October 7, 2025 • 20 mins
Customers don’t just want answers. They want action! That’s why agentic AI, the kind that doesn’t just recommend but actually does the work, is reshaping how teams analyze experiences, resolve issues, and connect systems.
In this CX Pulse Check, Jeannie talks with Medallia’s Chief Strategy Officer, Sid Banerjee, about today’s clearest CX use cases and the pitfalls leaders should avoid as agentic AI adoption accelerates across industries like banking, healthcare, retail, and hospitality.
We explore three big patterns:
- AI that speeds up insight generation across calls, chats, and surveys
- Assistive tools that help agents find and deliver answers in real time
- Orchestration that lets AI schedule, submit, and update without piling on handoffs
Voice is making a comeback too, from car dealerships booking recall appointments to restaurants taking phone orders with flawless recall. The key isn’t “more bots,” but better design, continuous coaching, and a balance of speed with trust.
We also cover empathy, accountability, and practical guardrails (like building feedback loops, retraining models, and giving customers clear choices so they never feel trapped by tech).
If you’re evaluating where to deploy AI next or whether your pilots are effective, this conversation provides examples, cautions, and a pragmatic perspective.
About Sid Banerjee, Chief Strategy Officer, Medallia:
Over a nearly 30-year career, Sid has amassed deep skills building companies and solutions based on customer experience, business intelligence, and AI-powered technologies. Sid was the Founder and served as CEO, Chairman, and Chief Strategy Officer at Clarabridge. He most recently served as Chief XM Strategy Officer at Qualtrics. He presently serves as Chairman at Prolific, a high-growth technology company that provides a platform for high-quality, human-derived AI and research models. He was a founding employee of MicroStrategy and founder/CEO of Claraview. He also held executive and management positions at Ernst & Young and Sprint. He holds a BS/MS in Electrical Engineering from MIT.
Learn More About Medallia at https://www.medallia.com/
Follow Sid on...
LinkedIn: https://www.linkedin.com/in/sidbanerjeewdc/
Articles Mentioned:
- Marriott checks out AI agents amid technology transformation (CIO Dive) -- https://www.ciodive.com/news/marriott-international-AI-strategy-agentic-cloud-cybersecurity/758922/
- Red Lobster Turns to AI to Power Phone Ordering (Hospitality Technology) -- https://hospitalitytech.com/news-briefs/2025-09-22?article=red-lobster-turns-ai-power-phone-ordering
- Long Island pizzeria goes viral with saucy videos giving bad diners a taste of their own medicine: ‘Go sit down’ (New York Post) -- https://nypost.com/2025/09/22/us-news/long-island-pizzeria-goes-viral-with-saucy-videos-giving-bad-diners-a-taste-of-their-own-medicine-go-sit-down/
Want to ask a question? Visit askjeannie.vip to leave Jeannie a voicemail! (And don't forget to follow Jeannie on LinkedIn! www.linkedin.com/in/jeanniewalters/)
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Jeannie Walters (00:10):
Welcome to my favorite episode of the month,
CX Pulse Check, where I talk toa smart co-host who helps me
break down some of the top CXstories of the last month or so.
Now, I am so excited about thisbecause my special co-host, who
you should be excited abouttoo, is none other than our
friend Sid Banerjee fromMedallia, Chief Strategy
(00:34):
Officer.
Sid, we're so happy you're herewith us today.
Sid Banerjee (00:36):
Thank you, Jeannie.
It's great to be here.
Always a pleasure to chat withyou.
Jeannie Walters (00:40):
Thank you.
Well, first of all, if peoplesomehow don't know exactly who
you are and what you've done,can you share a little bit more
about where you fit in thisworld of CX?
Sid Banerjee (00:51):
Sure.
Well, I've been a long-term,long time, I guess, CX
practitioner.
Um, currently serving as thechief strategy officer here at
Medallia, but um, over about a20-year career, I've started CX
companies, built them, grownthem, sold them, uh, really done
a whole bunch of things, uh,primarily in the technology
space, building software thathelps people improve customer
(01:13):
experiences.
Jeannie Walters (01:13):
Excellent.
Well, I'm excited about thisconversation because when we
talk offline, we always have agreat time.
So I figured it was it was agood idea to share with
everybody here.
And you're in such aninteresting space because
there's so much going on aroundjust how we listen to customers,
how we tap into really whatthey want and think, even when
(01:34):
they don't tell us that.
And then we're also in thisworld of AI and how much we can
do, how we can be more proactivein those customer journeys.
And so one of the first thingsI wanted to uh explore with you
today was one of the headlinesthat I saw off of this is off of
CIO Dive.
(01:55):
And the headline is banksaccelerate AI deployments as
agentic tools gain traction.
And what stood out to me aboutthis is if you dive into the
article, what is reallyinteresting is that one of the
explorations that are happeningis we're really figuring out how
far can we put AI into thebusiness.
(02:17):
And in this story, they talkedabout how in banks there are up
to two-thirds of employees nowwho have access to these tools.
So I'm just curious about youknow, what are you seeing in
this world and how can we kindof embrace this new world of
agentic AI and everything else?
Sid Banerjee (02:35):
Yeah, it's it's actually a crazy dynamic uh
transformation going on inbanks, but not just in banks, in
retail organizations andhealthcare organizations.
I'm seeing AI show up in three,maybe three or four discrete
use cases.
The first is um as a tool forcompany employees to better
(02:55):
understand how to kind of getbetter at their jobs, to learn
from experience, to get coachingtools.
The second is AI is being usedto help a particularly
customer-facing employee, butalso a back office-facing
employee find answers toquestions or kind of respond to
interactions or experiences inthe moment when they're trying
to solve a problem or they're onthe phone or they're trying to
(03:17):
basically exercise on a workflowin some kind of capacity as a
customer service employee.
The third is AI is being usedafter the fact to analyze, to
summarize, to recommend ways toimprove things.
It could be, you know, feedbackon how to build a better
product or how to improve aservice experience.
And then the fourth is justusing it to actually connect the
(03:37):
dots, right?
Historically, connectivity wasdone through traditional
software and coding where youbuilt rules and you did things.
And now agentic AI is actuallydoing that connecting for you.
And it's a it's it's a kind ofan exploding laboratory of uh
kind of just ideas andevolutions.
Some of it's being builtin-house, some of it's being
used uh from companies that arebuilding solutions.
(03:59):
I don't think we've seen astabilization yet, but I can
tell you that pretty much everytime I talk to a customer that's
looking at CX, they areexploring one or more or all of
those use cases in theirenterprises today.
Jeannie Walters (04:14):
And I think what's so exciting about it is
that once we start testing itand piloting it, we start seeing
even more and more use cases.
That's that's what I see.
And and part of that is becausethe employees, the people who
are in the trenches every daydoing the work, they go, wait,
couldn't I use it in this way?
And then it opens up a wholenew avenue.
And I think that's somethingthat the pace of that innovation
(04:37):
is really exciting andchallenging at the same time.
Sid Banerjee (04:41):
Yeah, and then I think you know, there there are
also organizations trying tofigure out at what point am I
giving too much up to AI?
Am I using AI in a way that'sactually improving business
outcomes and customerexperiences, or am I just
creating unnecessary complexity?
So I do think we're gonna gothrough a phase of trying and
learning and improving and alsodiscarding you know uses of AI
(05:02):
that don't work very well.
But there's no question that,you know, to use the the
overused phrase, the genie's thegenie's out of the bag.
Um, and uh there's no puttingit back, right?
I mean, it's it's it's it's atechnology that I think is
actually driving tremendousimprovement in the way people
can be more responsive tocustomer experience and being
more adaptive and learning fromcustomer interactions as well.
Jeannie Walters (05:24):
Yeah, and I think there's a lot around the
scale of this too, where youknow, there are there are times
in the past I remember workingwith like incredible business
analysts, and you'd be like,wow, how did you find that
pattern after only four months,right?
Like you'd and now we can do itin 24 hours, 12 hours, uh 30
seconds sometimes.
Sid Banerjee (05:44):
Yeah, I mean, I'll give you an example.
One of the things that I'vesort of concluded in looking at
AI in and customer experiencesthat there's really three things
right now that people are usingit a lot for.
The first is it is a way ofaccelerating um analysis, right?
Taking a lot of interactions, alot of uh experiences, and
saying, what are the goodexperiences, what are the bad
(06:04):
experiences, what are the causesof the good and the bad
experiences, and ultimatelyhelping uh a customer experience
practitioner or analyst figureout what to focus on fixing,
what to focus on continuing todo as well as they're doing, et
cetera.
The second big use case isusing AI to help um automate or
at least streamline theexperience of responding to
(06:25):
customers and to experiences andto inquiries or problems.
It is very good at findinganswers, it's very good at
suggesting ways to kind ofbasically respond to the
customer in the moment.
And then the third is in thatarea of connecting the uh the
different systems that touch oncustomer experiences to create
what you know, what people callagentic AI, right?
(06:48):
An agentic AI is not just anagent.
I like to say agentic AI is anAI that has agency, it can do
things, right?
So it will actually recognizethat when someone needs
something done, instead oftelling the human how to do it,
it'll actually do it itself.
And those are the three big usecases I think that are going to
be dramatically transforming umcustomer experiences.
(07:08):
I see it in, like I say, I seeit in banks automating inquiries
and balance transfers andapplying for credit cards.
I see it in healthcare withinsurance companies looking at
ways to uh to automate claimsinvestigations, reimbursements,
payments, you know, receivingpayments, et cetera.
And even in retail, you'restarting to see people using AI
(07:29):
to uh do a lot more onlineshopping, online servicing,
online receiving, or hybridexperiences where you might
start online and then picksomething up.
And the connectivity isAI-based nowadays, it's not more
static or or kind ofineffective business rules
because it's just more dynamicand more empathetic, and the
experience is better.
Jeannie Walters (07:48):
I love the way you say that it's AI with agency
because I think that givescontext to it in a different
way.
Because I know there are peoplewho just think it's one-to-one,
like, oh, you had an agent, nowyou have AI.
And that's not really whatwe're talking about.
It's so much more robust andconnected and all of those
things.
But well, you set us upperfectly for the next topic
here, which we didn't even planthat, but here we are, uh, where
(08:11):
you know we talked about allthe different industries that
are using this.
This is from hospitalitytechnology, and it's one of
their news briefs.
And it is really um, theheadline is Red Lobster turns to
AI for power phone ordering.
And so this is really about umphone ordering agents, quote
(08:33):
unquote, uh, as well as justusing voice in different ways.
And I think that there this issomething that is on the
horizon, maybe coming fasterthan people realize as well.
And so, you know, this is justanother angle where this
technology sounds like it'sgoing to make things so
efficient and convenient, bothfor the restaurant workers and
(08:55):
the customer.
But if we're not careful and ifwe don't continue to ask
customers, is this working foryou?
And how can it work better?
We might be disconnectingthere.
That's one of the things that Ithink we need to start maybe
highlighting for people is onceyou roll out this technology,
you still have to have thoseconversations.
So, what's your take on that?
Sid Banerjee (09:16):
I think that if you went back four or five years
ago, people were predicting thedeath of the telephonic
experience prematurely, right?
Um, people wanted everyone topick up their mobile phones and
order on their phones and doeverything online on websites.
And certainly many of us havedone that.
That's a big part of how we docustomer experience today.
(09:37):
But there are still a lot of uhexperiences that are pretty
effectively done on the phone.
And when you add agentic AI andyou know, voice uh AI into
phone experiences, you can do alot of things, a lot of things
better than you could even witha human-to-human conversation.
I'll give you an example beforeI go to Red Lobster, but a
friend of mine is starting acompany that actually he's
(10:00):
selling AI voice bots to cardealers.
And the business problem he'strying to solve is you know,
when you try to call a cardealer to schedule an
appointment or to maybe askabout a test drive, half the
time you go to the oneoverworked person in the center
of the car, you know, the cardealership who doesn't always
pick up the phone becausethey're getting phone calls all
the time.
And so there's a response lagto get the call back, to
(10:22):
schedule the thing.
This thing is essentially avoice bot.
It picks up all the time, ittakes your information, it
schedules your car repair, itcan look you up and say, Yeah, I
see you have a XYZ brand andyou can schedule it.
Or it can say, here, there's anappointment opening here.
And that means every time youcall, you get what you want
done, done, right?
(10:42):
When you're ordering food fromRed Lobster, you might want to
ask questions, you want to knowif the hush puppies come with
the lobster.
If you have an agentic AI,they're gonna know everything,
they're gonna have instantrecall of the menu.
They're not gonna be overworkedtrying to deal with answering
the phone and working the cashregister.
It's an effective experience.
And arguably, if those voicebots are set up correctly and
(11:06):
they're agentically designed,they're actually gonna be a
better experience than you mighthave talking to an overworked
employee in a restaurant.
So I don't think there'sanything wrong with that.
I actually think it's a greatuse case, and I think you're
gonna see more voiceinteractions coming back into
customer experiences powered byby AI, very likely.
Jeannie Walters (11:23):
Well, I love that dealership example because
we literally just went throughthis where they contacted us
about a recall thing and we hadto go in and it took forever
because you know, you've got towell, that service person isn't
here, or we've got to call youback, or those aren't our
service hours, and so we had tokeep calling, just kind of
hoping for the best.
(11:44):
So I love that use case.
And I think the whole idea ofyou know, customers are looking
for convenience, and sometimesthe convenience is when you're
in your car and you want tocall, and you can ask your voice
assistant in the car to makethat call for you.
And so if you can't get whatyou need through that, then you
(12:04):
it's that it's that open loop inyour brain of oh gosh, okay,
I've got to remember to schedulethis, or I've got to ask about
the hush puppies when I go in.
Sid Banerjee (12:12):
And by the way, you don't want to be using your
phone when you're drivingeither.
So there's another good reasonfor doing this, right?
You much rather be using thevoice, exactly.
Jeannie Walters (12:18):
Yeah, I I'm still making I still see
websites that don't have clickto call on their phone numbers.
I'm like, it's 2025.
Like we should agree that thisis the standard, right?
Sid Banerjee (12:29):
Completely
Well, and I think that again,we have to remember that like
this is a cooperation, this is acollaboration with our
customers, and so that meansthat we have to make sure that
we're collaborating with thetechnology on their behalf, we
are continuing to check in withthem, we are making sure that
when they tell us things, we usethose insights and we respond
(12:51):
to them.
Because if we don't see it thatway, it can become very
one-sided and it can feel on ourend like everything's working
fine.
And we don't know that thesecustomers want to call even.
And let me, yeah, let me kind of touch on that
point.
When you do replace a humanfrom the company side with it
with an AI to handle customerexperiences over the phone, that
(13:13):
doesn't um negate the need todo quality assurance, right?
And to do continuousimprovement.
Because just as a human canalways be coached and trained to
get better at their jobs, socan an AI, right?
I mean, fundamentally, an AIbot gets trained on experiences,
then you may want to analyzehow it's doing, and then you may
want to coach it or effectivelytrain it or retrain it to get
(13:35):
better.
And a big part of any AIinvestment should be doing the
customer experience analysisafter the fact and even during
the fact to make sure thatyou're making sure that you're
generating better experiences,not worse, and that you're
continuing to get it make itbetter as business conditions
change, you know, customerpreferences change, other things
can change.
(13:55):
So we, you know, coming from abusiness, obviously where I do
with Medallia, where there'scustomer experience analytics,
is a big part of how we thinkabout improving customer
experiences.
Deploying AI does not make thecustomer experience improvement
kind of mandate go away.
If anything, it just meansexpanding that aperture to make
sure the AI is continuing togrow and get better as well.
Jeannie Walters (14:14):
I think that's a very important point that we
need to coach and traincontinuously, even if it's you
know, we can't trust that eventhe information will all stay
accurate if we're not doing thatquality assurance just like we
would at any other part of thebusiness and with the people
involved as well.
So it's a really, really goodpoint.
And so this last story, it's alittle goofy, Sid, but I liked
(14:37):
it.
Sid Banerjee (14:38):
Uh so I'm laughing because you told me about it
ahead of time, but go ahead.
Jeannie Walters (14:42):
I did.
So um, and the site is a littlechallenging because even the
news source is one that Iwouldn't typically pick, I will
be honest.
But this is from the New YorkPost, which I don't quote very
often, but the headline is LongIsland pizzeria goes viral with
saucy videos giving bad diners ataste of their own medicine.
(15:04):
And there's a quote that says,Go sit down.
And the whole um kind of anglehere is that they were getting
fed up with bad customerbehavior at this local pizzeria.
It's called Phil's Pizzeria inLong Island, and they started
making these uh skitsessentially and filming them of
bad customer behavior.
(15:25):
And the owner and some otherpeople, the employees there,
they do these skits and theyswear in them and they say
things like, What are you doing?
You're wasting my time, and uh,it's very New York, it's great.
But what struck me about it wasby starting this, they actually
have this now really good viralrun.
They're getting tons offollowers on TikTok and
(15:46):
Instagram and everything.
And so people are actuallyphysically seeking out the
restaurant now when they're inLong Island because they're
sassy and they're being kind ofin you know irreverent about it.
And it just struck me that Ithink part of what this whole
conversation around AI andtechnology and everything else
that is so prevalent andeverything, I think customers
(16:07):
are also craving thatauthenticity, they're craving
just that one-to-one.
So I don't know if you had achance to watch any of the
videos, they're not reallysuitable for work, so I'm not
gonna share any here, but golook those up on your own time,
listener.
Sid Banerjee (16:22):
I will definitely, I I read the uh the headline,
but I did not get a chance tosample the videos.
So, my um my reaction to thatis I guess two twofold.
One, um, with every newtechnology, there is an
opportunity for you know the uhthe outliers, the crazy people,
the trolls to take advantage ofit to uh to maybe abuse or
(16:44):
harass or potentially do crazythings.
And I think as an organization,if you're a business, the best
thing you can do is not take anyof that too seriously.
Obviously, use feedback toimprove things, but there's a
point at which things getludicrous, and I think the best
thing you can do is not not gettoo upset or too crazy and just
(17:05):
have a little bit of fun withit.
I'm I was reminded as uh as youhad shared that that article
with me that um the same kind ofthing happened a few years ago
with Twitter or now X.
Jimmy Kimmel famously wouldhave celebrities on to read mean
tweets about themselves, right?
And at some point you just haveto be like, yeah, what are you
gonna do?
Right, you use the technologyto improve the world, but if
(17:27):
people want to do silly thingswith it, just don't take it too
seriously.
And I think that does create asense of authenticity for, you
know, in that case, thecelebrity, in this case, the the
restaurant, which actually isrefreshing.
It means you know, you can youcan take yourself you know not
too seriously when when thingsgo a little crazy.
Jeannie Walters (17:44):
That's true, and I think that's a really good
message for today's world.
Just there are always going tobe the complainers and the
trolls, and you know, I Iproduce a lot of content on
LinkedIn, and usually I getgreat conversation there.
And a couple months ago, therewas somebody who was just
throwing bombs all the time, andI was kind of like, What did I
do to this person?
And I was taking it verypersonally and I was trying to
(18:06):
engage, and I showed what washappening to one of my kids who
are now, you know, young adults,and literally without
hesitation, they both said,Well, that's a troll, so just
block that person.
Like, what's the problem?
Because it happens, it's justpart of their world.
And I think that it was a goodreminder to your point not to
(18:26):
take it so seriously, and alsorealize that there are people
who literally wake up in themorning and that is their plan
to troll, to complain.
Sid Banerjee (18:36):
Yeah, this is their purpose in life for that
day, at least.
But you know, I think it'simportant also, you know, the
the serious message here is youwill get feedback whenever you
use a new technology.
You know, I would imagine thatwhen you're talking to an AI
voice bot, because it's not ahuman, if you know it's not a
human, you as a as a customermay actually be a little bit
(18:58):
more, let's just say, direct ormaybe even obscene, right?
And and if I'm a company havingto process all those recordings
and and and calls, I'm gonnahave to not take that too
seriously.
I have to sift through theconstructive feedback with the
crazy feedback.
And that's just the nature oftechnology nowadays, for better
or worse.
Jeannie Walters (19:14):
So that's a very good point, too.
Yeah, we have to remember thecontext of the people who are
you know engaging and they mightthink they're behind a screen,
they're not acting exactly likethey would in the real world.
Sid Banerjee (19:27):
That's right.
Jeannie Walters (19:28):
So, well, this was fascinating.
I'm so happy that we we got youon here, that you had the time
to explore all these things withme.
And I know I will see you inFebruary at Experience in Las
Vegas
Sid Banerjee (19:39):
This year in Vegas, looking forward to it.
And yeah, I appreciate theopportunity to get a little bit
off color, if you will, and uhand talk about some interesting
um but but unique perspectiveson customer experience.
So thank you.
Jeannie Walters (19:51):
Well, thank you for being here, and thanks for
everything you do for ourindustry too.
There's a lot going on, and Ithink you uh continue to be a
leader there.
So we appreciate that as well.
Sid Banerjee (20:00):
Likewise, likewise.
Thank you so much.
Jeannie Walters (20:02):
Thank you.
And thank you, everybody who ishere with us, watching,
listening, subscribing to theExperience Action Podcast.
I always love hearing from you.
Don't forget, you can leave mea voicemail at askjeannie.vip.
And I might answer it for youhere on the podcast.
Thanks to Sid for being here.
Thanks to you, and I will seeyou next time.
Welcome to my favorite episode of the month,
CX Pulse Check, where I talk toa smart co-host who helps me
break down some of the top CXstories of the last month or so.
Now, I am so excited about thisbecause my special co-host, who
you should be excited abouttoo, is none other than our
friend Sid Banerjee fromMedallia, Chief Strategy
(00:34):
Officer.
Sid, we're so happy you're herewith us today.
Sid Banerjee (00:36):
Thank you, Jeannie.
It's great to be here.
Always a pleasure to chat withyou.
Jeannie Walters (00:40):
Thank you.
Well, first of all, if peoplesomehow don't know exactly who
you are and what you've done,can you share a little bit more
about where you fit in thisworld of CX?
Sid Banerjee (00:51):
Sure.
Well, I've been a long-term,long time, I guess, CX
practitioner.
Um, currently serving as thechief strategy officer here at
Medallia, but um, over about a20-year career, I've started CX
companies, built them, grownthem, sold them, uh, really done
a whole bunch of things, uh,primarily in the technology
space, building software thathelps people improve customer
(01:13):
experiences.
Jeannie Walters (01:13):
Excellent.
Well, I'm excited about thisconversation because when we
talk offline, we always have agreat time.
So I figured it was it was agood idea to share with
everybody here.
And you're in such aninteresting space because
there's so much going on aroundjust how we listen to customers,
how we tap into really whatthey want and think, even when
(01:34):
they don't tell us that.
And then we're also in thisworld of AI and how much we can
do, how we can be more proactivein those customer journeys.
And so one of the first thingsI wanted to uh explore with you
today was one of the headlinesthat I saw off of this is off of
CIO Dive.
(01:55):
And the headline is banksaccelerate AI deployments as
agentic tools gain traction.
And what stood out to me aboutthis is if you dive into the
article, what is reallyinteresting is that one of the
explorations that are happeningis we're really figuring out how
far can we put AI into thebusiness.
(02:17):
And in this story, they talkedabout how in banks there are up
to two-thirds of employees nowwho have access to these tools.
So I'm just curious about youknow, what are you seeing in
this world and how can we kindof embrace this new world of
agentic AI and everything else?
Sid Banerjee (02:35):
Yeah, it's it's actually a crazy dynamic uh
transformation going on inbanks, but not just in banks, in
retail organizations andhealthcare organizations.
I'm seeing AI show up in three,maybe three or four discrete
use cases.
The first is um as a tool forcompany employees to better
(02:55):
understand how to kind of getbetter at their jobs, to learn
from experience, to get coachingtools.
The second is AI is being usedto help a particularly
customer-facing employee, butalso a back office-facing
employee find answers toquestions or kind of respond to
interactions or experiences inthe moment when they're trying
to solve a problem or they're onthe phone or they're trying to
(03:17):
basically exercise on a workflowin some kind of capacity as a
customer service employee.
The third is AI is being usedafter the fact to analyze, to
summarize, to recommend ways toimprove things.
It could be, you know, feedbackon how to build a better
product or how to improve aservice experience.
And then the fourth is justusing it to actually connect the
(03:37):
dots, right?
Historically, connectivity wasdone through traditional
software and coding where youbuilt rules and you did things.
And now agentic AI is actuallydoing that connecting for you.
And it's a it's it's a kind ofan exploding laboratory of uh
kind of just ideas andevolutions.
Some of it's being builtin-house, some of it's being
used uh from companies that arebuilding solutions.
(03:59):
I don't think we've seen astabilization yet, but I can
tell you that pretty much everytime I talk to a customer that's
looking at CX, they areexploring one or more or all of
those use cases in theirenterprises today.
Jeannie Walters (04:14):
And I think what's so exciting about it is
that once we start testing itand piloting it, we start seeing
even more and more use cases.
That's that's what I see.
And and part of that is becausethe employees, the people who
are in the trenches every daydoing the work, they go, wait,
couldn't I use it in this way?
And then it opens up a wholenew avenue.
And I think that's somethingthat the pace of that innovation
(04:37):
is really exciting andchallenging at the same time.
Sid Banerjee (04:41):
Yeah, and then I think you know, there there are
also organizations trying tofigure out at what point am I
giving too much up to AI?
Am I using AI in a way that'sactually improving business
outcomes and customerexperiences, or am I just
creating unnecessary complexity?
So I do think we're gonna gothrough a phase of trying and
learning and improving and alsodiscarding you know uses of AI
(05:02):
that don't work very well.
But there's no question that,you know, to use the the
overused phrase, the genie's thegenie's out of the bag.
Um, and uh there's no puttingit back, right?
I mean, it's it's it's it's atechnology that I think is
actually driving tremendousimprovement in the way people
can be more responsive tocustomer experience and being
more adaptive and learning fromcustomer interactions as well.
Jeannie Walters (05:24):
Yeah, and I think there's a lot around the
scale of this too, where youknow, there are there are times
in the past I remember workingwith like incredible business
analysts, and you'd be like,wow, how did you find that
pattern after only four months,right?
Like you'd and now we can do itin 24 hours, 12 hours, uh 30
seconds sometimes.
Sid Banerjee (05:44):
Yeah, I mean, I'll give you an example.
One of the things that I'vesort of concluded in looking at
AI in and customer experiencesthat there's really three things
right now that people are usingit a lot for.
The first is it is a way ofaccelerating um analysis, right?
Taking a lot of interactions, alot of uh experiences, and
saying, what are the goodexperiences, what are the bad
(06:04):
experiences, what are the causesof the good and the bad
experiences, and ultimatelyhelping uh a customer experience
practitioner or analyst figureout what to focus on fixing,
what to focus on continuing todo as well as they're doing, et
cetera.
The second big use case isusing AI to help um automate or
at least streamline theexperience of responding to
(06:25):
customers and to experiences andto inquiries or problems.
It is very good at findinganswers, it's very good at
suggesting ways to kind ofbasically respond to the
customer in the moment.
And then the third is in thatarea of connecting the uh the
different systems that touch oncustomer experiences to create
what you know, what people callagentic AI, right?
(06:48):
An agentic AI is not just anagent.
I like to say agentic AI is anAI that has agency, it can do
things, right?
So it will actually recognizethat when someone needs
something done, instead oftelling the human how to do it,
it'll actually do it itself.
And those are the three big usecases I think that are going to
be dramatically transforming umcustomer experiences.
(07:08):
I see it in, like I say, I seeit in banks automating inquiries
and balance transfers andapplying for credit cards.
I see it in healthcare withinsurance companies looking at
ways to uh to automate claimsinvestigations, reimbursements,
payments, you know, receivingpayments, et cetera.
And even in retail, you'restarting to see people using AI
(07:29):
to uh do a lot more onlineshopping, online servicing,
online receiving, or hybridexperiences where you might
start online and then picksomething up.
And the connectivity isAI-based nowadays, it's not more
static or or kind ofineffective business rules
because it's just more dynamicand more empathetic, and the
experience is better.
Jeannie Walters (07:48):
I love the way you say that it's AI with agency
because I think that givescontext to it in a different
way.
Because I know there are peoplewho just think it's one-to-one,
like, oh, you had an agent, nowyou have AI.
And that's not really whatwe're talking about.
It's so much more robust andconnected and all of those
things.
But well, you set us upperfectly for the next topic
here, which we didn't even planthat, but here we are, uh, where
(08:11):
you know we talked about allthe different industries that
are using this.
This is from hospitalitytechnology, and it's one of
their news briefs.
And it is really um, theheadline is Red Lobster turns to
AI for power phone ordering.
And so this is really about umphone ordering agents, quote
(08:33):
unquote, uh, as well as justusing voice in different ways.
And I think that there this issomething that is on the
horizon, maybe coming fasterthan people realize as well.
And so, you know, this is justanother angle where this
technology sounds like it'sgoing to make things so
efficient and convenient, bothfor the restaurant workers and
(08:55):
the customer.
But if we're not careful and ifwe don't continue to ask
customers, is this working foryou?
And how can it work better?
We might be disconnectingthere.
That's one of the things that Ithink we need to start maybe
highlighting for people is onceyou roll out this technology,
you still have to have thoseconversations.
So, what's your take on that?
Sid Banerjee (09:16):
I think that if you went back four or five years
ago, people were predicting thedeath of the telephonic
experience prematurely, right?
Um, people wanted everyone topick up their mobile phones and
order on their phones and doeverything online on websites.
And certainly many of us havedone that.
That's a big part of how we docustomer experience today.
(09:37):
But there are still a lot of uhexperiences that are pretty
effectively done on the phone.
And when you add agentic AI andyou know, voice uh AI into
phone experiences, you can do alot of things, a lot of things
better than you could even witha human-to-human conversation.
I'll give you an example beforeI go to Red Lobster, but a
friend of mine is starting acompany that actually he's
(10:00):
selling AI voice bots to cardealers.
And the business problem he'strying to solve is you know,
when you try to call a cardealer to schedule an
appointment or to maybe askabout a test drive, half the
time you go to the oneoverworked person in the center
of the car, you know, the cardealership who doesn't always
pick up the phone becausethey're getting phone calls all
the time.
And so there's a response lagto get the call back, to
(10:22):
schedule the thing.
This thing is essentially avoice bot.
It picks up all the time, ittakes your information, it
schedules your car repair, itcan look you up and say, Yeah, I
see you have a XYZ brand andyou can schedule it.
Or it can say, here, there's anappointment opening here.
And that means every time youcall, you get what you want
done, done, right?
(10:42):
When you're ordering food fromRed Lobster, you might want to
ask questions, you want to knowif the hush puppies come with
the lobster.
If you have an agentic AI,they're gonna know everything,
they're gonna have instantrecall of the menu.
They're not gonna be overworkedtrying to deal with answering
the phone and working the cashregister.
It's an effective experience.
And arguably, if those voicebots are set up correctly and
(11:06):
they're agentically designed,they're actually gonna be a
better experience than you mighthave talking to an overworked
employee in a restaurant.
So I don't think there'sanything wrong with that.
I actually think it's a greatuse case, and I think you're
gonna see more voiceinteractions coming back into
customer experiences powered byby AI, very likely.
Jeannie Walters (11:23):
Well, I love that dealership example because
we literally just went throughthis where they contacted us
about a recall thing and we hadto go in and it took forever
because you know, you've got towell, that service person isn't
here, or we've got to call youback, or those aren't our
service hours, and so we had tokeep calling, just kind of
hoping for the best.
(11:44):
So I love that use case.
And I think the whole idea ofyou know, customers are looking
for convenience, and sometimesthe convenience is when you're
in your car and you want tocall, and you can ask your voice
assistant in the car to makethat call for you.
And so if you can't get whatyou need through that, then you
(12:04):
it's that it's that open loop inyour brain of oh gosh, okay,
I've got to remember to schedulethis, or I've got to ask about
the hush puppies when I go in.
Sid Banerjee (12:12):
And by the way, you don't want to be using your
phone when you're drivingeither.
So there's another good reasonfor doing this, right?
You much rather be using thevoice, exactly.
Jeannie Walters (12:18):
Yeah, I I'm still making I still see
websites that don't have clickto call on their phone numbers.
I'm like, it's 2025.
Like we should agree that thisis the standard, right?
Sid Banerjee (12:29):
Completely
Well, and I think that again,we have to remember that like
this is a cooperation, this is acollaboration with our
customers, and so that meansthat we have to make sure that
we're collaborating with thetechnology on their behalf, we
are continuing to check in withthem, we are making sure that
when they tell us things, we usethose insights and we respond
(12:51):
to them.
Because if we don't see it thatway, it can become very
one-sided and it can feel on ourend like everything's working
fine.
And we don't know that thesecustomers want to call even.
And let me, yeah, let me kind of touch on that
point.
When you do replace a humanfrom the company side with it
with an AI to handle customerexperiences over the phone, that
(13:13):
doesn't um negate the need todo quality assurance, right?
And to do continuousimprovement.
Because just as a human canalways be coached and trained to
get better at their jobs, socan an AI, right?
I mean, fundamentally, an AIbot gets trained on experiences,
then you may want to analyzehow it's doing, and then you may
want to coach it or effectivelytrain it or retrain it to get
(13:35):
better.
And a big part of any AIinvestment should be doing the
customer experience analysisafter the fact and even during
the fact to make sure thatyou're making sure that you're
generating better experiences,not worse, and that you're
continuing to get it make itbetter as business conditions
change, you know, customerpreferences change, other things
can change.
(13:55):
So we, you know, coming from abusiness, obviously where I do
with Medallia, where there'scustomer experience analytics,
is a big part of how we thinkabout improving customer
experiences.
Deploying AI does not make thecustomer experience improvement
kind of mandate go away.
If anything, it just meansexpanding that aperture to make
sure the AI is continuing togrow and get better as well.
Jeannie Walters (14:14):
I think that's a very important point that we
need to coach and traincontinuously, even if it's you
know, we can't trust that eventhe information will all stay
accurate if we're not doing thatquality assurance just like we
would at any other part of thebusiness and with the people
involved as well.
So it's a really, really goodpoint.
And so this last story, it's alittle goofy, Sid, but I liked
(14:37):
it.
Sid Banerjee (14:38):
Uh so I'm laughing because you told me about it
ahead of time, but go ahead.
Jeannie Walters (14:42):
I did.
So um, and the site is a littlechallenging because even the
news source is one that Iwouldn't typically pick, I will
be honest.
But this is from the New YorkPost, which I don't quote very
often, but the headline is LongIsland pizzeria goes viral with
saucy videos giving bad diners ataste of their own medicine.
(15:04):
And there's a quote that says,Go sit down.
And the whole um kind of anglehere is that they were getting
fed up with bad customerbehavior at this local pizzeria.
It's called Phil's Pizzeria inLong Island, and they started
making these uh skitsessentially and filming them of
bad customer behavior.
(15:25):
And the owner and some otherpeople, the employees there,
they do these skits and theyswear in them and they say
things like, What are you doing?
You're wasting my time, and uh,it's very New York, it's great.
But what struck me about it wasby starting this, they actually
have this now really good viralrun.
They're getting tons offollowers on TikTok and
(15:46):
Instagram and everything.
And so people are actuallyphysically seeking out the
restaurant now when they're inLong Island because they're
sassy and they're being kind ofin you know irreverent about it.
And it just struck me that Ithink part of what this whole
conversation around AI andtechnology and everything else
that is so prevalent andeverything, I think customers
(16:07):
are also craving thatauthenticity, they're craving
just that one-to-one.
So I don't know if you had achance to watch any of the
videos, they're not reallysuitable for work, so I'm not
gonna share any here, but golook those up on your own time,
listener.
Sid Banerjee (16:22):
I will definitely, I I read the uh the headline,
but I did not get a chance tosample the videos.
So, my um my reaction to thatis I guess two twofold.
One, um, with every newtechnology, there is an
opportunity for you know the uhthe outliers, the crazy people,
the trolls to take advantage ofit to uh to maybe abuse or
(16:44):
harass or potentially do crazythings.
And I think as an organization,if you're a business, the best
thing you can do is not take anyof that too seriously.
Obviously, use feedback toimprove things, but there's a
point at which things getludicrous, and I think the best
thing you can do is not not gettoo upset or too crazy and just
(17:05):
have a little bit of fun withit.
I'm I was reminded as uh as youhad shared that that article
with me that um the same kind ofthing happened a few years ago
with Twitter or now X.
Jimmy Kimmel famously wouldhave celebrities on to read mean
tweets about themselves, right?
And at some point you just haveto be like, yeah, what are you
gonna do?
Right, you use the technologyto improve the world, but if
(17:27):
people want to do silly thingswith it, just don't take it too
seriously.
And I think that does create asense of authenticity for, you
know, in that case, thecelebrity, in this case, the the
restaurant, which actually isrefreshing.
It means you know, you can youcan take yourself you know not
too seriously when when thingsgo a little crazy.
Jeannie Walters (17:44):
That's true, and I think that's a really good
message for today's world.
Just there are always going tobe the complainers and the
trolls, and you know, I Iproduce a lot of content on
LinkedIn, and usually I getgreat conversation there.
And a couple months ago, therewas somebody who was just
throwing bombs all the time, andI was kind of like, What did I
do to this person?
And I was taking it verypersonally and I was trying to
(18:06):
engage, and I showed what washappening to one of my kids who
are now, you know, young adults,and literally without
hesitation, they both said,Well, that's a troll, so just
block that person.
Like, what's the problem?
Because it happens, it's justpart of their world.
And I think that it was a goodreminder to your point not to
(18:26):
take it so seriously, and alsorealize that there are people
who literally wake up in themorning and that is their plan
to troll, to complain.
Sid Banerjee (18:36):
Yeah, this is their purpose in life for that
day, at least.
But you know, I think it'simportant also, you know, the
the serious message here is youwill get feedback whenever you
use a new technology.
You know, I would imagine thatwhen you're talking to an AI
voice bot, because it's not ahuman, if you know it's not a
human, you as a as a customermay actually be a little bit
(18:58):
more, let's just say, direct ormaybe even obscene, right?
And and if I'm a company havingto process all those recordings
and and and calls, I'm gonnahave to not take that too
seriously.
I have to sift through theconstructive feedback with the
crazy feedback.
And that's just the nature oftechnology nowadays, for better
or worse.
Jeannie Walters (19:14):
So that's a very good point, too.
Yeah, we have to remember thecontext of the people who are
you know engaging and they mightthink they're behind a screen,
they're not acting exactly likethey would in the real world.
Sid Banerjee (19:27):
That's right.
Jeannie Walters (19:28):
So, well, this was fascinating.
I'm so happy that we we got youon here, that you had the time
to explore all these things withme.
And I know I will see you inFebruary at Experience in Las
Vegas
Sid Banerjee (19:39):
This year in Vegas, looking forward to it.
And yeah, I appreciate theopportunity to get a little bit
off color, if you will, and uhand talk about some interesting
um but but unique perspectiveson customer experience.
So thank you.
Jeannie Walters (19:51):
Well, thank you for being here, and thanks for
everything you do for ourindustry too.
There's a lot going on, and Ithink you uh continue to be a
leader there.
So we appreciate that as well.
Sid Banerjee (20:00):
Likewise, likewise.
Thank you so much.
Jeannie Walters (20:02):
Thank you.
And thank you, everybody who ishere with us, watching,
listening, subscribing to theExperience Action Podcast.
I always love hearing from you.
Don't forget, you can leave mea voicemail at askjeannie.vip.
And I might answer it for youhere on the podcast.
Thanks to Sid for being here.
Thanks to you, and I will seeyou next time.
Popular Podcasts
My Favorite Murder with Karen Kilgariff and Georgia Hardstark
My Favorite Murder is a true crime comedy podcast hosted by Karen Kilgariff and Georgia Hardstark. Each week, Karen and Georgia share compelling true crimes and hometown stories from friends and listeners. Since MFM launched in January of 2016, Karen and Georgia have shared their lifelong interest in true crime and have covered stories of infamous serial killers like the Night Stalker, mysterious cold cases, captivating cults, incredible survivor stories and important events from history like the Tulsa race massacre of 1921. My Favorite Murder is part of the Exactly Right podcast network that provides a platform for bold, creative voices to bring to life provocative, entertaining and relatable stories for audiences everywhere. The Exactly Right roster of podcasts covers a variety of topics including historic true crime, comedic interviews and news, science, pop culture and more. Podcasts on the network include Buried Bones with Kate Winkler Dawson and Paul Holes, That's Messed Up: An SVU Podcast, This Podcast Will Kill You, Bananas and more.
24/7 News: The Latest
The latest news in 4 minutes updated every hour, every day.
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