire hose of data.
The AI is constantly learning from that fine five million,
and that scale allows for incredibly specific tailored offers. Like
the example given was, maybe the AI notices you only
ever buy a nitro cold brew when the weather forecast
predicts rain.
Speaker 1 (24:40):
Okay, random but plausible.
Speaker 2 (24:42):
Right, So on the next rainy day, maybe the app
offers you a free upgrade to a larger sized Nitro.
It's that granular level of personalization they're aiming for and.
Speaker 1 (24:50):
Are they seeing results from this kind of thing.
Speaker 2 (24:53):
They are in the pilot locations where they tested these
personalized offers. Customer satisfaction score is apparently lifted by fifteen points.
That's significant, and this is really the ultimate goal, the
conversions point Nicol is shooting for. The AI handles all
the complex data, the logistics, the predictions.
Speaker 1 (25:09):
Which frees up the human barista exactly.
Speaker 2 (25:11):
To deliver that moment of connection, that quick chat how's
your week going, The stuff that actually defined the third
place idea in the first place, the warmth.
Speaker 1 (25:20):
Okay, But analyzing seventy five million daily transactions, predicting orders,
tracking preferences, that immediately throws up huge red flags around privacy,
especially with regulations like GDPR in Europe and CCPA in California.
Speaker 2 (25:35):
Absolutely, privacy and trust are non negotiable here. Starbucks is
stressing that they're taking steps to anonymize the vast majority
of this data. They have to given the scrutiny under
those global frameworks. Nicol even said a dreamforce directly, trust
is our brew. Yeah, but they're constantly fighting the perception
and the reality of past security incidents like that breach
(25:57):
back in twenty twenty two, that affected about one thousand users.
They need robust, easy to use opt out mechanisms and
total transparency about how data is used and secured. If
they lose customer trust on this, the whole personalization engine collapses.
Speaker 1 (26:12):
Okay, so we've spent a lot of time on Starbucks's
specific situation and their AI solution. Let's zoom out a
bit now put their moves into the broader context, the
coffee cosmos, as you call it. Is this green dot
system really that revolutionary or are competitors doing similar things?
Speaker 2 (26:28):
That's a great question. Starbus is definitely making the biggest,
most integrated, and most high profile push right now, but
AI is absolutely percolating part in the pun across the
entire hospitality sector. Competitors are definitely using similar types of tools,
even if they're not as centralized or branded as Green Dot.
Speaker 1 (26:44):
Okay, like, who what are others doing well?
Speaker 2 (26:46):
Duncan, for instance, they're reportedly using AI mainly for demand forecasting,
trying to predict how busy a store will be based
on whether local events, time.
Speaker 1 (26:55):
Of day, to optimize staffing levels exactly.
Speaker 2 (26:57):
They claim it's helped reduce overstaff costs by about ten percent.
Then you have Dutch Bros. They're apparently piloting voice AI
systems in their drive thrusts, trying to automate the basic
order taking part to speed things up.
Speaker 1 (27:10):
And even smaller players high end places.
Speaker 2 (27:13):
Yeah, even some independence like Bluebottle or using simpler AI
chat pots, but maybe just for managing reservations or answering
basic questions online. It's becoming table stakes at various levels.
Speaker 1 (27:23):
And then there's the really far end of the spectrum,
full automation, like that Cafe x place in San Francisco,
the robot barista, right.
Speaker 2 (27:30):
That's the alternative model. Cafe x has had a robotic
arm making drinks since twenty seventeen, and it's efficient, makes
about one hundred and twenty drinks an hour, but it's expensive.
That fifty thousand dollars per unit cost you mentioned is
a huge barrier for widespread adoption. It keeps it pretty niche.
Speaker 1 (27:45):
So Starbucks's strategy is fundamentally different.
Speaker 2 (27:48):
It really is. It's not about replacing the human with
a robot. It's about using AI's efficiency to empower the
human barista, trying to create this symbatic relationship. Augmentation, not
full automation.
Speaker 1 (27:58):
That's the bet aiming to be the world's greatest customer
service company means scaling this globally across eighty countries that
must present enormous logistical and regulatory challenges for green Dot.
Speaker 2 (28:11):
Oh, the hurdles are immense Logistically, just getting the hardware,
the software, updates, the training rolldout consistently across such a
diverse network is a nightmare.
Speaker 1 (28:20):
And the regulations.
Speaker 2 (28:22):
The regulations are a minefield, especially in places like the
European Union with their EU AI Act. That act classifies
AI systems based on their potential risk. A system like
green Dot, because it tracks performance, interacts with employees, potentially
influences things like scheduling or even job reviews.
Speaker 1 (28:40):
Down the line, it could be deemed high risk.
Speaker 2 (28:42):
It's possible, and if it is, it triggers much stricter
requirements around transparency, data governance, accuracy, human oversight. The compliance
burden becomes way heavier than just rolling out a new
app feature in the US.
Speaker 1 (28:53):
That definitely complicates a global.
Speaker 2 (28:55):
Rollout absolutely, And then you have the simple realities on
the ground. Think about rural stores, maybe in less developed areas,
do they even have the reliable high speed internet bandwidth
needed to run this kind of cloud based real time
AI effectively maybe not?
Speaker 1 (29:12):
And cultural differences. Does the AI understand that a Macha
latte in Japan is totally different from one in Peoria?
Speaker 2 (29:17):
Exactly? They can't just translate the interface. They need truly
localized AI models that understand regional tastes, ingredient availability, complex
local preferences. It's not a one size fits all solution.
Speaker 1 (29:29):
So all this complexity, the costs, the potential pitfalls, it
naturally breeds skepticism, doesn't it. Can Nickel actually pull this off?
Speaker 2 (29:36):
The skepticism is definitely out there, and it's loud. You
saw that piece on AOL back in October, basically asking
if this whole AI push is just mere hype, wondering
if Starbucks is just slapping an AI powered label on
everything to create buzz, suggesting maybe we're in a bit
of an AI bubble generally.
Speaker 1 (29:52):
And the fear of job losses doesn't go away either.
Speaker 2 (29:54):
Not at all. You see it all over social media
accounts like at zero hedge on x just quipping next robots.
It reflects that underlying anxiety about automation about where this
all leads for workers.
Speaker 1 (30:07):
But Nicol he has the counter argument, right, he has
the data from the pilot stores.
Speaker 2 (30:11):
That's his strongest weapon in this perception battle. Those early
tangible results that Q three twenty twenty five data showing
copparable sales. We're up two percent in the store's testing
green dot. That's his proof of concept, proof.
Speaker 1 (30:24):
That it's not just TYU but actually moves the.
Speaker 2 (30:26):
Needle right, proof that customers are responding positively to the
improved consistency, the faster service, and that data fuels the
other side of the ex conversation. You get users like
a Junior Stock celebrating it, hailing it as data driven delight.
It really is a battle of narratives being fought with
performance metrics.
Speaker 1 (30:42):
Let's talk finally about the cost, because this isn't cheap.
The financial investment and also the environmental cost.
Speaker 2 (30:49):
The financial commitment is staggering. Nichol isn't just tweaking software here.
He's committed to a massive physical transformation too. They're aiming
to remodel one thousand stores by twenty twenty six.
Speaker 1 (31:00):
One thousand stores, and each remodel.
Speaker 2 (31:02):
Costs around half a million dollars each.
Speaker 1 (31:05):
Wow, so that's half a billion dollars just on refreshing
the physical stores yep, and.
Speaker 2 (31:10):
The entire return on that massive investment. It hinges completely
on whether these AI driven improvements, the speed, the accuracy,
the personalization actually drive enough new customer traffic and get
people to spend slightly more each visit to justify that
colossal expense, huge gamble.
Speaker 1 (31:28):
And the environmental footprint. Running AI at this scale, analyzing
seventy five million transactions daily, predictive maintenance checks, inventory scans,
it takes a lot of server power, right, A lot
of energy.
Speaker 2 (31:39):
It does, and that's a key part of their public commitment.
Starbucks knows this. They've pledged to offset that increased energy
usage through investments in renewable energy. They're sticking to their
ambitious target of being carbon neutral by twenty thirty. So
they have to constantly balance this push for high tech
cloud based efficiency with their very public, very scrutinized environmental promises.
Speaker 1 (31:59):
Okay, so as we wrap up this deep dive, let's
bring it back to that core tension we identified at
the start, the balancing act that defines this whole turnaround effort.
Speaker 2 (32:08):
Right, Starbucks is essentially fighting an operational war on two fronts.
On one hand, Nickel is deploying his proven playbook, data
driven efficiency embodied by the green Dot system, to fix
years of inconsistency and customer frustration, fixing the mechanics, but
on the other hand, he's simultaneously trying to reclaim that softer,
(32:29):
human centric artismal third place feeling, the vibe that originally
made Starbucks well Starbucks.
Speaker 1 (32:35):
So the success of this whole thing, the green Dot,
the AI strategy, it really hinges on finding that perfect
equilibrium between the tech optimization and the human touch.
Speaker 2 (32:44):
It absolutely does. If the system tips too far one way,
if it feels too much like rigid scrutiny or cold efficiency,
if it kills employee morale or makes the stores feel sterile,
then the entire turnaround could fail, regardless of how technically
brilliant the AI is.
Speaker 1 (32:58):
Because of that ambitious goal they have hitting one hundred
billion dollars in revenue by twenty thirty.
Speaker 2 (33:03):
That relies completely on these AI gains translating into both
genuine speed and genuine customer satisfaction. The tech has to
truly serve the people, involve both the breezes making the
coffee and you, the customer drinking it. It can't just
supplant them or treat them like you know, predictable variables
in an algorithm.
Speaker 1 (33:23):
Which leaves us and you with a final, maybe provocative
question to think about. It builds on the strange duality,
doesn't it? Efficiency and humanity? If AI is now handling
almost every logistical piece of getting your coffee, checking the stock,
predicting your order using NLP based on millions of data points,
telling the breis exactly how many pumps, troubleshooting the machine,
(33:44):
how far can those buristas really push the idea of
personalization and connection before that human element itself starts to
feel well, like just another optimized feature, precalculated, prompted and
scheduled into your experience by that same powerful algorithm.
Speaker 2 (33:59):
Yeah, when the entire process is so perfectly optimized by data,
what's left that is truly uniquely human about grabbing a
cup of coffee?
Speaker 1 (34:07):
Something to definitely mull over next time you tap your
app for that perfectly calculated, perfectly consistent cup. Thanks for
joining us for the deep dive
Welcome back to the deep dive. Today, we're strapping in
for a look at the well, the pretty high stakes
world of global retail, and we're zeroing in on a
company that, let's face it, basically defined coffee culture for decades, Starbucks.
Speaker 2 (00:14):
Right, that whole third place idea, you know, the spot
between home.
Speaker 1 (00:18):
And work, exactly, that cozy hub. But well, lately, if
you've been in one, maybe you've noticed, and our sources
agree that magic it's been let's say, significantly diluted.
Speaker 2 (00:31):
And that dilution it's not just a feeling, it translates
directly into well, financial strength. Yeah, we're looking at a
really critical moment for Starbucks, a moment of trying to
reinvent their operations.
Speaker 1 (00:40):
The pressure must be immense, Oh absolutely.
Speaker 2 (00:42):
I mean, look at the numbers, same source. Sales dipped
and dipped hard, a painful seven percent in the fiscal
year ending September twenty twenty four. That's not just a
slow quarter. That's bordering on a crisis of identity, you know,
and execution.
Speaker 1 (00:55):
Okay, So enter the crusader, right, a new CEO, Brian Nicol.
He's been brought in specifically because he's got this reputation,
this track record of resurrecting major brands.
Speaker 2 (01:04):
And his playbook is very clear. Yeah, data driven, efficiency
focused discipline.
Speaker 1 (01:11):
And the core strategy we're really diving into today is
what he calls his Ai Barista, this green dot system.
He unveiled it back at the Dreamforce conference October sixteenth,
twenty twenty five.
Speaker 2 (01:22):
Right, So our mission here is to really understand this
tech because Ai Barista, I mean, it makes you think
of robots making your lante.
Speaker 1 (01:30):
Tually, like the Jetson's or something.
Speaker 2 (01:31):
Yeah, exactly. But what this deep dive shows is something, well,
something far more subtle and maybe frankly, far more impactful
in the near term.
Speaker 1 (01:40):
So the big question is, is this green dot tech
the first step towards, you know, automating the human barista
right out of a job, or is.
Speaker 2 (01:48):
It a smarter, more sophisticated evolution. Yeah, something that actually
solves the real pain points, the long lines, the messed
up drinks, and maybe allows that human element to actually
come back to reclaim the sort of artistry of making coffe.
Speaker 1 (02:00):
That's the core tension we're tracking then, automation versus augmentation.
So let's start with the context. We need to get
a handle on just how serious the situation was that
Nicole walked into. He took over September ninth, twenty twenty four,
replacing Laxman Nearasimhan.
Speaker 2 (02:15):
And that's seven percent dip in same store sales. I
mean for a company the size of Starbucks, that's catastrophic.
It signals a massive, massive operational failure somewhere.
Speaker 1 (02:25):
And it wasn't just numbers on a spreadsheet, right, it
was the experience exactly.
Speaker 2 (02:29):
It was experiential. You the customer, were dealing with these
just frustratingly long lines. Yeah, but maybe even more damaging
inconsistent drinks.
Speaker 1 (02:38):
Ah the worst. You pay what seven bucks for a complicated.
Speaker 2 (02:41):
Drink and it's different every single time. The menu complexity
just completely outstripped their ability to deliver consistently that ritual,
that reliable morning coffee. It just dies.
Speaker 1 (02:52):
And that whole third place vibe that Howard Schultz built
the empire on, yeah.
Speaker 2 (02:56):
Just crumbled under the chaos. Hard to feel cozy when
it's frantic and your drink is wrong.
Speaker 1 (03:00):
So Nicole comes in with this very public mission. He
calls it the back to Starbucks crusade, and he's not
being shy about AI is he not at all?
Speaker 2 (03:08):
He told Yahoo Finance quote, He's all in on AI
and the goal pretty ambitious, becoming the world's greatest customer
service company. Again.
Speaker 1 (03:19):
Wow, Okay, so that's basically declaring war on inefficiency.
Speaker 2 (03:22):
It absolutely is, and it's really vital we clarify this
up front. The AI beuris isn't a robot arm.
Speaker 1 (03:28):
Pouring milk right, not yet anyway, not yet.
Speaker 2 (03:30):
It's a real time digital assistant. Think of it like
an advanced layer of intelligence designed to guide and optimize
what the human beriast is doing, making sure every step,
you know, grinding the beans, steaming the milk, is perfect consistent.
Speaker 1 (03:43):
And to really get the scale of this tech bet,
we absolutely have to look at the guy placing the bet.
Brian Nickel, the architect. His playbook is all about resurrection
through data, right.
Speaker 2 (03:53):
His resume is incredibly specific on this point. He's known
for taking these big established brands that have kind of
lost their way, stagnated exactly, and driving them back to profitability.
And he does it largely by focusing on digital efficiency
and crucially simplifying the in store experience.
Speaker 1 (04:09):
His time at Taco Bell over a decade there. That
seems like a precursor. He didn't just like climb the ladder.
Speaker 2 (04:15):
No, he fundamentally shifted how people interacted with Taco Bell.
He pushed the breakfast menus, which was huge for them,
and critically he rolled out app based ordering that became
the standard across the entire fast food industry later on.
Speaker 1 (04:28):
And that app wasn't just about making it easier to
order at Chilupa, was it.
Speaker 2 (04:32):
Not at all? It was about data capture, that whole
digital pivot. It's credited with boosting Taco Bell's sales by
a huge twenty percent. He proved really early on that
you streamline ordering and gather customer data, you unlock serious growth.
Speaker 1 (04:47):
But the real high wire act, the thing that probably
got in the Starbucks job, was turning around Chipotle.
Speaker 2 (04:52):
Oh absolutely, that was his masterpiece. Arguably. He took over
his CEO in twenty eighteen. And remember Chipotle was reeling
major Ecali health scares. The brand was just stagnant. People
were scared to eat.
Speaker 1 (05:03):
There, and the results under his watch phenomenal, just phenomenal.
Speaker 2 (05:07):
Under Nickel, Chipotle essentially doubled its revenue, doubled to nine
point eight billion dollars by twenty twenty four. He didn't
just stabilize it. He injected massive growth, mainly through modernization.
Speaker 1 (05:18):
And we know from the sources that AI was really
the backbone of that Chipotle turnaround, digital kiosks, loyalty programs,
and that really advanced AI for inventory forecasting.
Speaker 2 (05:29):
Yes, that inventory AI at Chipotle is key. It wasn't
just some fancy tech gimmick. It was a fundamental cost saver.
It cut food waste by fifteen percent.
Speaker 1 (05:39):
Fifteen percent with fresh ingredients like they use. That must
be huge.
Speaker 2 (05:42):
It's enormous. A fifteen percent reduction in spoilage. They go
straight to the bottom line hundreds of millions of dollars.
It just confirms Nichol's pattern. AI isn't hype for him.
It's a serious, quantifiable operational tool.
Speaker 1 (05:54):
Okay. So this history, this sort of Midas Touch reputation,
it explains the frankly staggering commitment the Starbucks board made
to get him.
Speaker 2 (06:02):
It really speaks volumes about how desperate they were operationally speaking.
Nicholas compensation package in twenty twenty four it was nearly
ninety seven point eight million dollars, Yeah, including a ten
million dollars signing bonus, and then a jaw dropping seventy
five million dollars in performance based equity. A figure like
that just signals absolute faith that he can do the
Chipotle thing all over again, but on an even bigger scale.
Speaker 1 (06:25):
And Wall Street bought it immediately, didn't they. The stock jumped.
Speaker 2 (06:28):
Twenty four point five percent instantly, just on the announcement
he was appointed. Wall Street definite believes in the power
of the data driven executive.
Speaker 1 (06:36):
So he's already running experiments, these starting five programs.
Speaker 2 (06:39):
Yeah, testing innovations internally, but this Green Dot AI this
is the big one, the bold centerpiece of this massive
financial bet, this whole strategic shift.
Speaker 1 (06:49):
Okay, but we have to be impartial here. We need
to talk about the controversies that seem to follow this
efficiency first playbook, especially when it comes to labor, right.
Speaker 2 (06:58):
And that history is important context for what happening at
Starbucks now at Chipotle. His time as CEO included that
really high profile closure of a store in Augusta, Maine,
back in twenty twenty two, and that store was in
the process of unionizing.
Speaker 1 (07:13):
And the NLRB, the National Labor Relations Board, they ruled
on that didn't.
Speaker 2 (07:17):
They they did. They ruled it was illegal union busting,
so that hangs in the background.
Speaker 1 (07:20):
And now he walks straight into Starbucks where labor tensions
are already I mean at a fever pitch. How many
stores have unionized.
Speaker 2 (07:28):
Over four hundred since twenty twenty one. It's a significant movement,
and the source material really highlights the let's say, stark
optics of the friction. You've got reports of Nickel commuting
via private jet from his home in Newport Beach to Seattle.
Speaker 1 (07:44):
While baristas are pushing for fair contracts.
Speaker 2 (07:46):
Exactly pushing for better wages of better conditions. It creates
this immediate, pretty dramatic clash of cultures, priorities, and.
Speaker 1 (07:54):
The union's reaction to the AI push not positive.
Speaker 2 (07:57):
Union spokesperson Jasmine Leli was quick to jump on it.
She basically called the whole AI initiative a distraction, a
way to divert attention from quote declining sales and brand perception.
Speaker 1 (08:08):
Her argument being that fancy tech doesn't matter if the
workers aren't valued.
Speaker 2 (08:11):
Pretty much, that operational efficiency is kind of irrelevant if
the workforce feels exploited or isn't fairly paid. So Nichol's
challenge is not just technical, it's absolutely a cultural battle
for the soul of Starbucks.
Speaker 1 (08:24):
Okay, let's pivot to the technology itself. Then let's look
under the hood. If this is the big solution, what
exactly is the green dot Assist? How does it actually
work when a barista is, you know, slammed during the
morning rush.
Speaker 2 (08:35):
Well, the green Dot Assist it's actually surprisingly simple on
the front end how it looks, but incredibly complex behind
the scenes. It's basically an iPad beamed AI interface and
it just appears, like the name suggests, as a small
green dot on the barista screen.
Speaker 1 (08:52):
Okay, a green dot.
Speaker 2 (08:53):
And then what when things get hectic or maybe they
get a really complex order, the barista just taps the
dot and it summons up contextual, real time help right
there on the screen.
Speaker 1 (09:03):
So it's like having i don't know, an instant expert,
a perfect recipe book and troubleshooter rolled into one right
when you need it.
Speaker 2 (09:09):
Exactly that tailored for that exact moment. Think about it.
Say it's eight thirty am peak rush. A new hire,
maybe it's their first week, gets hit with like the
dreaded venty iced white MoMA.
Speaker 1 (09:21):
Oh yeah, those modifiers right.
Speaker 2 (09:23):
Venty iced white mocha, but with almond milk, extra pumps
of sugar free vanilla, and oh, just a light amount
of camel drizzle. The barista's head is spinning.
Speaker 1 (09:33):
Okay, so they tap the green dot.
Speaker 2 (09:35):
They tap the dot and boom. The screen instantly shows
the precise number of syrup pumps needed for that size
in those mods, the exact volume of almond milk, and
crucially it flags the allergen info because of the almond
milk swap.
Speaker 1 (09:48):
Wow, that is huge, especially for those super customized drinks
that are basically the norm.
Speaker 2 (09:54):
Now.
Speaker 1 (09:54):
It takes the mental load off right, the calculation under pressure,
that's where mistakes happen.
Speaker 2 (09:59):
That's exactly where the happen. It doesn't just do recipes.
It helps troubleshoot equipment too. Say the steam wand pressure
suddenly feels low. Tap the dot. It might suggest exactly
what to check first. It's like having a dynamic, always
available expert shoulder surfing.
Speaker 1 (10:12):
With you and the brain behind This is massive, you said.
It's powered by Microsoft's Azure Open Ai. They announce that
partnership back in June twenty twenty five. What does that
tell us about the data it's using.
Speaker 2 (10:23):
It tells us this system has access to truly enormous
data sets. It's pulling from every single recipe variation. Starbucks
is ever offered every equipment manual for every machine in
the store. And this is key, the historical record of
every single order air the company has.
Speaker 1 (10:42):
Ever logged, every mistake ever made.
Speaker 2 (10:44):
Essentially, yeah, it uses that huge pile of data to
predict where mistakes are most likely to occur in a
given situation, and it tries to step in before they happen.
Speaker 1 (10:53):
So, just to be clear, it's not like chat GPT.
It's not generative AI writing palms about coffee.
Speaker 2 (10:58):
No, No, definitely not. It's a highlight targeted, highly optimized
knowledge retrieval system purely focused on operational consistency and efficiency.
Speaker 1 (11:06):
Got it, and the pilot programs the tests as they ran,
they're already showing some pretty impressive quantifiable results stuff Nickel
can point too.
Speaker 2 (11:13):
Absolutely, the numbers speak directly to his data driven approach.
Speaker 1 (11:16):
Okay, let's look at those numbers. The training time reduction
seems like a big one.
Speaker 2 (11:20):
It is green dot cut the time it takes to
train a new hire from like weeks down to just.
Speaker 1 (11:25):
Days weeks to days.
Speaker 2 (11:27):
Yeah, think about the impact of that. Instead of spending
maybe two weeks just trying to memorize hundreds of recipes
and troubleshooting steps, the new Beerrista has this instant digital
cheat sheet, gets some up to speed productively making drinks
way faster.
Speaker 1 (11:42):
And the financial impact the waste production that sounds enormous too.
Speaker 2 (11:46):
It cut drink remake rates by twelve percent in the
test stores. Now you factor in the cost of wasted ingredients,
the barista's time remaking it, the customers lost time. The
company estimates that errors and waste costs roughly fifty thousand
dollars a year per store.
Speaker 1 (12:02):
Fifty thousand per store. Wow.
Speaker 2 (12:05):
Yeah, so a twelve percent cut in that figure across
thousands of stores nationwide. That is a massive financial win.
And it's better customer service right, you get your drink
right the first time. It's a direct margin improvement.
Speaker 1 (12:15):
And the rollout is moving fast. It's not just a
test anymore.
Speaker 2 (12:18):
No, this is happening. It's rolling out across North America. Now.
Over two thousand US stores had it by October twenty
twenty five, and the plan is a global expansion in
twenty twenty six. This has basically become the core operating
system for the stores.
Speaker 1 (12:31):
And Nicol sees it going beyond just making drinks correctly.
Predictive maintenance.
Speaker 2 (12:35):
Yeah, that's his longer term vision. He wants the system
to evolve so it can analyze subtle signs from the
equipment itself. Like maybe it monitors the specific vibration pattern
of a coffee grinder over time.
Speaker 1 (12:46):
And flags it before it breaks down.
Speaker 2 (12:48):
Exactly, alert the manager, Hey, grinder three's vibration signatures, looking
if he might need service soon before he just completely
fails midshift.
Speaker 1 (12:56):
Preventing that downtime is huge. If the espresso A machine
goes down or the main grinder fails, the store is
instantly crippled, isn't it.
Speaker 2 (13:06):
It happens more often than you'd think. The sources say
equipment failure downtime affects about ten percent of shifts. So
if the AI can predict that a steam waand needs
preemptive service based on its past performance data, it avoids
that failure. It maintains that consistency the customer expects.
Speaker 1 (13:22):
And there are other AI tools already in play too, right,
like the inventory AI you mentioned from Chipotle.
Speaker 2 (13:27):
Right, that's a direct import from his Chipotle playbook. It's
already in five thousand stores. It uses computer vision basically
Cameron's mounted in the stockroom to automatically count inventory.
Speaker 1 (13:36):
So no more manual counting with clipboards.
Speaker 2 (13:39):
Pretty much. The AI literally recognizes the shapes and volumes
of syrup bottles, milk cartons, cup stacks. It calculates inventory
levels in real time without a human needing to scan
every barcode.
Speaker 1 (13:52):
That's seriously sophisticated. What's the impact. What does that actually
do for the store.
Speaker 2 (13:56):
Well, the big claim is it cuts the physical time
baristas and manager spend back in the stock room counting
stuff by thirty percent.
Speaker 1 (14:03):
Thirty percent less time counting boxes.
Speaker 2 (14:05):
Yeah, and that time theoretically gets redeployed out front, making drinks,
maybe talking to customers. And crucially, it helps ensure the
staples you know, like the infamous pumpkin spice, are always
ordered on time and kept in stock. Prevents those frustrating
sorry we're out moments.
Speaker 1 (14:22):
Okay, but this level of automation, this minute tracking of inventory,
even the help from green Dot, it immediately brings up
the other side of the coin surveillance, Yeah, employee monitoring.
We have to talk about neurospot.
Speaker 2 (14:33):
Yeah, this is where that tension between efficiency and the
human element really gets sharp.
Speaker 1 (14:38):
Ye.
Speaker 2 (14:38):
Back in June twenty twenty five, there's this viral post
on x the platform formerly known as Twitter. It highlighted
this existing AI system Starbucks was using called Neurospot, and
Neurospot apparently tracks barista productivity and something they called dwell
times twelve times.
Speaker 1 (14:54):
That sounds orwellian, like how long you stand in one place?
Speaker 2 (14:57):
Exactly how long an employee spends in one spot? Is
it too long? Talking to a customer? Too long? Resocking?
It immediately sparked these big brother fears. People worried that
green dot, even if helpful, could be weaponized use for scrutiny,
not support.
Speaker 1 (15:12):
So what's Starbucks's official line on this? How do they
separate green dot from Neurospot.
Speaker 2 (15:17):
Their official response is that green dot is opt in,
meaning the barista chooses to tap the dot for help,
and that the data collected is anonymized. They insist it's
focused purely on system support overall performance improvements, not on
tracking or punishing individual baristas. The stated goal is support.
Speaker 1 (15:34):
Not scrutiny, but the shadow of neurospot is still there.
It leads straight to that question from the ex user
at Rainmaker in nineteen seventy three, efficiency or exploitation. Where's
the line?
Speaker 2 (15:44):
It's the tightrope nick All has to walk, isn't it
net The data is being collected, no doubt about that.
Whether it's ultimately used to guide the barista or to
judge the barista, that defines whether this whole strategy gets
embraced or rejected by the actual workforce on the ground.
Speaker 1 (15:59):
Okay, this this brings us squarely to the human equation.
If the tech is driving this relentless efficiency, how is
Starbucks convincing it's what two hundred and forty thousand baristas
globally that this AI is their friend, not their replacement.
Speaker 2 (16:14):
Well, Nicole's public messaging is relentlessly focused on augmentation, not replacement.
He keeps saying he wants more real baristas back in stores,
more baristas, not fere That's the claim, and it's significant
because it lines up with this other strategic shift they made,
pulling back from those super efficient but maybe soulless mobile
only store formats they were testing. They actually halted those
(16:35):
in July twenty twenty five, so they're consciously saying they
want to pivot back towards that human centric third place idea.
Speaker 1 (16:43):
So the pitches let the AI handle the annoying complex stuff,
the drudgery, as the sources put.
Speaker 2 (16:49):
It right, the remembering of obscure recipes, the troubleshooting.
Speaker 1 (16:52):
And that frees up the human barista to actually focus
on the customer interaction, the craft, the welcoming vibe.
Speaker 2 (17:00):
That is absolutely the promise, and look at the potential
benefits for the workforce if it works as advertised. Starbucks
has this notoriously high annual turnover eate. It's like one
hundred and fifty percent.
Speaker 1 (17:11):
Wow, one hundred and fifty percent. That means they replace
basically their entire staff every eight months or so pretty much.
Speaker 2 (17:17):
It's incredibly high. So green Dot could offer new hires
and remember many are teenagers, juggling school gives them instant expertise.
It reduces that initial feeling of being completely overwhelmed, which
is often what leads to burnout and people quitting quickly.
Speaker 1 (17:31):
And stabilizing the workforce, reducing that churn. That's crucial for
the customer experience. Too, isn't it high turnover means you
rarely see a familiar face, kills that neighborhood third place
feeling exactly.
Speaker 2 (17:41):
So if green doc can actually reduce that one hundred
and fifty percent turnover by making the job less stressful,
more manageable, it fosters a more experienced, stable team, and
it helps veterans too. There's this anonymous brist to post
on X that really summed it up. They said something like,
finally tech that doesn't make us feel stupid for forgetting.
Speaker 1 (17:59):
A pump count ah. Interesting, So it's like error proofing
for everyone, right.
Speaker 2 (18:03):
It allows the experienced priests to stop worrying about tiny
details and concentrate on real personalization like remembering your regular
word extra hot, no foam, oat milk, that kind of thing,
the stuff that makes you feel recognized.
Speaker 1 (18:16):
Okay, that's the potential upside for the employees. But the
union opposition is still strong. They seem to view this
whole AI push through a very different lens, one of
distraction and ongoing conflict.
Speaker 2 (18:28):
They are deeply skeptical of the motives. Absolutely. Starbucks Workers
United even put out this satirical video around October seventeenth,
twenty twenty five, basically portraying Nichol as some kind of
robot overlord. It just highlights the profound mistrust that exists.
Speaker 1 (18:42):
And their spokesperson, Jasmine Leylai, her main argument is that
this tech focus is just a smoke screen to hide
the real issue.
Speaker 2 (18:51):
Yes, her argument is that it's a calculated move to
distract everyone from the core issue of wage stagnation.
Speaker 1 (18:56):
Okay, so that's the heart of the labor fight. Then, wages,
that's the crux of it.
Speaker 2 (19:00):
Baristas are currently earning somewhere between fifteen and eighteen dollars
an hour typically, and this is while facing what's seven
percent inflation. The unions are demanding significantly more that, pushing
for twenty dollars plus per hour minimum and some form
of profit sharing.
Speaker 1 (19:14):
Their argument being if the company can afford nearly one
hundred million dollars for the CEO and millions more for
AI systems, they.
Speaker 2 (19:21):
Can afford a living wage for the people actually making
and serving the coffee. That's the core contention.
Speaker 1 (19:26):
So how is management responding to that specific wage critique.
Are they saying the AI savings will pay for raises or.
Speaker 2 (19:34):
Nickel's trying to frame it differently. He's trying to show
that the efficiencies gained through AI are actually enabling growth
and allowing for higher wages, rather than being about cutting
staff to save costs. He announced plans to hire ten
thousand more baristas by twenty twenty six, ten thousand more yeah,
claiming these new jobs are essentially funded by the AI
efficiencies making the stores more productive. And importantly, he's planning
(19:56):
to raise the minimum wage floor to eighteen dollars an
hour across the board. Eighteen dollars eighteen, So it's a
strategic move right. It directly counters the union narrative that
AI automatically means job losses, but it still falls short
of that twenty dollars plus demand the union is making.
Speaker 1 (20:12):
And that ethical tightrope round monitoring it doesn't go away.
Even if green dot is opt in. If other systems
like neurospot are tracking dwell times or productivity, how do
you guarantee that data isn't used, maybe even unintentionally, to
pressure buristas or penalize them for taking that extra moment
to chat with the customer, the very thing Nichol says
he wants more of.
Speaker 2 (20:32):
That is the absolute key question for whether this whole
thing works long term. Starbucks says they use diverse data sets,
human oversight, all sorts of things to try and mitigate
bias risk in the AI. But let's be real, as
long as performance is being measured somehow, even if it's
aggregated or anonymized, the system creates a metric, and metrics
inevitably get used to justify decisions.
Speaker 1 (20:54):
Right, what gets mattered gets managed.
Speaker 2 (20:56):
Exactly, So the success of this entire turnaround it really
rests on whether the barista community, the actual people on
the floor yeah, truly come to believe that green Dot
is an ally helping them do their job better, or
if it feels like a sophisticated warden constantly monitoring their
every move.
Speaker 1 (21:13):
Okay, let's shift focus from the internal friction. Now, let's
talk about the customer. Why should the what is it
one hundred million people who visit Starbucks every week actually
care about this AI revolution behind the counter? How does
it change my experience?
Speaker 2 (21:26):
Well, it starts with making ordering smoother, more frictionless, and
this is where Nicol dropped what the source is called
his boldest tease, an app that actually predicts.
Speaker 1 (21:36):
Your order, predicts my order, Yes, Starbucks.
Speaker 2 (21:38):
Already has what thirty million active users in its loyalty app.
That's a ton of data. This new system uses advanced
natural language processing NLP to understand your past orders, the
time of day, maybe even the weather.
Speaker 1 (21:50):
Okay, so the app opens up and it might already
suggest my usual VENTI iced coffee to pumps vanilla splash
a cream exactly.
Speaker 2 (21:57):
It anticipates what you likely want. It moves that little
bit of friction, that decision fatigue makes the whole process
feel seamless.
Speaker 1 (22:04):
And that seamlessness they expect that to drive loyalty significantly.
Speaker 2 (22:08):
They're predicting it could boost loyalty program uptake by another
twenty percent. The whole digital overhaul is about shrinking the
gap between you thinking I want coffee and you actually
holding the cup. And to make this happen, they're rolling
out updated point of sale systems, you know, the screens
and registers the barist has used to all forty thousand
stores globally by the second quarter of twenty twenty six,
(22:30):
forty thousand stores.
Speaker 1 (22:31):
Wow. And these new POS systems they also use AI
for dynamic pricing and upsells. Is that right?
Speaker 2 (22:38):
Yes? And this is where the personalization gets really sophisticated.
The AI looks at what's in your basket, what you're
buying right now, plus your past history, and it generates
these really relevant, hopefully subtle up cells.
Speaker 1 (22:50):
So instead of just asking everyone want a pastry.
Speaker 2 (22:52):
With that right, it might analyze your history and see
you often by baked goods on Tuesday mornings, So the
screen might the barista or maybe even the app suggests
add a blueberry scone for just one dollar. It's tailored,
designed to gently increase that average ticket size without feeling
a knowing or generic.
Speaker 1 (23:11):
Okay, it makes sense now. Shifting to speed. You mentioned
the drive through is the big bottleneck, right, It accounts
for half of all US sales. What's the AI doing
to tackle those long lines.
Speaker 2 (23:23):
They're implementing something called smart Q. Think of it like
an intelligent air traffic control system, but for coffee orders.
Speaker 1 (23:29):
Smart Q Okay, how does that work?
Speaker 2 (23:31):
Instead of just a simple first order in first order
made system, smart Q dynamically routs orders. If a really
complex freppuccino comes in, it might rout it to a
station that's less busy or better equipped for blending simpler
orders like a black coffee get routed to the absolute
fastest path. It maximizes the throughput of the whole system.
Speaker 1 (23:50):
So it's not just about individual barista speed, it's about
managing the flow of all the orders together.
Speaker 2 (23:56):
Precisely, it's traffic management for drinks, and the goal is
to minimize the time cars spend just sitting there idling
in the drive through lane, which is obviously better for
customer satisfaction, but it also helps with their environmental goals
cutting down on vehicle emissions, especially in busy urban areas.
Speaker 1 (24:14):
And the scale of this personalization using all that transaction data,
it's kind of mind boggling. How many transactions did you
say they analyzed daily?
Speaker 2 (24:24):
Seventy five million daily. That's the fire hose of data.
The AI is constantly learning from that fine five million,
and that scale allows for incredibly specific tailored offers. Like
the example given was, maybe the AI notices you only
ever buy a nitro cold brew when the weather forecast
predicts rain.
Speaker 1 (24:40):
Okay, random but plausible.
Speaker 2 (24:42):
Right, So on the next rainy day, maybe the app
offers you a free upgrade to a larger sized Nitro.
It's that granular level of personalization they're aiming for and.
Speaker 1 (24:50):
Are they seeing results from this kind of thing.
Speaker 2 (24:53):
They are in the pilot locations where they tested these
personalized offers. Customer satisfaction score is apparently lifted by fifteen points.
That's significant, and this is really the ultimate goal, the
conversions point Nicol is shooting for. The AI handles all
the complex data, the logistics, the predictions.
Speaker 1 (25:09):
Which frees up the human barista exactly.
Speaker 2 (25:11):
To deliver that moment of connection, that quick chat how's
your week going, The stuff that actually defined the third
place idea in the first place, the warmth.
Speaker 1 (25:20):
Okay, But analyzing seventy five million daily transactions, predicting orders,
tracking preferences, that immediately throws up huge red flags around privacy,
especially with regulations like GDPR in Europe and CCPA in California.
Speaker 2 (25:35):
Absolutely, privacy and trust are non negotiable here. Starbucks is
stressing that they're taking steps to anonymize the vast majority
of this data. They have to given the scrutiny under
those global frameworks. Nicol even said a dreamforce directly, trust
is our brew. Yeah, but they're constantly fighting the perception
and the reality of past security incidents like that breach
(25:57):
back in twenty twenty two, that affected about one thousand users.
They need robust, easy to use opt out mechanisms and
total transparency about how data is used and secured. If
they lose customer trust on this, the whole personalization engine collapses.
Speaker 1 (26:12):
Okay, so we've spent a lot of time on Starbucks's
specific situation and their AI solution. Let's zoom out a
bit now put their moves into the broader context, the
coffee cosmos, as you call it. Is this green dot
system really that revolutionary or are competitors doing similar things?
Speaker 2 (26:28):
That's a great question. Starbus is definitely making the biggest,
most integrated, and most high profile push right now, but
AI is absolutely percolating part in the pun across the
entire hospitality sector. Competitors are definitely using similar types of tools,
even if they're not as centralized or branded as Green Dot.
Speaker 1 (26:44):
Okay, like, who what are others doing well?
Speaker 2 (26:46):
Duncan, for instance, they're reportedly using AI mainly for demand forecasting,
trying to predict how busy a store will be based
on whether local events, time.
Speaker 1 (26:55):
Of day, to optimize staffing levels exactly.
Speaker 2 (26:57):
They claim it's helped reduce overstaff costs by about ten percent.
Then you have Dutch Bros. They're apparently piloting voice AI
systems in their drive thrusts, trying to automate the basic
order taking part to speed things up.
Speaker 1 (27:10):
And even smaller players high end places.
Speaker 2 (27:13):
Yeah, even some independence like Bluebottle or using simpler AI
chat pots, but maybe just for managing reservations or answering
basic questions online. It's becoming table stakes at various levels.
Speaker 1 (27:23):
And then there's the really far end of the spectrum,
full automation, like that Cafe x place in San Francisco,
the robot barista, right.
Speaker 2 (27:30):
That's the alternative model. Cafe x has had a robotic
arm making drinks since twenty seventeen, and it's efficient, makes
about one hundred and twenty drinks an hour, but it's expensive.
That fifty thousand dollars per unit cost you mentioned is
a huge barrier for widespread adoption. It keeps it pretty niche.
Speaker 1 (27:45):
So Starbucks's strategy is fundamentally different.
Speaker 2 (27:48):
It really is. It's not about replacing the human with
a robot. It's about using AI's efficiency to empower the
human barista, trying to create this symbatic relationship. Augmentation, not
full automation.
Speaker 1 (27:58):
That's the bet aiming to be the world's greatest customer
service company means scaling this globally across eighty countries that
must present enormous logistical and regulatory challenges for green Dot.
Speaker 2 (28:11):
Oh, the hurdles are immense Logistically, just getting the hardware,
the software, updates, the training rolldout consistently across such a
diverse network is a nightmare.
Speaker 1 (28:20):
And the regulations.
Speaker 2 (28:22):
The regulations are a minefield, especially in places like the
European Union with their EU AI Act. That act classifies
AI systems based on their potential risk. A system like
green Dot, because it tracks performance, interacts with employees, potentially
influences things like scheduling or even job reviews.
Speaker 1 (28:40):
Down the line, it could be deemed high risk.
Speaker 2 (28:42):
It's possible, and if it is, it triggers much stricter
requirements around transparency, data governance, accuracy, human oversight. The compliance
burden becomes way heavier than just rolling out a new
app feature in the US.
Speaker 1 (28:53):
That definitely complicates a global.
Speaker 2 (28:55):
Rollout absolutely, And then you have the simple realities on
the ground. Think about rural stores, maybe in less developed areas,
do they even have the reliable high speed internet bandwidth
needed to run this kind of cloud based real time
AI effectively maybe not?
Speaker 1 (29:12):
And cultural differences. Does the AI understand that a Macha
latte in Japan is totally different from one in Peoria?
Speaker 2 (29:17):
Exactly? They can't just translate the interface. They need truly
localized AI models that understand regional tastes, ingredient availability, complex
local preferences. It's not a one size fits all solution.
Speaker 1 (29:29):
So all this complexity, the costs, the potential pitfalls, it
naturally breeds skepticism, doesn't it. Can Nickel actually pull this off?
Speaker 2 (29:36):
The skepticism is definitely out there, and it's loud. You
saw that piece on AOL back in October, basically asking
if this whole AI push is just mere hype, wondering
if Starbucks is just slapping an AI powered label on
everything to create buzz, suggesting maybe we're in a bit
of an AI bubble generally.
Speaker 1 (29:52):
And the fear of job losses doesn't go away either.
Speaker 2 (29:54):
Not at all. You see it all over social media
accounts like at zero hedge on x just quipping next robots.
It reflects that underlying anxiety about automation about where this
all leads for workers.
Speaker 1 (30:07):
But Nicol he has the counter argument, right, he has
the data from the pilot stores.
Speaker 2 (30:11):
That's his strongest weapon in this perception battle. Those early
tangible results that Q three twenty twenty five data showing
copparable sales. We're up two percent in the store's testing
green dot. That's his proof of concept, proof.
Speaker 1 (30:24):
That it's not just TYU but actually moves the.
Speaker 2 (30:26):
Needle right, proof that customers are responding positively to the
improved consistency, the faster service, and that data fuels the
other side of the ex conversation. You get users like
a Junior Stock celebrating it, hailing it as data driven delight.
It really is a battle of narratives being fought with
performance metrics.
Speaker 1 (30:42):
Let's talk finally about the cost, because this isn't cheap.
The financial investment and also the environmental cost.
Speaker 2 (30:49):
The financial commitment is staggering. Nichol isn't just tweaking software here.
He's committed to a massive physical transformation too. They're aiming
to remodel one thousand stores by twenty twenty six.
Speaker 1 (31:00):
One thousand stores, and each remodel.
Speaker 2 (31:02):
Costs around half a million dollars each.
Speaker 1 (31:05):
Wow, so that's half a billion dollars just on refreshing
the physical stores yep, and.
Speaker 2 (31:10):
The entire return on that massive investment. It hinges completely
on whether these AI driven improvements, the speed, the accuracy,
the personalization actually drive enough new customer traffic and get
people to spend slightly more each visit to justify that
colossal expense, huge gamble.
Speaker 1 (31:28):
And the environmental footprint. Running AI at this scale, analyzing
seventy five million transactions daily, predictive maintenance checks, inventory scans,
it takes a lot of server power, right, A lot
of energy.
Speaker 2 (31:39):
It does, and that's a key part of their public commitment.
Starbucks knows this. They've pledged to offset that increased energy
usage through investments in renewable energy. They're sticking to their
ambitious target of being carbon neutral by twenty thirty. So
they have to constantly balance this push for high tech
cloud based efficiency with their very public, very scrutinized environmental promises.
Speaker 1 (31:59):
Okay, so as we wrap up this deep dive, let's
bring it back to that core tension we identified at
the start, the balancing act that defines this whole turnaround effort.
Speaker 2 (32:08):
Right, Starbucks is essentially fighting an operational war on two fronts.
On one hand, Nickel is deploying his proven playbook, data
driven efficiency embodied by the green Dot system, to fix
years of inconsistency and customer frustration, fixing the mechanics, but
on the other hand, he's simultaneously trying to reclaim that softer,
(32:29):
human centric artismal third place feeling, the vibe that originally
made Starbucks well Starbucks.
Speaker 1 (32:35):
So the success of this whole thing, the green Dot,
the AI strategy, it really hinges on finding that perfect
equilibrium between the tech optimization and the human touch.
Speaker 2 (32:44):
It absolutely does. If the system tips too far one way,
if it feels too much like rigid scrutiny or cold efficiency,
if it kills employee morale or makes the stores feel sterile,
then the entire turnaround could fail, regardless of how technically
brilliant the AI is.
Speaker 1 (32:58):
Because of that ambitious goal they have hitting one hundred
billion dollars in revenue by twenty thirty.
Speaker 2 (33:03):
That relies completely on these AI gains translating into both
genuine speed and genuine customer satisfaction. The tech has to
truly serve the people, involve both the breezes making the
coffee and you, the customer drinking it. It can't just
supplant them or treat them like you know, predictable variables
in an algorithm.
Speaker 1 (33:23):
Which leaves us and you with a final, maybe provocative
question to think about. It builds on the strange duality,
doesn't it? Efficiency and humanity? If AI is now handling
almost every logistical piece of getting your coffee, checking the stock,
predicting your order using NLP based on millions of data points,
telling the breis exactly how many pumps, troubleshooting the machine,
(33:44):
how far can those buristas really push the idea of
personalization and connection before that human element itself starts to
feel well, like just another optimized feature, precalculated, prompted and
scheduled into your experience by that same powerful algorithm.
Speaker 2 (33:59):
Yeah, when the entire process is so perfectly optimized by data,
what's left that is truly uniquely human about grabbing a
cup of coffee?
Speaker 1 (34:07):
Something to definitely mull over next time you tap your
app for that perfectly calculated, perfectly consistent cup. Thanks for
joining us for the deep dive
Popular Podcasts
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Las Culturistas with Matt Rogers and Bowen Yang
Ding dong! Join your culture consultants, Matt Rogers and Bowen Yang, on an unforgettable journey into the beating heart of CULTURE. Alongside sizzling special guests, they GET INTO the hottest pop-culture moments of the day and the formative cultural experiences that turned them into Culturistas. Produced by the Big Money Players Network and iHeartRadio.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com