Play
Show Navigation
Show Transcript
00:00
Play
Seek 10 seconds backwards
Seek 10 seconds forward
00:00 / 01:14:10
Mute
Disable Captions
Settings
Picture in picture
Fullscreen
Search in video
You can search the video captions for specific words or phrases.
Popout transcript
Hide Transcript
Skip transcript
[ROCK MUSIC]
Salesforce.
For 25 years,
we've been a force
for sales, a
force for service,
a force for marketing, for
commerce, for analytics.
Well, starting now, we're
also your agent force.
(SINGING) Super
agent force.
Super agent force.
A force for humans
with AI agents
to drive customer
success together.
Agentforce is not about
AI replacing people.
It's about how do
we enhance employee
experience to deliver
better customer
experience?
We're not just talking
about a business
evolution.
We're talking about
an agent revolution,
agents with the power
to act on behalf of you
and your business,
agents with the context
to make customer
relationships more
personal, agents
with the ability
to do it all
intelligently, accurately,
and autonomously.
Because when you've got
trusted AI agents at work,
your business can
succeed at scale.
Agentforce gives
us the ability
to deliver that
high-touch, personalized
service to everybody.
You can improve
the communication
between supply and
demand, between patients
and doctors, employees
and employers, clients
and customers.
You see, this isn't
another basic bot.
No, this is a bona
fide Agentforce A team.
This is using
data to reach
every possible
customer with exactly
the right message, then
taking immediate action
so they have exactly
the right experience.
This is augmenting
and elevating
humans to new heights
while always staying
grounded in trust.
It will help humans
become more human.
Welcome to a force of
meaningful and effective
intelligence.
Welcome to a
force for purpose,
for trust, and for good.
Welcome to Agentforce, to
what AI was meant to be.
Please welcome
Co-founder Salesforce
and Chief
Technology Officer
Slack, Parker Harris.
[CHEERING AND APPLAUSE]
All right.
Well, it's so good
to be back here
in New York City.
And I'm so excited to
see people still pouring
into the room.
So for those of you who
have seats, good job.
You did a good job
getting here early.
Well, we have an
incredible show today.
And what I'm so excited to
talk about is Agentforce.
And I think
you've maybe heard
a little bit about it.
If you haven't, you're
going to hear a lot more
about it today.
It's really probably the
biggest innovation that
I've seen at Salesforce in
our 25-plus years building
technology.
But as we always
do at Salesforce,
before we do anything--
before we do anything--
we want to thank you.
We want to thank you.
There we go.
We want to thank each
and every one of you
because we couldn't
do it without you.
We have some
employees in the room.
So yes, I want to thank
my fellow employees that
are building
the technology,
selling, supporting you.
But more
importantly, I want
to thank you, the
customers that are here,
the MVPs.
Do we have some MVPs?
I think we do.
Yeah, we normally do.
[APPLAUSE]
Wow.
We got to get the energy
up in the room, MVPs.
What happened?
The nonprofits, the
partners, the ISVs,
all of you together--
it's our community--
have built something
pretty incredible.
And I just want
to thank you.
So what are we
doing today?
Well, we've got
a big keynote.
We've got some customer
stories from Saks,
from Prudential, from
Unity, Fisher and Paykel.
We've got 150 sessions.
You need to go to every
single one of them.
Maybe you have
some agents you
could send to some
of those sessions
and they can report back.
We've got over 30 demos.
We've got a campground.
It's very-- it's
overwhelming.
But you know what's cool
is we've got an app.
We've got our events app.
Download the events app
if you haven't done it.
And you know what's
inside there?
Agentforce.
Agentforce will help you
personalize your journey
here at World
Tour New York.
That's the magic of
agentic innovation.
And I want you to
try it right now.
And finally, I want to
thank our sponsors--
AWS, Deloitte
Digital, and Writer
for sponsoring
this conference.
We really appreciate
your support.
Well, over 26 years
ago, I met Marc Benioff.
And when we met, we
talked about starting
this incredible company.
And when we
talked about it,
the first thing
we said is we
need to found the company
on a set of values.
And I'm so thankful
that we have always
followed our values,
especially right now
in this world of AI.
Trust is our
number one value.
And I think you're all
thinking like, oh, my god.
This AI world
is incredible,
but I'm worried about it.
Is my data going to get
sucked up into some LLM
and given to my
competitor, god forbid?
No.
Not with Salesforce,
because it's
trusted with Salesforce.
Because we want to use
that layer of trust
to then give you
the innovation
you need, to give
you Agentforce,
so that you can be
successful in transforming
how you connect
with your customers
through Agentforce.
So it's really,
really cool.
And of course,
we're always
working for a more
equitable and more
sustainable future for
everyone, something
that we believe
very deeply in.
So based upon
these core values,
we also realized, when
we started the company,
that we could also be
a platform for change.
It's not just
about innovation,
even though that's what
I absolutely adore,
is innovating.
It's not just
about innovation.
It's also about being
a platform for change.
And we're doing it
here in New York City.
We're doing it back in San
Francisco, where I'm from.
In Paris, in
Tokyo-- everywhere
we have employees,
we're driving change.
It's called the 111 model.
We started it when we
started the company.
There was no money,
equity was worth nothing.
Three people in an
apartment, and Marc
was the CEO.
He wasn't even an
employee at first.
It was kind of funny.
But look at what
we've done since then.
$751 million in
all-time giving,
9.5 million
volunteer hours.
59,700 nonprofits around
the world are using
Salesforce to do good.
Yeah.
Do we have any
nonprofits in the room?
[APPLAUSE]
We must have some.
Can you raise your
hand or stand up
if you're a nonprofit?
Yes, thank you right here.
And when you see them,
you know what they need.
They need your help.
Don't expect Salesforce
to be the only one
being a platform
for change.
Each and every
one of you can
be a platform for
change in your business,
in your line of work, and
in your day-to-day life.
You can actually do good.
And it will
actually change
your life and
your company,
just as it's changed mine
and changed Salesforce.
And through those values
and through that platform
of driving change
in the world,
we've become an
incredible company.
I'm very, very
proud of that.
Number one CRM--
that's pretty cool.
Number two,
enterprise software--
not cool enough,
but not bad.
We're going to
be number one.
Don't worry.
We're going to
be number one.
We're on our way.
We're on our way.
Most innovative
company-- so we're
valued for our innovation.
We're valued for our
philanthropy-- top
100 companies that care.
Ethics, part of
our value system--
world's most
ethical companies.
And we're so excited that
we're on our way this year
guided to $38 billion,
$38 billion in revenue.
That's really incredible,
really incredible.
Now it's been a
journey with AI.
So I want to now seed
this conversation more
into what's happening
in the world of AI.
Salesforce
didn't just start
doing AI when OpenAI
hit the market
and everyone was like,
whoa, what happened?
We have been doing it
for a very long time.
We did it
starting in 2014.
We started our
research group
that does research in
artificial intelligence.
Some of that
research actually
led to some of
the innovations
you're seeing
today in market.
But back then, it was
about machine learning,
doing predictive AI, which
many of you probably use.
It's all over
our products.
You do case routing and
send time optimization
and marketing.
That's predictive
AI, machine learning.
It is still
incredibly valuable.
It's still
incredibly valuable.
But then we went from
wave one to wave two.
Wave two was all
about copilots.
These copilots were
sitting beside you.
And you're
saying, well, hey,
can it help me write my
email a little bit better?
I want to have some
empathy for that customer
when I'm responding to
them on that support case.
So that was helpful.
That was generative.
And it was good.
And we were
excited about it.
And AI on our
platform, we're
delivering over 2.2
trillion results per week.
That's a lot of AI
happening on our platform.
So you know
that we know AI.
You know that you're
doing AI with us.
You have been
for a long time.
But we're moving
into another phase,
another wave.
We were in
predictive, as I said.
Then we were in copilot.
But the new wave
is agentic AI.
And we are succeeding.
We are seeing success
right now with customers.
I was just with
RBC yesterday
at a conference here
in New York City.
And they were talking
about how quickly they
were able to be successful
with AI, with Agentforce,
that it makes every
advisor at RBC an expert.
Pretty cool.
Wiley, the bookseller--
their peak season
just happened
when everybody's
going back to school.
And people need to
get these textbooks.
And they get inundated
with service requests.
And they-- how
do we scale this?
Agentforce.
And so these are
customer stories.
You're going to hear
more about them today.
But we're already
seeing these incredible
successful
customers who are
taking Agentforce and
amplifying their business.
Now we're going to keep
going in future waves.
I'm not going to talk
much more about them.
In San Francisco, I
finally bit the bullet
and I got in one of those
automated cars, a Waymo.
It's a little bit scary if
you know computer science
because I'm like, is
this thing going to work?
But it was incredible.
It's the future
of robotics.
This car just pulled up.
There was no
one driving it.
The steering
wheel is moving.
And it took me right--
And after about five
minutes, I was like,
this is kind of boring.
It's just kind of normal.
That's the world
we're headed into.
And then many
believe we're
going to hit AGI, or
artificial general
intelligence.
Maybe that'll happen soon.
Maybe it'll take a while.
But it doesn't
matter, because the AI
that we have today
can transform
you and your business.
Now for 26 years, our
number one priority
has been to make
you successful
and to help you connect
with your customers
in a whole new way.
And we like the number one
being really tall there.
We're not just number one.
We're a big number one.
So we like that.
But our vision
has always been
to help you connect
with your customers
in a whole new way.
Now in 1999, connecting
with your customers
in a whole new way
was about surfing up
onto a website.
And we wanted to let
you point, click, close.
Anyone remember that?
It's like, hey, it
shouldn't be that hard.
So that's what we were
doing 26 years ago.
That was 1999 when
this thing called
the cloud could
actually be used
not just for buying
a book on amazon.com,
but for doing CRM, or
Salesforce automation.
Actually, it was just
Salesforce automation
back then.
But as we've moved
through these evolutions
of technology, a
lot has happened.
Since I've been
at Salesforce,
I mean, we didn't have
smartphones 26 years ago.
Mobile happened.
And you're all
like, well, man,
how do I bring mobility
into my enterprise?
Well, we took you there
because we added mobility.
We added the
mobile platform
to Salesforce's platform.
Social happened.
It wasn't just Meta.
It was social within
an application.
So we added social.
AI hit, and we just
talked all about AI.
So we brought you into the
world of predictive AI.
Data-- I'm so happy
we've been investing
for many, many years on
data with our data cloud,
because we had that
platform of data
ready to go when
generative AI really hit
that tipping point,
because generative AI,
as you know-- and we'll
talk more about it--
needs that data
to be successful.
And now we're
bringing you forward
into the world of agentic
AI with Agentforce.
We are going to
take you there.
You don't have to worry.
We are going to
take you there.
And when we
take you there--
we just go to this
next slide here--
you all realize that
it's tough out there.
I mean, I don't
know about you,
but the last few
years have been tough.
We haven't all been
hiring a ton of people.
I don't know about you.
If anyone's been
hiring a ton of people,
I want to know
who you are.
But we have not been
hiring a lot of people.
But the demands
on our businesses
haven't changed.
So the workforces are
feeling overwhelmed.
They have all these
low-value tasks.
They're like, oh, my god,
it's all repetitive stuff.
I need to hire
more people.
But I'm not productive.
People are burning out.
But the customers aren't
saying to you, it's OK.
Don't worry.
It's fine.
Don't worry.
We don't mind
waiting on hold.
I like waiting on hold.
Don't you?
Like, totally fine.
Don't worry about it.
We don't mind if you
don't know who I am.
No, that's not true.
Customer expectations
keep rising.
So you're feeling
this pressure.
You're like, oh, my god.
I'm overwhelmed.
And the customers keep
demanding more and more
and more and more.
So how are we
going to solve it?
Well, what if workforces
had no limits?
You say you can't
hire a lot of people.
But what if workforces
had no limit,
and maybe you can
hire some agents?
Maybe those agents could
actually help you scale.
Maybe those agents
could help you both grow
your revenue,
but also keep
your costs low
or even lower
your costs at
the same time.
Both are super important.
And Agentforce will
take you there.
And it'll take
you there safely.
Of course,
trust, as I said,
is our number one value.
So we're going to bring
you there with trust.
We're going to
make sure we
take care of those
hallucinations, that bias.
We're going to make sure
your data is secure.
You're going to use
the Salesforce security
model, not the LLM, which
has no security model.
We're going to make
it super easy for you
to build your own agents.
At Dreamforce, customers
built over 10,000 agents
on site in like 10, 15
minutes on their own data.
It wasn't just a sample.
It was like, we're
going to build one
for your website, for your
company, for your data.
And they were like,
whoa, that was fast
and that was easy.
That's what we
do at Salesforce.
We want to make
things super easy.
And how do we do that?
Well, you know
how we do that.
It's with metadata.
It's our platform.
It's all integrated.
And you just configure
some metadata.
And now we have metadata
you're configuring
about agents, about AI.
And when you configure
that metadata for AI,
our engineers--
I don't know how they
did it so quickly--
interpret that
AI metadata now.
And then bam, you've
got that agent.
And if that isn't
enough, of course
it's built on
this one platform.
You know about
our ecosystem.
So the ISVs
here in the room
are already building their
own agents on our platform
that you can use, building
skills and actions
that you can use
in your agents.
It's a whole
open ecosystem.
That's what we do.
You know that's
what we do.
And this is very
familiar to you.
We have the
Salesforce platform.
We have Data Cloud.
We have all the Customer
360 touchpoints.
But we didn't take
Agentforce and go,
here it is over here.
Go use it.
And then it's separate
from everything else.
That's not what we do.
We want it to be
one beautiful,
integrated whole thing.
So that's what we did,
is we added Agentforce
to the integrated
platform.
You've got that platform
for that metadata-driven
customization for Slow.
How many love-- are
there any Slowmatics
in the room?
That's what we call
the people who love
automating with Slow.
Slow can help you in AI.
And I think
Patrick might show
you an example of that.
You have Data Cloud,
which I talked about.
How does the AI
know who you are,
know who the customer
is, know your business,
know all your
business processes
and all of the other data
across your enterprise
through zero copy?
That's with Data Cloud.
And you still need
those touch points.
You need to be
able to interact
in sales and service and
marketing and commerce.
So you need all
those applications.
But you know what?
You may not,
in the future,
interact directly with
those applications.
It might be the agents
actually working
with all these services.
Those applications
get broken down
into a bunch of skills,
a bunch of actions
that these agents know how
to use, just like humans.
All of you in
the room know how
to use our applications.
What if these
agents are learning
how to use our
applications for you
and taking action
on your behalf?
That is the magic
of Agentforce.
Companies like
OpenTable have
grown so quickly
that they're getting
inundated with calls.
People are calling like,
hey, I got a problem.
My reservation, I
need to cancel it.
This isn't working.
Well, all of these
agents are exhausted.
So how do we solve that?
Agents.
Agentforce.
24/7.
These agents don't get
tired on Agentforce.
They work 24/7.
They're empathetic.
They know the customer.
OpenTable is loving it.
Those support agents are
getting a little bit more
relaxed because Agentforce
is there helping them.
Another customer
story-- Adecco.
One of the biggest
recruiting firms and job
placement firms
in the world,
they get over 300 million
resumes submitted a year.
And these are people
who want jobs.
And so the people who
are the candidates who
are saying, I
need a job, they
want to get an
immediate response.
They cannot
hire, at Adecco,
enough recruiters to
answer quickly enough all
of those candidates
coming in.
So those candidates
are getting frustrated.
The recruiters are getting
tired and frustrated.
They're like, oh, my
god, I can't help.
This is not a happy place.
Agentforce comes in
and those candidates
can be pre-screened
by the agent.
The resumes
can be reviewed
and actually associated
with a job placement.
Agentforce can even
automatically do
scheduling for
those candidates
to actually speed
up that process,
because Adecco knows
that the longer it takes
to respond to a
candidate, the less likely
they're going to
actually place
that candidate,
which is not
good for the candidate.
It's also not good
for Adecco's business.
So why should you
settle for copilots?
We had chatbots.
And we've had chatbots
for a long time,
and they were kind of OK.
And then copilot
came along.
And wow, that's great.
But it was kind of
fixed and rigid.
And it worked.
It was helpful.
But why don't
you want to move
into the world
of Agentforce,
where it actually
knows your business?
It's got all that
data and grounding
of personalization
to know who you are,
who your customer is.
It can actually
kind of think.
It can plan and reason.
And we have something
called our Atlas reasoning
engine.
So it's not just one
LLM doing thinking.
It's actually a whole
agentic reasoning system
or loop that actually
is reasoning.
So it can actually
be smart and safely
answer your questions
and help you or help
your customers,
and then take
action based
upon that data
and based upon
that reasoning.
It can actually
take action for you
or for your customers.
And it can do it at scale.
That's pretty incredible.
And as you know, one
integrated platform,
but also working in
every single industry.
So of course, Agentforce
works in every industry.
Here in New
York City, we've
got a lot of banking
customers, insurance
customers.
We know that.
I saw a lot yesterday.
So in banking,
proactive financing
to improve loan terms
and insurance, like, hey,
wouldn't it be nice
to pay out faster
with automatic handling
of insurance claims?
That would be pretty cool.
Or how about telecom?
How many people want to
call your telecom, hey,
my internet's down?
My television
is not working.
Customers hate it.
And the providers actually
don't like it either
because it's making for
poor customer success.
Why not just do that with
Agentforce and reduce
those wait times
with 24/7 support?
Working in every industry.
And as we know, I think
you worry about, well,
will Agentforce
be compliant?
It's kind of boring.
Oh, compliance.
But you worry about that.
That's part of what
we think of trust.
We want you to
be able to take
Agentforce wherever in
every line of business.
And so we, of course,
have 40 compliance
certifications
that Agentforce
inherits because it's
part of that integrated
platform.
It all works together.
And bam, we've
got an ecosystem.
How did we do
that so fast?
Well, you know
how we did it.
Agentforce is just
part of our platform.
Of course we have
an ecosystem.
We already have
partner agents.
We have agent actions.
Actions are skills or
APIs or capabilities
that Salesforce has.
But look at that.
All these partners
are giving Agentforce
a lot more power by
the actions in AWS
on Box, on Canva,
on Code Science.
We have data partners
because you need
that personalization.
So you don't have
to copy that data.
If you've invested--
as you know,
if you've invested
in Snowflake,
we love Snowflake.
If you've invested
in Databricks,
Databricks is great.
You don't have to redo it.
Data Cloud will
just connect to it--
you got that
zero copy data--
to make those agents
smarter with data.
And if you need a
little bit of help,
we got some incredible
implementation partners--
Accenture, Capgemini,
Deloitte Digital, IBM,
PwC, Slalom, and yes,
even Salesforce too.
We can help you and
we want to help you.
But before we help
you do all this,
we want to show you
some technology.
And to do that, I would
like to bring up my friend
Patrick Stokes,
EVP of product
and industries marketing.
I've lost him.
There he is.
Patrick, please come
up and take it away.
[APPLAUSE]
All right.
Thank you so much, Parker.
All right.
Good morning, New York.
So autonomous AI
agents, that is--
it's kind of a doozy.
Who the heck knows
what these things are?
The good news is, in
like two or three years,
our kids are
probably going
to know everything there
is to know about this.
Maybe an AI agent is going
to come over for a play
date, play some Fortnite,
maybe do some gymnastics
in the front lawn.
I don't know.
But in the meantime,
it's our job here
to figure out, how do
we use these agents
in the business?
And that's exactly
what we're here
to do this morning.
So the good news is
these things actually
aren't that complicated
when you really
start to break them down.
There's really just
five key characteristics
that we need to
think about when
we're thinking about AI
agents and Agentforce.
So the first is the role.
What's the job that we
want the agent to be
able to do?
What's the function?
The second is the data.
What data does it need
access to in order
to go do that job?
The third is the actions.
What skillsets
does it need?
What do you want it
to be able to actually
physically go out
and do and execute?
The fourth is
the channels,
which maybe
isn't so obvious,
but that's just, where
do you want to interact
with the agent?
Do you want to talk
to it on the phone
or in the chat experience?
So how do you want
to interact with it,
if at all?
And then the fifth is
really, really important,
which is the trust
and security.
So what do you want the
agent to be able to do?
What do you not want
it to be able to do?
What guardrails do you
want to be able to set up?
That's really important.
If you're going to put
these things running
autonomously
in your system,
you need to be
able to trust them.
Now, if you tried
to wire all five
of these things up
all by yourself,
it would be really, really
hard, really expensive.
What you want
is a platform
that can do all
of that for you.
And that's what we're
so excited about,
because with the
Agentforce platform,
you're getting started
really from miles ahead.
All of the roles
are largely
already inside of
Salesforce, whether it's
your salespeople,
service, marketers,
or commerce folks
or people in really
specialized industry
types of use cases.
A lot of the definition
of those roles
and the data
that they need,
and the job that they do
is already in Salesforce.
Much of your data is
already there as well.
Of course, your CRM data,
your structured data,
but not just data
inside of Salesforce,
but with Data Cloud,
you can connect data
from anywhere,
whether it's
your data lakes
like Snowflake
or third-party
systems like your ERP,
but also more recently,
your unstructured data
too, your
knowledge articles,
your conversations in
Slack, or your emails.
All of that makes up
a tremendous amount
of knowledge
that you'd like
your agents to be able to
read, just like you do.
Then your actions.
This is where the
platform really shines.
You've been
building workflows
with Flow and Apex
and MuleSoft for years
to define how your
business works.
All of those are available
as actions right inside
of Agentforce.
And then your channels.
Of course, Salesforce
brings all this together
for you.
You can deploy
on your website,
in WhatsApp line,
Apple Business Manager,
on your phone system, or
in the very best place
to work and
collaborate with all
of your employees,
which is Slack.
And then finally,
that trust layer.
You've heard about our
Einstein Trust Layer,
but let's not forget
all of our data
permissions built right
into the platform,
role-based
access controls.
All of that is built
right into Agentforce.
Now, once you've
defined these agents,
you want to give
them a place
to actually go
out and run.
And this is where
we're so excited
because, as the world
has been enamored
with the idea of
the LLM itself,
and how big is the LLM,
how many parameters is it,
the LLM itself is a part.
It's an important
part, but it's not
the whole part.
The whole part of
an energetic system
is being able to connect
data to an LLM to action.
And that's what we've
built with our Atlas
reasoning engine.
And this is a
looped system.
So as you define
these agents,
your agent is
looking at your data
and coming up with a plan.
It's reasoning and
making a decision
and coming up with a plan.
And it doesn't
just send that plan
back to the human
beings to go execute.
It can actually go out and
execute that plan itself.
And when it's done,
it doesn't stop there.
It sends the signals,
the outcome of that
plan right back
into Data Cloud.
And what does that mean?
It means that Agentforce
gets better over time.
As you use it, it's
learning your business,
and it's going to get
better and better.
It's actually an
appreciating technology
asset, which is
pretty incredible.
When you put all
of this together,
this is how our
customers are
getting so much success so
quickly out of Agentforce.
Instead of trying to build
all of this themselves,
they're getting to
value 39 times faster.
And they're doing it
with better accuracy.
Agentforce delivers 33%
more accurate answers,
and they're two
times more relevant.
This is really incredible.
Nobody would have thought
just six months ago
that you could
move this quickly.
Now, to do this, we need
to be able to build.
We need to be able
to build and deploy
these agents.
And that's why
everyone in this room,
every trailblazer
in this room,
is now an Agentblazer.
Because at
Salesforce, what we've
been trying to
do for 20 years
is make the hard
things really easy.
And one of the core
tenets of our platform
is this idea of low code.
So we've built our
low-code Agent Builder.
It's GA now.
This is the
place where you
can go and build and
define and start to play
with all of these agents.
But we're also
today very excited.
We just launched
this today,
our all new Agentforce
Testing Center.
One of the things that
happens when you start
to build these agents
and talk to them
is you realize, you
ask it a question,
and it gives
you an answer.
But if you ask
that question in
just a slightly
different way,
you might get a
different answer.
And you realize,
oh my gosh,
I've got to generate
all sorts of variations
of these questions.
Well, that's what Testing
Center helps you do.
And this means you can
deploy agents much,
much faster and
with the confidence
that you can trust
them as they go out.
Now, as always, we want
to tell you a story.
We want to tell
you a story
of an incredible
agentblazer.
And that's the
story of Saks.
Now, Saks is delivering
luxury experiences
for every one of their
customers with Agentforce.
In just over a week, they
built a brand new customer
service agent on
top of Agentforce.
And I don't want to tell
you this story myself.
I want you to hear
it right from them.
So let's go ahead
and roll the video.
[VIDEO PLAYBACK]
- We believe there's going
to be a fundamental shift
in the way that
you interact
with humans at Saks.
That's really going to
be enabled by Agentforce.
- At Saks, we
deliver a dream.
We sell things that
people want to love,
the emotional connection
you have with finding
that one thing you want.
- There are lots of
new changes happening
across the
technology ecosystem.
You need to have
the best technology.
- We saw the luxury
consumer beginning to move
to e-commerce.
We said, we got to pivot.
- Really important for us
to meet the customer where
they want to be
and service them
in the best ways.
- Agentforce offers
personalized service
to customers like Maria.
In this case, she has
a birthday coming up.
- If you have
millions of customers
and you want to
create millions
of different experiences,
it's not scalable.
But Agentforce gives
us that ability.
- Agentforce will enable
us to service customers
in ways we couldn't
imagine before.
It will recognize photos
and understand text
and make personalized
recommendations
to customers like Maria.
- We really believe there
is an immense opportunity
for Agentforce to allow
for our service agents
to become stylists, really
focus on clienteling
with our customers.
We are recommending
things for you
that are personal to you,
that are authentic to you.
Agentforce is really
allowing our humans
to have the deeper, more
impactful interaction
with our customers.
- It gives us the ability
to continue to offer
a very high-touch luxury
experience that people
expect from a Saks.
And I know it
sounds simple,
but that full 360 degree
view of the consumer,
everything they're doing,
from how they're browsing,
from how they're shopping,
from what they return,
from what they call
the call center about,
from how they pay,
from everything, really
understanding them.
We're going to give them
that comfort that they're
the most important to
us, which they are.
- When Maria
contacts us, it's
intelligently routed to
a customer representative
based on the nature
of her question.
- Agentforce is going
to free up our people
to work in a different
way with their clients.
It's not going to replace.
It's going to augment.
- Here, we have Agentforce
making a recommended
response based on customer
data, previous cases,
as well as
knowledge articles.
All we have to
do is hit Post.
- We believe
that Data Cloud
is going to be the
platform for us
to activate that data
through servicing,
through commerce,
through order management.
- If you can get people
comfortable and if you can
work with them on
an individual basis,
a personalized
basis, the sky's
the limit on what
you can do with them.
- I don't believe that
the experience that we
are delivering
would be possible
without Salesforce.
- It goes back
to the belief
that humans with agents
drive customer success
together.
Together, that's how
we're going to win.
[MUSIC PLAYING]
[END PLAYBACK]
All right.
So that was the
incredible story of Saks.
And we're going to show
you their new agent
in just a moment, but
first, what I want to do
is I want to
share a little bit
of my own experience
with this.
So we're approaching
the holiday season.
Thanksgiving is
next week, and we're
going to be in full
swing any minute now.
It's probably going to
get cold in New York.
Who knows?
It's been like summer
the whole time.
But I needed to
order a new outfit
because I've got some
holiday parties coming up.
In fact, I've
got an early one
starting right this week,
which is kind of crazy.
So I went on saks.com, and
I ordered a new sweater.
And it came to my
house just a little bit
earlier this week, and
it didn't quite fit.
It was a little
bit too small.
So I thought it
would be really
fun to stand up in the
middle of a room with--
I don't know-- a
couple thousand people
and call customer
service together.
Sounds fun, right?
So let's go ahead
and do that.
So here we go.
Hello, and thank you for
calling customer service.
If you know the
extension of the party
you are trying
to reach, you
may dial it at any time.
To hear our store
hours, please press 1.
To start a return,
please press 2.
For product inquiries,
please press 3.
If you would like to speak
to a customer service
representative,
please press--
OK.
I don't know
about you guys,
but I don't want
to do any of that.
What I really want
to do is actually
just start talking.
I want to describe
what my problem is,
and I want to get
some help really,
really quickly.
I think we
actually all know
where that's going to go.
I'm just going
to go to option 4
and ask for an
agent, anyway.
The irony of this
thing is there's
a whole kind of technique
in the customer service
industry for this.
It's called
case deflection.
They're just trying
to deflect you
from the more expensive
channel of actually
talking to a human being.
And the result is that you
kind of, most of the time,
don't actually get the
help that you need.
And that's
really not great.
That's not what we
want as consumers.
What we really want
is we want help.
We want solutions.
We want resolution.
Well, the good news
is it does not have
to be that way anymore.
That experience
that I just
had doesn't have to
be that way anymore.
And it doesn't have
to be that way anymore
because of Agentforce.
Now, earlier I mentioned
that Saks used Agentforce
to build a new AI
agent, and they did that
in just a little
bit over a week.
And so they call
this agent, Sophie.
What I want to do now
is give Sophie a call
and see if the experience
is any different,
hopefully a
little bit better.
So let's do that.
All right.
There's Sophie.
Hi, I'm Sophie, your
Saks AI assistant.
How can I help you today?
Yeah.
Hey, Sophie,
this is Patrick.
I'm wondering if you
can help me with my most
recent order.
I think I might have
gotten the wrong size.
I'm sorry to hear
that, Patrick.
I'm happy to help.
Is this regarding the Saks
collection seed stitch,
crewneck sweater you
purchased last Saturday?
Yeah, it sure is.
I think it might be just
a little bit too small.
I'm sorry the sizing
didn't work out.
Based on your
purchase history,
A size medium should
provide a relaxed fit.
How does that sound?
That sounds pretty good.
One question though,
is it going to shrink?
This sweater is 100%
fine Italian wool,
so we recommend
dry cleaning only
to preserve the
integrity of the fibers
and avoid any
risks of shrinking.
Would you like to
go with a medium?
Yeah, fine Italian wool.
That sounds great.
Yeah, let's do the medium.
Great.
Based on your
shipping address,
you should receive it
within three to five
business days.
Does that work?
Ooh, no.
Actually, I've got
a party coming up
in just a couple
days that I
was hoping to wear it to.
Is there any way that
we could get it to me
a little bit faster?
Unfortunately, that's
the fastest shipping
available for this item.
Would you like me
to transfer you
to someone who can
assist you further?
No.
That's OK for now, Sophie.
Thanks.
OK, got to hang
up on Sophie.
There we go.
So much, much
better experience.
The thing that I
like most about it
is that I was able to just
start talking and describe
my problem.
I didn't have to listen
and interpret a menu.
I just was able to start
describing my problem.
And Sophie knew
exactly who I was.
This is one of
the cool things
about Agentforce,
is agent force knows
who your customers are.
And I didn't have to
give an order number.
She was able to just
run all the queries
and find my order
and immediately
start getting me help.
That was a really
awesome experience.
But you may have
also noticed
there was one thing
there at the end
that she couldn't
help me with, which
is she couldn't come up
with an option for getting
the item to me, the
sweater to me a little bit
quicker.
But Saks actually
does have a way
to do that, which is
they've got a store right
down the street,
about a half
a mile away from here.
And so imagine if
Sophie could actually
identify that there's
a store nearby,
and then imagine if
Sophie could look
into the real time
inventory of that store,
see if the
sweater is there.
If she knew that the
sweater was there and knew
that the store
was close by,
she might be able to
offer me a new option,
to go to the store
and pick it up,
and then I'd
have it in time.
Now, that sounds
kind of crazy.
It sounds like that might
take a long time to build,
but it really doesn't
with Agentforce.
And that's because
of our low-code tool,
Agent Builder.
I want to show you
just how easy it
is to configure these
agents to do exactly that.
So we're going to
go to the demo here.
And no tricks, no magic.
I love our demo
team, but I
wanted to do
this demo myself
right here in front
of all of you.
Although to be
fair, they're there
in case I screw it up.
So here we go.
So here we are
in Agent Builder.
You can see we've
got Sophie here,
and we've got a little bit
of a definition of what
her job is and the
channels that she
interacts on.
But what I really
want to look here at
is our topics.
Now, our topics
are just kind
of groups of
knowledge, the things
that Sophie can
actually go and do.
And you'll see we've
got an Order Management
topic here.
And if I look inside
of Order Management,
it's just got
a description
and a couple of things.
See this is how the
AI is connecting
what I was asking about.
I was asking
about an order,
and so Sophie was able to
narrow in on this topic
and follow these
instructions.
There's no code.
She's just doing
it with reasoning.
Within this topic is a
whole bunch of actions.
And I can see, in fact,
some of the actions.
If you recognize these
icons here, a lot of these
are just flows.
These are actions
that we actually
just saw Sophie perform.
She looked up my order.
She checked on the size.
She even modified
the order for me
to get me a medium.
But you'll notice
that there's
a few missing
actions here.
There's nothing about
identifying nearby stores
or the inventory.
So if I wanted to give
Sophie the ability
to do that, actually
all I'd have to do
is add some new actions.
And I've got
some actions here
that look like they might
be helpful, like change
delivery method, get
nearby stores, which you
can see as a MuleSoft
API, schedule
an in-store pickup, but
I don't have anything
to check to see if
I've actually got
the sweater in the store.
So to do that,
what I need to do
is add a new
Agentforce action.
So I'm going to go
over here to Setup,
and I'm going to click Add
New agent force action.
And this is where you just
expose all of the MuleSoft
APIs and flows that you've
built to Agentforce.
And you can see
we've got Flow here.
All of the flows that I've
already built in my system
are available to me.
With one click,
I could make them
available to Agentforce.
But in this case, I want
to add a MuleSoft API
because I happen to have
a store inventory API
actually already all
set up for my stores.
So I can do that.
I'll click Next.
I can change the inputs
and outputs if I need to.
The good news is
we don't need to.
We're just going to
go ahead and hit Next.
And I can jump back over
here to Agent Builder.
And now, when I
go in, and if I
refresh these
actions, you'll
see I've got a
new one here.
So let's add a couple.
Let's add Store Inventory.
Let's add Schedule
In-Store Pickup.
We're going to need
to see if we've
got a store nearby.
So we're going
to add that API,
and then we're going
to add Change Delivery
Method.
And I'm going
to click Finish.
And this is where
something really magical
happens.
I don't have to give
Sophie anything else.
I don't have to
write any code.
I don't have to write
any more instructions.
I just have to
provide the ability
to look at those actions
and identify what they
are, which Sophie can
do because they've
got descriptions.
Now, to prove that,
let's actually
chat with Sophie right
here in the chat interface
and see if this all works.
So I'm going to
put my name in,
and I'm confident enough
to do a demo on stage
but not type on stage.
So I'm cutting
and pasting.
So hi, my name is
Patrick Stokes,
customer ID yada yada.
And we're going to
see Sophie work here.
And what's so cool
inside of Agent Builder
is I can actually
see Sophie thinking.
So she selected that topic
based on my question.
I mentioned
the word order,
so she went into
Order Management.
She's finding
my order number.
This is where she's
querying against the flow
to get my order number.
And she's got a
response here.
So it looks like
you still need
to select a
delivery method,
but there's
something new here.
She's saying that now I
can pick it up in store.
That is pretty incredible.
How did she
figure this out?
Well, let's look.
She checked the
order status.
She looked at my
question, which is,
can I get it there sooner?
She realized that the
order status and sooner
doesn't quite match up.
So she's looking
now to see
if there's nearby stores.
She found a nearby
store, and now she's
looking into the
store inventory
to see if the thing
is actually there,
the sweater.
But it's not quite done.
Now she realizes she's
got to come back to me
and ask me if I would
prefer an in-store pickup.
I sure would.
We're going to
select pickup,
and we're going to see
the rest of the process
unfold here.
She's now changing
the delivery message,
running the schedule,
in-store pickup flow,
and giving me my response.
Now, we just tested this
inside of Agent Builder,
but let's see if this has
actually propagated out
to our Sophie customer
service call as well.
So let's go back and
call Sophie one more time
and see if this works.
Hi, I'm Sophie, your
Saks AI assistant.
How can I help you today?
Yeah, hey, Sophie.
It's Patrick again.
I just wanted to
see if I could
get an update on the
delivery time of my most
recent order.
Is this regarding the Saks
collection seed stitch,
crewneck sweater?
Yep, that's the one.
We have two delivery
options available.
I can have it delivered
within three to five
business days, or it
can be picked up as soon
as today at Saks New York
in Midtown Manhattan.
Which option
suits you better?
Yeah, let's do
Midtown Manhattan.
I don't know how I'm
going to get across town
that quick, but
let's do that.
Sounds great.
Great.
I've placed the order and
notified in store client
advisor that you will
pick up the item.
They will have
it ready for you
within the next
three hours.
Is there anything else
I can help you with?
Nope.
Sophie, you
have been great.
That was very helpful.
Thank you so much.
I always forget to
hang up on Sophie.
I hope she doesn't mind.
All right.
So you can see, just
with a few clicks
inside of Agent
Builder, we've
given Sophie a
whole new ability.
She actually just reasoned
her way through that
and figured out that
there's a nearby store,
and there's a way to
get me that item sooner.
And so you can
really see how
this is what AI
was meant to be,
how agent Salesforce is
what AI was meant to be.
Think about the ways
that this can augment
all of your employees.
Think about a customer
service rep actually
solving problems
so that they
don't have to go to that
live person all the time.
Think about what this
means for the customer
experience, for the
experience that I just
had.
And most
importantly, think
about what this can
mean for your margins
and ultimately
for your revenue.
But to really make
all of this work,
we have to
understand, how do we
get the data to these
agents, to Agentforce?
How do we bring that
data together in a way
that they can
actually use it?
And the good news is,
we've got the world's best
person here to
explain that to us
and walk us through it.
Please welcome to
talk about Data Cloud,
our VP of product
marketing, Sanjna
Parulekar.
Sanjna, welcome.
[APPLAUSE]
All right.
Thank you so
much, Patrick.
Now, you've seen
Agentforce in action,
but data is really
the beating heart
of how Agentforce does
all of that cool stuff you
just saw Sophie do in that
interaction with Patrick.
But in order to fully
understand its power,
I want to debunk
a little myth here
for everyone today
on how AI can
work for our business.
Now, what a lot
of folks think
is that I need to
train an LLM with all
of my business's
data in order for AI
to work for me.
And that's just
simply not true.
It's costly.
It's time intensive.
It requires specialized
developers and data
scientists.
And more importantly,
as your data changes
and as dynamic as
your customers are,
the AI doesn't change with
it without retraining.
So this really
isn't the way
that we're going to get AI
to work for our business.
Instead, we're going
to flip the script here
and put the power
inside of our prompt.
Now, a prompt is nothing
but a set of instructions
and questions that we send
to a large language model
to teach it about
our business.
So everything you
need for an LLM
to know about
your business
can reside
within a prompt.
But what we
also don't want
is a prompt that's
hundreds and hundreds
of pages long
because it has
all of that context
that we want
to give back to the LLM.
So what we want
to employ here
is a new technique called
Retrieval Augmented
Generation, or RAG.
Now, what RAG
does is it enables
us to put a ton of
context into that
prompt that searches
through all of the data
that I have
available to me
and retrieve that
data into the prompt
and teach our LLM
about our business.
This is how we get
AI to work for us
and for our customers.
So you see here
that we don't
need to do any DIY
projects to get
AI to work for us, but I
know what you're thinking.
It sounds great to
get all of this data
into an agent,
but my data is
sitting in totally
different systems.
We know that all
of our customers
have islands of
disconnected data sitting
everywhere, whether
that's in an external data
lake, several different
Salesforce clouds, or even
some legacy systems
that might not even
have an API.
We understand this
pain very, very well.
And it's actually
the inspiration
behind why we created Data
Cloud in the first place.
And it's our
hyperscale data engine
that resides deeply
inside of the Salesforce
platform.
And the mission we had
in mind with Data Cloud
was really simple.
We want to enable you
to put all of your data
to work, no matter
what kind of data
it is, telemetry data,
health data, website
engagement data.
We want you to put that
data to work for a better
customer experience.
Now, to really understand
the magic of Data Cloud,
we got to talk about how
it works under the hood
too.
And we make it really
easy to connect
to any of these systems.
And we have a few
of these logos
here, whether it's
your Salesforce clouds
or different Salesforce
orgs that you have
for different regions.
It could be external
applications,
it could be external
data lakes like Snowflake
or Databricks,
or it could even
be systems that
are storing
your unstructured data
like your PDFs, your voice
data, audio files,
video files, anything
you can think of.
Now, once you've
connected to this data,
we do this in a really
special way using our zero
copy integration.
Because we don't
want all of you
to incur the
cost of moving
your data every time you
want to use it inside
of Salesforce.
We want to make it easy.
And so what
Data Cloud does
is we virtualize that
data inside of Salesforce
so you can use it
like any other object
and activate it across
the Customer 360.
Now, another vital
part of Data Cloud
is harmonization,
because once you've
connected to your
data, you still
have different versions of
the same customer residing
in those systems.
All of your data systems
speak different languages.
And what Cloud does is it
enables them to all speak
the same language.
So you have that
central, unified view
of every single
customer that you
can activate in
any experience
that you choose.
You can govern this data.
You can govern
these profiles
with define policies
inside of Data Cloud.
And you can activate
this data, of course,
inside of an agent
or any other actions
that you want to drive
for a better customer
experience.
Now, Patrick mentioned the
Atlas reasoning engine.
And Data Cloud is
really at the heart
of the way that
agents reason
and plan with Atlas.
And retrieval
augmented generation
is such a vital
part of this,
and we're going to show
you this in action in just
a couple of minutes here.
Now, Data Cloud really
is at the heart of what
makes Agentforce better.
And I think
the easiest way
to think about
it is that so
many of these capabilities
around personalization,
cross-sell, upsell,
or even just closing
deals faster with
the right information
for your reps at
the right time,
it wouldn't be possible
without connecting
to all of those systems
with Data Cloud.
So I want to show you
this in action in a demo,
and I'm going to have
my friends [? Claire ?]
and Rebecca behind the
demo desk count me out.
Let's give them a hand.
[APPLAUSE]
All right.
Now, Patrick had his story
with Sophie and his Sak
shopping experience, but
I got a story of my own.
And I think it's a
pretty relatable one.
I think we can all
relate to that feeling
when you buy an
expensive item,
and then it immediately
goes on sale,
and you want
your money back,
a price adjustment
workflow.
Now, that type
of question,
if I ask it to
an agent here,
it might take a service
rep at Saks several hours
to solve this case.
They'd have to know the
return policy practically
by heart to figure out
how to process my refund.
But the great thing
about agent Salesforce
is that a task that's
really complicated
for a service rep
is a perfect task
for Agentforce.
So why don't we
go ahead and see
how we can enable Sophie
with this new ability
to handle this
customer question?
Now, our journey here
starts, of course,
in Data Cloud
in connecting
to the right
data that we want
to empower Sophie with
to solve this question.
Now, it looks
like we've already
gotten a head start here.
We've connected to our CRM
data, which is awesome,
but in order to really
answer this question,
I need my Order Management
system data that
sits inside of Snowflake.
So I'm going to go ahead
and create a new data
stream.
And Data Cloud makes
this really simple,
with all of these
out-of-the-box connectors
for any data system you
can practically imagine
here.
So I'm going to go ahead
and select Snowflake
and create this new data
stream with my Order
Management system data.
And what you're
going to see here
is it's automatically
creating this data graph.
And it shows a
visual representation
of all of the
relationships
across my different data
systems to that individual
that we want to solve
this query form.
So our next step here
is going to pop back
into Agent Builder.
Now, Patrick
showed you this.
Agent Builder
is where we give
our Agentforce
new abilities,
through topics, actions,
and instructions.
So we already
have a topic here
around pricing questions.
And we've actually
drafted an action as well
on price adjustment.
And this action
is a prompt.
I told you before the
power is in the prompt.
So why don't we go ahead
and open up Prompt Builder
and make this prompt a
little bit more powerful?
Now, Prompt
Builder has been
available to our
agentblazers for a year
now, and they're
absolutely loving it.
And they're loving
it for a couple
of different reasons.
The first is
that we enable
you to select
any model you
want to use under the hood
with a simple dropdown.
You don't need
to write any code
or figure out a
complex evaluation.
All you have to do
is test it inside
of Prompt Builder.
Now, the second reason
that our agentblazers
love this is because this
is prompt engineering
the Salesforce way.
It's completely low code.
So you can
create and revise
these prompts in
natural language.
So we've already
gotten a head
start here and paste it in
the draft of our prompt.
And I want to make
it a little bit more
intelligent with
some data that I
connected to
with Data Cloud
just now with you all.
So I'm going to
go ahead and drag
in the name of the contact
to really ground that
prompt and the information
about my customer as well
as some of that
Order Management data
that we just connected
to inside of Snowflake.
So when I drag
all this data in,
I can also test this
on a contact record.
So I'm going to test it
on my own contact record
here and see what the
response could look
like from Agentforce.
So I want to call
your attention
to the resolution here,
because I can actually
see how the prompt is
thinking and retrieving
the right data to
create that response
you see on the right.
And I'd say, I'd give this
response like a B-plus.
It's OK.
It's giving me
some context.
It's giving me the
amount of the refund.
It's addressing
me by first name,
but I think we can do
a little bit better.
And we can do better
with retrieval augmented
generation.
So what I've
done previously
is I've created what's
called a retriever.
Now, all a
retriever does is--
it kind of does what
it sounds like it does,
which is retrieve the
right data into the prompt
from what I specify.
So I'm going to drag in
my retriever for my return
policy in here.
And when I drag
that in, I also
want to give this
retriever some parameters,
because I don't want
the LLM to waste time
searching through
every single document
every time I
run this prompt.
I want to give it
some specifications.
So I wanted to
specifically look at price
adjustment eligibility.
Now, let's go
ahead and save this
and preview it again
and see how we did OK.
If we look in
the resolution,
we're going to see exactly
where this prompt is
looking for the
right context
and grounding our
prompt with it.
And our response
is much better.
So why don't we
see it actually
inside of the experience
that we started out with?
So I'm going to ask
my question to Sophie.
I want this
price adjustment.
I'm going to confirm which
product it is because I've
been shopping a lot.
So we want to make sure
it's the right product
for the price adjustment.
And immediately
what you're
going to see here is
that Sophie can not only
adjust the price here,
give me the amount,
all the context
that I need, but can
process that refund for
me in a matter of seconds.
This is the power of Data
Cloud under the hood,
empowering us with the
right data for agent
for us to really take
action for the customer.
And Saks was able
to do all of this
quickly with Agentforce.
They were able to ground
their prompts, empower
Sophie with new
abilities, and do
all of this on
the platform
without training
a single model.
So now that
we've shown you
how the platform
works, the power
of Agentforce,
all of this really
does come to life
when you see it
across the Customer 360.
And to do that,
I want to invite
our SVP of Trailblazer
Community, Leah
McGowen-Hare.
Thank you.
All right.
New York, it's always
great to be back
at my old
stomping grounds.
Brooklyn in the house!
Oh, things have changed.
All right.
Well, as Parker
said, if you've
been rocking with
us since 1999,
you know that our
mission, our vision
has always been
to help you
connect with your
customer in new ways.
And this has not changed.
If anything, we have
doubled down on this.
And we do this by helping
our customers build
their Customer
360 by pulling
in their service, sales,
marketing, commerce,
all into one tightly
integrated platform.
And now, we're
bringing Agentforce
to every one of
these applications.
And we want you
all, everybody here
to meet their
Agentforce today.
We're talking one
Agentforce on one trusted
platform.
And what this is going
to allow you to do
is it's going to
allow you to create
a seamless experience
for your customers
across all lines
of business.
So let's look at this
for each application,
shall we?
Let's kick it
off with sales.
So Agentforce for
sales acts as your SDR,
helping your sales team
convert and increase
their pipeline, as well
as provide coaching
to all of their sellers.
And then there's
Agentforce for Service.
And here, they're working
with your service reps
to help close cases faster
as well as give proactive
and personalized
service 24/7.
And I know what
you're thinking,
but what about
marketing, Leah?
Well, Agentforce
for marketing
works with your marketers
to help them create
and deploy highly
personalized
and super precise
campaigns at record speed.
And then,
Agentforce-- oops--
Agentforce for commerce.
Now, this is really
amazing because what
it does is it's
going to work
with your merchandising,
your buying,
and your personal
shoppers,
all to help your
customers find the items
that they're looking for.
And then, I know, what
about revenue and orders,
Leah?
Well, we didn't forget
about them either.
With Agentforce, we have
automated the revenue life
cycle, from quotes to
contract to billing.
And this is
super powerful.
And with Agentforce
in Tableau,
we are bringing
AI to analytics.
Now, those are two
powerful words.
And bringing those
together, mind blowing.
And what this
allows you to do
is to see your data
easier so that you
can move from insight to
action on one platform.
And lastly but
definitely not least
is Agentforce in Slack.
And you're going to see
how you can collaborate
not just with your
fellow colleagues,
but you're going
to collaborate
with agents as well,
right in the flow of work.
Now, we're going to
bring this in-- oh,
and let's not forget
our industries.
Now, we understand that
each industry is unique.
So we have over 100
industry-specific actions
for Agentforce available
right out of the box.
Now, we're going to
see this innovation all
come together through the
lens of Fisher & Paykel,
which is a premier
brand, premier
brand in appliances.
And this is really
timely because we're
talking about holidays.
Everybody's cooking.
So we're going
to look at this.
And we're going to have
a lot of great innovation
that we're going to see
through four mini demos,
or as I like to
call them, menos--
I just made that
up, four mini demos.
So we're going to see
lots of innovation
through four demos.
Let's kick it off
with our first one,
with Agentforce
in field service.
So here we are.
Now, a customer's
refrigerator
has been beeping.
Now, we've all been there.
Maybe you haven't,
but I know I have.
And you kind keep
pushing through
and you sort of ignore it.
Well, you know who
is not ignoring it?
Agentforce.
Agentforce sends
the customer a text
and is like, hey,
you know what?
We're noticing--
and does say that.
It says, hey,
you know what?
Hey, we noticed that your
freezer temperatures are
high, and we want to send
a technician out there
to check out your freezer.
Well, it's the holidays.
We don't need anything
defrosting unnecessarily
so.
So they come to a date,
they go back and forth
with the customer,
finding a date
that works for the
customer, and voila, it's
settled.
Now, this is all
proactive with Agentforce.
And now we want to
fast forward to the day
that the technician
is coming out.
So when the
technician comes out,
they receive a
brief of everything,
a prebrief of the work.
Now, what's really
cool about this
is they're not
just getting
a brief about the
work they have to do.
They also know
about the customer.
Now, I know--
I've had technicians
come to my house,
and they didn't
even know my name.
But having here,
this technician
knows who the
customer is, knows
about past incidents,
all of that information.
And what this is
doing is, Agentforce
is preparing
that technician
to provide
excellent customer
service,
personalized customer
service, and
the technician
knows all of the
items that they
need to fix the problem.
Now, by show of hands,
how many people have
had a technician
come to their house,
they get there,
work on something,
and then go, oh my
bad, you know what?
I'm going to have to
reschedule because I'm
missing a part.
How many people have
experienced that?
OK, but guess what?
Not on Agentforce's watch.
Because in the past,
the technicians
had to manually
check on the truck,
make sure they
had everything,
match up all of their
jobs with the inventory.
Now with Agentforce,
Agent force automatically
identifies every
item that is needed
for that technician
on their truck
for the entire week.
And what that means
is they show up
to the job ready and able
with all of the tools.
So once the job
is successful,
here, Agentforce
automatically
writes a summary up
removing that admin task
from the technician
so that they
can start prepping
for their next job.
But this is even
powerful because, inside
of this summary here,
it actually triggers
a workflow that
then goes and adds
the customer to any
follow-up visits
or any checkups that
they might need to know
or any notifications.
That's the power of being
in an integrated platform,
being able to leverage
those workflows,
those MuleSoft APIs, all
of the validation rules.
All right.
That was meno number 1.
We ready for
meno number 2?
Mini demo--
remember, mini demo.
Number 2.
Let's jump into
number two with Slack.
So welcome to
Agentforce in Slack.
Here we have an army
of agents ready to go.
So we're going
to go ahead.
Let's check out the
Inventory Agents profile.
Now, earlier, we
saw Patrick show us
how these agents
are built.
Here we have the inventory
agent with the skills,
all their skills,
their topics
and the knowledge
base articles
that power this agent.
But let's see all of
this come in action.
So we're going to go to
the channel [HICCUPS]
here-- excuse me--
the American--
amer-east-distribution
channel.
Now, in here, we
have 119 employees
collaborating in there,
but I want you to notice.
Right next to it,
you see that 5?
That means we
have five agents
in there working alongside
the employees as well.
So let's see what
this looks like.
The inventory manager
asks the agent, hey,
how are our inventory
levels looking
for the rest of the year?
Well, check this out.
The inventory agent comes
back with information,
but it has receipts
for that information,
in the form of Tableau
dashboard right there.
So here's the thing.
This agent has
the CRM skills,
but it also has
Tableau skills.
And this
inventory agent is
able to use predictive
analytics because it says,
we notice that 17% of jobs
are scheduled within two
days of their incidents.
And this is also a
really busy time of year.
So all that to
say, you need
to order some
more inventory.
But all of that
predictive built right
into that agent.
And here's the thing.
The agent doesn't
just give insights.
It also takes
action because it's
on the integrated
platform.
It has access to
all of your flows,
all of your Apex, all of
your validation rules.
And here, it kicks
off the order process,
which is an actual flow.
So here, this is super
integrated right here.
This inventory
agent is kicking off
the order process.
Awesome.
Let's jump into
vignette or meno number
3, when we're going
to look at Agentforce
and marketing.
So here we have Ageforce
in marketing, which
is going to help
your marketers create
and deploy very
personalized campaigns
with great precision.
This is an
existing campaign,
but agent force
notices this campaign
is missing a segment.
If you look on the
right here, it's saying,
you're missing a
segment of customers
that only have one
connected appliance.
Now, I want to pause
here for a minute.
Marketers know identifying
segments is difficult.
Unless you've got like a
ride or die data scientist
with you 24/7, these
things take a while.
And even if you do have a
data scientist-- no shade,
data scientist--
but this would
have took days, weeks, or
days to figure this out.
Agentforce figures
this out in seconds.
So here,
Agentforce not only
identifies the segment
but also identifies
the attributes to
complete this segment.
So when we go in and
look at the attributes,
we can see the first
attribute is pulling back
all of the
individuals that have
one connected appliance.
And then the
second segment
attribute-- now
this is where it
gets really interesting.
Here, we're using
predictive analytics
because Agentforce
identified that customers
typically go
into a showroom
when they want to
purchase an oven.
They want to touch it.
They want to see it.
You're not going to just
purchase it online and not
look at it.
So this attribute
is going to find
folks that are close
to the Fisher & Paykel
showrooms.
So we are all good
with this segment.
We want to go ahead
and create the segment.
And now, as
we're doing this,
Agentforce is
still at work.
It's like, how can we help
you move inventory faster?
Well, Agentforce
comes back and says,
you know what?
You should also offer a
promotion to this segment.
You should give a
20% off to help move
this inventory faster.
Well, I'm like,
if I'm a marketer,
I could definitely
use a partner
that's going to help
me sell my inventory
a lot faster.
And that's exactly
what Agentforce
is doing right here.
So let's see what this
all looks like when it
comes together in the ad.
Now, here we can see
that lovely ad, 20% off.
That oven is beautiful.
Now, Patrick and Sanjna
had their shopping moment.
I would like to have
my shopping moment
and purchase this oven.
But then that means
I need to cook.
That's a whole
other thing.
But we're going to go
ahead and book that.
That looks great.
That's the power of
Agentforce in marketing.
Now, let's wrap it up
with meno number 4.
We're going to look at
Agentforce in sales.
Now, in here,
you're going to see
how an SDR agent helps
sellers qualify prospects.
Now, a customer has
received this highly
personalized
email telling them
about a promo for oven.
The customer is very
much interested in it
and has some
questions and wants
to go back and forth.
Now listen,
it's Q4, sellers
are out there trying
to close deals,
hit their numbers.
Going back and forth
and trying to figure out
if this person is really
serious about oven
takes a lot of time.
Well, this is
where Agentforce
comes in and lightens
the load for the sellers.
So Agentforce goes back
and forth answering
the question.
And when the customer is
serious about purchase,
it has moved from
the looky loo stage
into I really want to
purchase this oven.
We can go ahead and
schedule the appointment.
Agentforce schedules
the appointment
to bring them
into the showroom.
That's awesome.
So now, if we fast
forward to the day
they're coming
in, the seller
can get fully
briefed on everything
that the SDR
agent has said
or the conversation has
had with the customer, all
the things the customer
is looking for,
right there in the middle
is the conversation.
And if the seller
is short on time,
they can see right
there on the right hand,
there's a summary that
Agentforce also put
together for the seller.
So the seller is
fully briefed on who
the customer is.
And what you're
seeing here
is Agentforce is really
lightning the load so
that the seller
can work with all
of their customers and
they can have qualified,
quantity time,
quality time
with quality customers.
Now, before I
close this out,
I need New York
to help me out.
Can you guys help me out?
All right.
Can you help me out?
All right.
So when I do this, I need
you all to say Agentforce.
Are you ready?
Let me test it.
Agentforce.
I need you to say
it with conviction.
Agentforce.
All right.
So that's how
humans with agents
drive customer
success with--
Agentforce.
Thank you.
Back to you, Parker.
All right.
Great job, Leah.
You're a hard
act to follow.
All right.
Well, Patrick took us
through how easy and also
powerful Agentforce is.
And you can
build them, too,
if you go out into
the room over here
just after the show,
but don't leave yet.
But Agentforce
is super easy,
powered by Data Cloud.
Sanjna showed you the
power of Data Cloud
to personalize
those agents.
And Leah just knocked
it out of the park
to show you how those
agents can appear
throughout the
Customer 360,
with all those
touch points.
So all of you want this.
You want to get to the
world of AI customer
success.
I want it.
You want it.
That's what we want.
We want to help
you connect
with your customers,
with these agents.
But we've all
been trying to do
this for probably the
last couple of years,
and it's not easy.
You might have been
training some models,
building your own
custom models,
retraining because you
want to keep them up
to date with
the latest data.
You might have data that
you don't have access to.
It's trapped.
You might be
disconnected from all
those applications
you've been trying.
How do I hook
up the skills
that all those
applications?
You're like,
wait, these LLMs
have no security model.
Oh, no.
My data is all
in this LLM,
and I have no
security model.
You're hiring
specialized AI engineers,
scaling, rapid change.
This is hard
stuff, people.
So don't do it yourself.
26 years ago we
said, why are you
trying to build and
run your own Salesforce
automation?
This is crazy.
Why are you
buying computers?
Why are you hiring people?
We are doing the
same thing today
with Agentforce.
Don't do it yourself.
Use Agentforce.
It's right there.
You have it.
You have the platform.
You have Data Cloud.
You have the Customer 360.
You have Agentforce.
And you might be
saying to me, Parker,
I don't have this.
Oh, no, I don't have this.
Oh, well, you do.
We have Salesforce
foundations.
For enterprise
customers and above,
you just upgrade for free.
And you get sales.
You get marketing.
You get service.
You get commerce.
You get Data Cloud.
And you get Agentforce
with 2,000 conversations
built in to get
started for free.
We want you to understand
the power of Agentforce.
We want you to
understand that today.
But before you leave
the room and go out,
yesterday, I
had the pleasure
of going to this thing
we call a hackathon.
So we had a bunch of
people coming and hacking
Agentforce, trying
to build agents
all over the place.
Christophe Coenraets
here ran that.
Christophe, come on up.
[APPLAUSE]
Christophe-- I walked
into the room-- normally,
I like people to
say hi to me when
I come into the room.
And I came in, and
everyone was like,
heads down.
And they're like
working hard.
And I'm like,
what's going on?
Like, don't they
want to say hi to me?
Well--
What were they doing?
Parker, they were hacking.
We, in fact, had 45 teams
competing for a $20,000
grand prize.
And all these teams
built amazing agents
with Agentforce.
But we have a winner.
And the winner built
an incredible event
coordinator agent.
So please, give it
up for the winner
of the Agentforce
hackathon, Team 4DS.
All right.
Team 4DS.
All right.
Congratulations.
You can't get this yet.
Sorry, don't try
to take the check.
Do not take this check.
I hope you're not
flying anywhere
because this
check may not fit
in the overhead
compartment.
But $20,000,
that's pretty cool.
What are you going
to do with $20,000?
You don't have
a mic, but I'll
tell them what you said.
Build more agents.
Build more agents.
Actually, he said
buy more Agentforce
is what he said.
He's going to
build more agents.
Well, congratulations.
And I love that it's an
employee-facing agent.
And I don't know
if you noticed,
you can actually light
that up in Slack.
Did you see that
demo earlier?
Yeah, it's perfect.
So I want you to
light it up in Slack.
Congratulations.
It's incredible.
Were you up
till late hours?
Midnight?
No.
No, it was easy, huh?
Easy peasy.
Easy peasy.
All right.
Well, we're going to
let you take this check.
Why don't we
get a picture?
Yeah, we need a picture.
Christophe, you go
on the other side
because this
is pretty cool.
This is a big
memory for you guys.
All right.
Big shout out
to the team 4DS,
winner of the hackathon.
[INAUDIBLE]
[APPLAUSE]
This is pretty cool.
All right, guys,
congratulations.
All right.
Well, thank you so much
for being here Agentforce
World Tour.
Thank you for
everyone online
who's been online
this whole time.
I forgot to give a
shout out to you.
Thank you for being here.
There's more ahead.
There's a talk by
Sanjay Gupta, future
of health and technology.
You can build
your first agent.
Go out and build
an agent right now.
Please, download the
app and please give us
feedback on the keynote.
We love feedback.
And thank you very much,
World Tour New York.
[APPLAUSE, CHEERING]
[UPBEAT MUSIC]
Resume AutoScroll
Questions about Salesforce+? We have answers.
What is Salesforce+?
Salesforce+ is the free streaming service for business. It’s a place to watch live events and original series that help you grow your career and company, on your own time. Salesforce+ is produced by Salesforce, so you’ll find our company’s big events like Dreamforce here — as well as new original shows and virtual learning made for your role and industry. Everything is free and unlimited.
What can I watch on Salesforce+?
You can watch all kinds of helpful, entertaining, and inspiring videos on Salesforce+. One thing our members love is our events. Whether you catch them live or on demand, Salesforce+ gives you a free front-row seat to our biggest global events, like Dreamforce, TrailblazerDX, and Connections. Our members are also big fans of our original series. These entertaining business shows star people in roles and industries just like yours, giving you fresh ideas for how to grow your company and do better work. The best part is, it’s all free.
Do I have to pay for Salesforce+?
Salesforce+ is free. Watch any Salesforce+ events or series at no cost. To enjoy unlimited free access to our entire library of videos, become a member by signing up for Salesforce+ with a Trailblazer account. You don’t need a Trailblazer account to watch original series. To join us for virtual events like Dreamforce or Connections, you need to create your own Trailblazer account (it’s free!).
What is a Trailblazer account?
Your Trailblazer account gives you access to a unified profile across the Salesforce ecosystem and allows you to manage your settings and access across multiple Salesforce sites.
How do I sign up to register for live and on-demand events, like Dreamforce?
Sign up or log in to your Trailblazer account and complete a few additional details to watch Salesforce+ events live and on demand. We recommend preregistering for all events on Salesforce+ to get your own personalized journey, reminders, recommendations, and more.
Do I need to be a Salesforce customer to watch Salesforce+?
You don’t need to be a Salesforce customer to watch Salesforce+. Our virtual learning is open to everyone.
