Latest / The DEX Show: A Show for IT Change Makers / When AI meets automation w/ Charles Araujo

Transcript

When AI meets automation w/ Charles Araujo · The DEX Show: A Show for IT Change Makers

0:01
Welcome it, change Makers to the deck Show with Tim Flower and Tom McGraw.
0:06
Let's get into it. Hello, change Makers, it's another deck show
0:11
for you, and we're Tim and I here together again and we're very excited
0:17
to welcome one of our one of our favorite guests, one of our smartest
0:21
guests, and the guests from one of our most successful episodes of two thousand
0:26
and twenty two, although the episode was actually recorded at the very very end
0:32
of two thousand and twenty twenty one, and that's Charles or Ayo. Hi,
0:35
Charles, It's great to see you, Agamon, great to see you.
0:39
And I feel like a little bit like, you know, I'm one
0:42
one of those dates with a serial dater where it's like I com ond you
0:45
tell all the girls that no, no, no, we don't have we
0:48
don't we don't do second dates too often. Here I'm a deck show.
0:53
It was also it was also it was not only one of my favorite,
0:56
one of our favorite episodes, one of our listeners favorite episodes, more important,
1:00
one of the as I say, when the first popular last year,
1:02
but it was also it's also my favorite episode title Commodore sixty four and a
1:07
Screwdriver obviously referring if we record this for cocktail, not the not the mechanically.
1:15
You know, for all those reasons and more, childs, we have
1:18
you back on and listen. Just if anybody doesn't it hasn't heard the first
1:23
of your appearances. You are, you're a publisher for Digital Experience Report,
1:26
You're an analystous speaker, you're a best selling author, and I think always
1:30
a great way to start a conversation with your childs just to hear about what
1:34
you what you what you're thinking about at the moment, was your what's your
1:37
attention focused upon? So I guess you know I'm boring. It's the same
1:44
stuff, you know, it's it partially, it's this frustration I get.
1:48
I know we're probably gonna talk about it later because we can't help but talk
1:52
about things like chat, GPT and AYI and all the hypeend buzz, and
1:56
you know, our industry is just horrible. It's see, it's like we
2:00
go chase the next shiny thing. But if you want to be successful,
2:05
if you want to actually drive transformation, if you want to actually deliver outcomes,
2:10
it's almost always about the basics. It's the blocking and tackling and so
2:15
I'm still very focused on a lot of the basics. I just published a
2:20
piece of analysis talking about the fact that why data is so critical, and
2:27
yet there is a study actually by Salesforce that just got released a couple of
2:30
days ago that's saying, you know, like eighty some huge percentage of executives
2:34
totally agree with how important data is, and then how like astronomically small of
2:38
a percentage of organizations are actually using it in any meaningful way. And you
2:44
find this across every different spectrum. It's like we get all hyped up about
2:47
something, but the stuff that we should be doing, the blocking, tackling,
2:52
the things like organizational design. I mean, I can rant on forever
2:55
about these, right, the basic stuff that actually delivers results we just give
3:01
a lot of lip service to and don't put enough focused on. So I'm
3:04
still very much focused on these kind of the basics and just let's just actually
3:09
get this stuff done once and for all. And I'm talking about getting it
3:15
done. There was I remember, you know, obviously when we last spoke,
3:19
middle of the pandemic, even though it was kind of about to enter
3:22
the post pandemic era, I think fair to say, and there was all
3:27
this talk of COVID, the transformative effect of COVID, you felt it was
3:31
a little overstated, maybe for this reason that you know, there's a lot
3:35
of hype and fluff and not enough attending to those basics that you mentioned.
3:39
I mean, is that where you feel we are today has in some wave
3:43
of a post pandemic era been more transformative from the pandemic era, if only
3:49
because some of the effects of the former have been felt more fully. Or
3:53
are you still just quite skeptical of there's a lot of a lot more talque
3:58
than action. I'm sort of middle of the road both on this um being
4:05
an old consultant. It depends, you know. I think the challenge around
4:10
this is that, on the one hand, any change right takes a long
4:15
time. And that was that was what I was bristling at when people were
4:18
making these these big, grandiose statements about, you know, more transformation in
4:23
three months, and we saw in ten years not really right. That was
4:26
surface level stuff. And so I think we're seeing that the actual transformative process
4:31
happens very slowly. It requires deliberation or requires focus, and that is happening
4:36
now. What I do think is interesting is and This is the part of
4:41
that story I agree with is that there has been more significant movement in the
4:46
last say, twelve to eighteen months of people being fully prepared to make much
4:55
more significant change. The challenge, I mean, going back to this whole
4:59
data thing, just because it's time of mind for me, is that the
5:02
foundation of what it takes to make that change is still hard and will always
5:08
be hard. I'm working on another I'm calling them deep dives now, right,
5:12
but the sort of deeper analysis focused on some of the challenges around you,
5:15
the gains or the lack of gains in automation that we're starting to see
5:18
now, and that a lot of that is that what we're running into is
5:23
that the the what is necessary to sort of get the same gains that we
5:29
used to get by simply applying any technology to drive efficiency while all the low
5:32
hanging fruits been picked, and so now we need to start dealing with the
5:36
big hairy problems, the cross functional end to end process gaps. And that's
5:43
why things like process mining are becoming such a huge deal now. And that's
5:46
hard. That's really really hard to do from not only a technology perspective,
5:50
but from the organizational change and the behavioral change that's necessary to do that.
5:56
So the good news is, I think we're starting to have the conversations.
5:59
I think we're I feel like when I'm talking to folks now, we're past
6:03
that hype that, you know, oh my gosh, all the stuff,
6:06
and they're starting to get to the real change that is starting to take effect.
6:12
But it's always and I'm not a big fan of almost most of the
6:15
stuff that Gardner does, but the hype cycle was probably one of the most
6:19
you know, dead on analyzes of how things go. And we see that
6:24
play out over and over and over again. And so I think we're in
6:27
that right now, right where a lot of these things I got super hyped
6:30
up during the pandemic. We're approaching that trough of despair or whatever it's called,
6:33
right where we all sort of forget about it, and that's when all
6:36
the real change starts to happen. The one I am watching of all of
6:41
that, and I'm sorry if I'm jumping ahead, I don't remember, but
6:44
that I am watching very carefully is the whole remote work with the return to
6:49
office thing and the dynamics changing there. I'm you know, we'll see I
6:53
maybe I may have been wrong on that one as far as I thought that
6:56
we were going to see a stall, but this was the tipping point to
6:59
a lasting change, and the threat of the recession has caused a massive turn
7:03
back where employees have lost a little bit of their leverage and organizations right now
7:09
are putting pretty deep stakes in the ground saying, hey, you need to
7:12
get back to the office. So we'll see if this is the last gasp
7:16
or if this represents a miss in terms of we're not going to see a
7:20
bigger shift, We'll see totally. Yeah, we had We had Jacob Morgan
7:26
on the show very recently, and he was saying that the pendulum is swinging
7:30
back towards employers, or that it's swung too far in favor of employees.
7:34
And one of the things that's important for everybody concerned that we have a kind
7:39
of a united front and a kind of a proper relationship fundamentally between employers and
7:45
employees is that we have a common workspace and workplace. There's an opinion a
7:49
lot of people have. It's when I find difficult to relate to myself since
7:55
I work for an international company and I've been working around from home for years
7:59
and feel quite ncted simultaneously to it. I mean, what's your view in
8:03
general employee power dynamic jaws, Well, so I think so. I think
8:09
we should differentiate between three different things here. One is exactly that there's a
8:15
power dynamic, the leverage. Who has the leverage to dictate the terms of
8:18
the relationship the employee of the employer. For arguably the last decade, it's
8:22
been the employee, particularly in tech companies, particularly in anything where we were
8:26
looking at these highly skilled workers that were in limited supply, the employee was
8:31
in control. So there's no question that the power dynamic the leverage has shifted
8:37
across the board. And we can talk more about that in a second.
8:41
But I think it's important not to lump everything together. There's two other there's
8:46
the then the element of employee productivity, collaboration, innovation, and then there's
8:50
the idea of the employee experience, and all three of those need to be
8:54
looked at distinctly. So my opinions on the last two haven't changed it at
8:56
all. Right, that the driver of productivity and innovation in some cases is
9:03
going to be benefited by in place work, but most of the time it's
9:07
not. There's plenty of good data now that shows that people are actually more
9:11
creative, more innovative when they're working independently, have the opportunity to come together
9:16
to collaborate in very specific moments of the creative process and the creation process,
9:22
and there's plenty of evidence that shows that we actually drop we would see losses
9:26
in productivity with people in the office because of all the you know, water
9:31
cooler stuff, and so I think there's cases. I'm a huge fan of
9:33
hotelling. I'm a huge fan of organizations having an office space to allow people
9:37
to come together. But I think this is a you know, this is
9:41
a power dynamic play that people want to exert the executives want to exert control,
9:46
and that's all this is. And then the last element of this,
9:48
the employee experience, I think is critical because despite all the purported layoffs,
9:56
and I don't want to say that's like, obviously the layoffs are happening,
9:58
but I think people may did a much bigger deal than it actually is from
10:01
macroeconomic standpoint. The reality is that if you are in a competitive space,
10:07
you are still competing for employees, and employees are still selecting to choosing to
10:11
work with organizations who are delivering a better cohesive employee experience because they're able to
10:18
get more more done, they're more productive, they have a more enjoyable process,
10:22
and that is better for them as from a career standpoint, because that
10:26
then leads to better outcomes that they can tout on their resume, and you
10:28
know, and it ultimately is a great indicator of a successful company. So
10:33
if I look at these three things, there's no question that paradynamic has shifted.
10:37
I don't think the fundamentals of the other two have shifted at all.
10:41
And then my last little comment on the paradynamics is there's a really fascinating article
10:46
that came out a couple of weeks ago from Fast Company that talked about the
10:48
fact that even that all these layoffs, it's it's fast followers stuff, it's
10:54
it's groupthink. That there's this really interesting number that the average is, like,
11:00
they're almost all at right at six percent, and six percent is the
11:03
number that the street, i mean Wall Street investors say, oh, they're
11:07
cutting the fat, that's great, and employees, it's not a deep enough
11:11
cut that they're freaking out and running for the hills and so it's like this
11:13
carefully calibrated hit that these large tech companies are having to do just to sort
11:18
of strike this balance. But it's not based on any fundamentals where there's weakness
11:22
in the market or anything. So it's a very interesting sort of space,
11:26
and I think people are mostly reading too much into sort of what's happened.
11:31
So Charles, let's touch out a little bit on that employee experience. We
11:33
just recorded an episode that talked specifically about how an employee experience capability benefits enterprise.
11:41
It and specifically the enterprise it helped USK, but many enterprises have a
11:48
customer facing HELPDUSK as well, and when they have a poor experience, it
11:52
translates into bad consumer experience. Think about when you call your cell phone provider
11:58
or your cable company or your insurance carrier and their PC has to reboot or
12:01
blue screens in the middle of the phone call. It impacts that customer experience.
12:07
Where do you feel that kind of the dynamic between employee experience and customer
12:11
experience? Where does that sit now in twenty twenty three, Is there really
12:15
an appreciation in large enterprise and is that increasing or is it decreasing? So
12:24
again, I think what it's really important to segregate perception from reality, or
12:28
the fact from the way people see it. The fact is, again the
12:33
fundamentals, in my opinion, are undeniable. You can prove it with data,
12:39
you can prove an anecdotally. I tell a story of my father several
12:43
years ago, and I won't name them, but a large cell phone provider,
12:46
and he had some minor issue whatever was I forgot what. He had
12:50
to go try to return something. And he went in there and the employee
12:54
was super helpful, super friendly, super engage, all the things you'd want
12:58
to deliver a great customer experience, could not make the system on the back
13:01
end do what needed to be done to solve my father's problem. Finally had
13:05
to say, you know, I think this is because you bought this other
13:09
story. You need to go there. So now my dad gets in a
13:11
car, drives fifteen minutes to the other store, gets there. That guy
13:15
still can't figure out, finally does the right thing and says, you know
13:16
what, here you go, here's this thing that you need to just forget
13:20
about it. I'll deal with on the back end. But this wasn't a
13:22
point of where the bad employee experience was very directly creating a horrible customer experience,
13:28
even when all that expensive customer service and customer experience training they invested in
13:33
paid off. These guys were doing all the right things and their systems were
13:35
in their way. And I can you know, I've countless stories that executives
13:39
have shared with me about these sorts of things. So to me, it's
13:43
undeniable the relationship between the employee experience and the customer experience. What I do
13:48
think we're seeing right now, though, is is because that relationship is still
13:56
hard to quantify. It's hard to make these very direct connections for a whole
14:01
bunch of reasons. Executives were always sort of suspect, always sort of like,
14:07
Okay, I get it, it makes sense, but is it going
14:09
to rise to the top of my investment plan or restiment portfolio. And I
14:15
think with this leverage shift that we talked about, it's actually become harder even
14:20
if somebody wants to justify it harder to do so, right, because the
14:24
argument used to be, hey, we need to go compete with these employees.
14:26
They've got all this leverage, and now that argument has fallen away.
14:30
So now you the only way you can really make these these claims is by
14:35
directly connecting to that customer outcome. And as we've shown, that's hard to
14:39
do, not impossible, but it takes a lot more work. And it's
14:41
where tools like yours help, right you start getting data. What we need
14:43
is data around all of it. But it is definitely a challenge, and
14:48
I think as a result, we've seen certainly a lot less interest in the
14:52
employee experienced side of things. Yeah, there's this concept that there's I think
14:58
it was a ten x difference in the aversion to losing revenue or losing benefit
15:05
versus the excitement of gaining revenue or gaining benefits. For every dollar you would
15:09
gain, you have a ten x aversion to losing revenue. So it's maybe
15:15
that's a concept that we need to push on a little bit more. Um
15:18
But let's let's talk a little bit about kind of the the other parts of
15:24
employee experience and specifically things that touch on the the IT folks. And I
15:31
know you've been interested for a long time and the impact of automation and AI
15:35
chat GPT is on the tip of everyone's tongues right now. It really took
15:39
off at its launch. What are you What are your thoughts on chat GPT
15:43
and how it's been received. Is it a game changer or is it?
15:46
Is it not as big a deal as everyone is making it out to me?
15:52
Oh gosh. So I'm working on another deep dive on this particular topic,
16:00
and I'm really struggling because I generally don't don't weigh in on these hyper
16:07
trendy topics because there's so much incoming spilled on it. But I'm a contrarian
16:11
on this. I you know, I've talked to a couple of executives who
16:14
like it, executives who are all in on it, who think it's fabulous.
16:17
And I think it has its uses, um, particularly particularly with like
16:22
closed data sets, Like if if I'm going to feed it a meeting notes
16:26
and say summarize this, it's going to do a pretty good job with that.
16:30
UM. There's so there are use cases where it makes sense. UM.
16:33
I've got two, maybe three major issues with the whole generative AI space.
16:41
UM. And I guess I should be more specific specifically with you know,
16:45
large learning model based generative AI that the fundamental way it works is it
16:52
has zero context, It has no understanding what some people in the AI community
16:56
called the semantic layer, no understanding of the underlying concepts. It is literally
17:02
just putting words together, and because those language models are so high, it
17:06
does a pretty dang good job of guessing what that next word is. But
17:11
because it's missing that contextual layer, we have absolutely no way to validate its
17:15
veracity. And the only way to validate is by somebody actually looking at that
17:21
output and making sense of it and saying this is this is correct or not.
17:23
So I think it has uses, like if I'm taking a lot of
17:26
big texts, I need to summarize it and then someone's going to do a
17:30
proof layer to go and read it. So, yep, that makes sense,
17:33
you know. I think it has its place, and so the places
17:36
that I've heard executives starting to experiment with it, that's sort of the use
17:40
case. I think, the broader implications about it, or the broader use
17:42
cases of it. I think if I'm an enterprise executive and I was like
17:47
diving into it, you'd be making a big, big mistake. You're opening
17:49
yourself up to a ton of risk, and you're seeing that already with a
17:55
lot of the major organizations sort of quietly on the back end saying do not
17:59
use chat GBT for anything that is, you know, actually important. So
18:04
I think that it is overblown from that standpoint, I also think it has
18:08
a potential negative consequence in two ways. One is that it's sort of like
18:15
when ALEXE and Sirie first came out, that initial reaction was so positive and
18:21
everyone got so excited that it actually set the expectation bar so high that nothing
18:25
could actually live up to it, including those products. And so now we're
18:29
seeing, right, I mean, Alexa is losing a ton of money right
18:32
on their side because they're not seeing the pull through that everyone sort of expected
18:34
from this, and that's because it's proving to be not nearly as useful as
18:37
everyone thought at that first blush. And I think it's going to be the
18:42
same thing. So now suddenly there's a whole bunch of really good organizations out
18:47
there doing conversational AI that do have a contextual layer, that do pride reliable,
18:52
verifiable data on the back end, and you know, but what they
18:56
don't have is this super clean interface from the the language learning models, where
19:02
they're not going to be they're not going to appear to be as intuitive rights
19:07
as someone I don't I need to find out who said this, but they
19:10
said chet GBT is basically mansplaining as a service right has no idea what it's
19:14
talking about it, but it says it with complete conviction. And the problem
19:18
is is that if you're going to have a conversational AI platform that is in
19:22
fact rooted in defensable data, it can't be that effective, right, And
19:29
so I think it actually runs the risk of taking some really solid technologies and
19:33
people like pooh poohing them. It's like, got way minute, it's why
19:36
isn't as good as chet GPT on this conversational side, and that's going to
19:41
create an expectation gap that is unfortunate. And I think that that is the
19:45
second problem is it extends more broadly. There's some amazing uses of AI being
19:48
being applied and deployed across all spectrums of the enterprise right now, and they're
19:56
all going to sort of pale and comparison in terms of that whizbang factor,
20:00
but they are in fact much more impactful on a real basis of how you
20:06
can apply them to think. You know, if you look at some of
20:07
the leaders in the AIOps space right of how they're using this technology to truly
20:12
do amazing stuff of dealing with alert storms and all of these different applications,
20:18
and people are going to sort of again, I think they may pale in
20:22
comparison to this against They can't live up to that hype, and so they
20:26
may actually, you know, it could go the other way. It could
20:27
be that this opens up this whole door of people really excited about AI.
20:32
But I fear that it's going to have the opposite effect. So I'm a
20:36
bit of a contrarian. I'm fairly bullish over the long haul on how this
20:41
technology gets applied, fairly barish in the short term, and even outside of
20:45
enterprise it where I am interested to see where it goes is say, for
20:51
instance, something like next Think takes employee sentiment and device data to get better
20:56
insights and context. I'm wondering when a company like Boston Dynamic Mix with their
21:00
robotic dog starts to use device metrics and context of the environment along with input
21:07
to be able to tell that dog in AI terms how to respond and get
21:11
better context and more more insight into how that how that robot is actually interfacing
21:17
and reacting with the environment. Yeah. No, and look I think.
21:19
I mean you're sort of you're started making my point and that there is an
21:22
amazing stuff that can happen as we start applying this technology. I mean,
21:26
at the end of the day, all AI is effectively pattern matching. I
21:30
mean, that's how our brain works. We're big giant pattern matchers. Um,
21:33
So there's tremendous potential in all of that. But what you realize is
21:41
that once you actually require it to be defensibly accurate, this becomes inorderately more
21:49
difficult. Right. It's why you know, we had massive predictions about,
21:52
for instance, using AI to do early detection of things like radiology scans,
22:00
and the technology is coming along, but this explosion where people thought is going
22:03
to just replace radiologist almost overnight hasn't proven to be true because that need for
22:08
it to be accurate requires that contextual layer where we can defend it, and
22:14
that is incredibly difficult. Right. Chat GPT had this great advantage. It
22:19
doesn't have to be right. You know, they're just gonna thrown it out
22:22
there, and so so you know the use cases that we're seeing. Somebody
22:26
just wrote this that that I thought was dead on. It's like when when
22:29
you're using chat gypt for a creative endeavor all day long, because it's not
22:36
you're not relying on that. You're using it as an idea generator or as
22:38
a starter for something you're working on or whatever. I think it's fabulous in
22:42
that use case, but people are immediately starting to go to, oh,
22:45
I could just ask it whatever, and it's going to give it and you
22:48
know, we'll see I think I think I'm I'm somewhat excited that Microsoft deployed
22:53
it as part of Being because I think it's going to be a great test.
22:59
And I do appreciate the way they're using it, meaning they are using
23:03
the language model as the front end and back end layer, meaning they're using
23:07
it to simply do the intent recognition and to then give it a cohesive response,
23:12
but they're connecting it to this other data. So I think that that
23:17
sort of a model, that sort of application makes a lot of sense.
23:22
But we're still in the very very early day. So if I'm advising anybody
23:25
on this, it's like, take a breath and you have a good healthy
23:29
dose of skepticism around it, and you touched on it a little bit there.
23:33
But in terms of automation, in my experience, it executives not at
23:37
all, but in many instances think about automation as the easy button. Right,
23:42
I'll automate it. I won't have to have my people do it.
23:45
It'll do it quickly and just go deploy it. What do you think about
23:48
specific advances in workplace automation where do you think we could see some comparable impact
23:56
there? So I'm actually super excited about out where this is going from a
24:00
couple of perspectives. The first I mentioned I'm working on this new report and
24:07
I forgot the title of it, but it's like the three reasons that your
24:10
automation efforts are stalling, and the fundamental cruxt of it is like I mentioned
24:15
that that the the gains of the future, the veins that get we're going
24:19
to be seeing because you're right, that easy button is gone, right,
24:22
it's it doesn't work anymore, it broke. We used it too many times.
24:25
That those gains are going to be found in looking at these cross functional
24:29
business processes, these things that go end to end within an organization, hitting
24:33
multiple, multiple functional units. Because in the past, we focused our automation
24:37
efforts in these very targeted functional tasks, and if you look at the early
24:41
days around automation, whether that was enablement of human automation or actual like URPA
24:48
type automation, it was done the exact same way, right our PA has
24:53
you came out of this very narrow focus of let's do one thing and automated
24:59
and and they're two across any of those different spectrums, whether we're looking at
25:03
workflow automation tools or RPA or whatever same thing that they're running. They're sort
25:07
of running out of the runway because all the simple stuff is being handled and
25:11
now we're seeing that the real gains are how do you connect this across multiple
25:15
dots? And and this is where AI can do a great job of helping
25:21
with that as well as tooling that simply makes it easier for organizations to do
25:26
that right, to look across functions and to go end to end. And
25:30
so you put these two together and I think we're going to see a massive
25:33
explosion. I mean, so a handful of people are starting to say,
25:36
hey, guys, what is chat GBT or what is AI really all about?
25:38
It's about making people more productive in their jobs, taking away all of
25:41
the stuff that is makes them inefficient and is wastele and allows them to focus
25:48
on these more creative tasks. That's what we're going to see here, I
25:49
think, and I think so many of our traditional back end business processes are
25:56
ripe for this. They're still highly inefficient, particularly in as handoffs, and
26:00
I think there's just a tremendous opportunity. So I think we're going to see
26:03
this sort of golden age of taking automation into these this real end to end
26:11
kind of domain. Right If if I you know this report, I'm talking
26:15
about going back to the days of process for engineering, and they failed for
26:18
a lot of reasons. If you're old like me and remember the nineties when
26:21
we try to do these big, giant projects. But one of the reasons
26:25
was that they were just so complex that you couldn't possibly develop leverage technology effectively.
26:30
They became these one hundred million, sometimes billion dollar projects that took years
26:34
and years to complete. And in this day and age, by the time
26:37
you get even close to being done, before you're even halfway done, everything
26:41
has changed. And so I think we're in this tremendous period where the technology
26:47
is sort of caught up with a need. So I'm very excited to see
26:49
where this all goes. Amazing stuff charg's just wonderful to have you back on
26:55
the show, wonderful to listen to your thoughts again, wonderful to get your
26:57
updates on I think very eventful twelve fifteen months of past since we last spoke.
27:03
So treat for us, treat for our listeners, and we will of
27:07
course point people to your wonderful website where they can find your writing and look
27:14
at. In terms of the chat GPT conversation, it keeps, it keeps
27:17
arising again and again you brought a really fresh perspective to it. But we
27:19
will also link in the show notes to some of the recent articles and discussions
27:25
we've had on this show and elsewhere around it. So look, just remains
27:29
for me to say, lovely to see you again, Charles, hope to
27:30
see you for a third date sometime in the future. And look, you
27:37
have a great rest of for you in the meantime. Thanks great being with
27:40
you. To make sure that you never miss an episode, subscribe to the
27:45
show in Apple podcast Spotify, your favorite podcast player, and if you're listening
27:49
on Apple Podcasts, make sure to leave a rating at the show. Just
27:53
have the number of stars you think the podcast deserves. If you'd like to
27:56
learn more about how next think can help them your digital employee experience head over
28:02
tonextthink dot com. Thank you so much for listening, Until next time,