KCAA: Inside Analysis with Eric Kavanagh (Sun, 21 Jul, 2024)

KCAA: Inside Analysis with Eric Kavanagh on Sun, 21 Jul, 2024

Episode Transcript

(00:00):
I mean of Biden's announcement may beintentional. Democrats up here were privately urging
the President and his team to bowout of the race before they were forced
to essentially call on him to exitpublicly. According to Serkin, the timing
of the announcement may be making itdifficult for Democrats to coordinate and present a
unified front. Biden's announcement comes asan increasing number of Democrats have been calling

(00:23):
for his withdrawal due to concerns abouthis advanced age, But top Republicans want
President Biden to step down now beforethe end of the term. House Speaker
Mike Johnson posted, if Joe Bidenis not fit to run for president,
he's not fit to serve as president. This is an NBC News special report
NBC News on CACAA Lomelada, sponsoredby Teamsters Local nineteen thirty two, Protecting

(00:48):
the Future of Working Families Teamsters nineteenthirty two, dot org. The information
economy has a rived. The worldis teeming with innovation as new business models
reinvent every industry industry. Inside Analysisis your source of information and insight about

(01:08):
how to make the most of thisexciting new eric. Learn more at Inside
Analysis dot Cossideanalysis dot com and nowhere's your host, through Eric Kavanaugh,
and all right, ladies and gentlemen, Hello and welcome back once again to
the only coast to coast radio showin the US. Today that's all about

(01:30):
the information economy. It's time forInside Analysis. You're truly Eric Kavanaugh here
with an all star cast, somegood buddies on our call. Today.
We'll be talking with Aaron Wilson ofAthena Solutions. They are a strategic consultancy
systems integrator. They do a lotof data governance, a lot of data
warehousing, master data management, althoughthese days they say mastering data because MDM

(01:51):
apparently is a bad word now,so they say mastering data. Everyone understands
that. We'll also be hearing fromJim Smith of Click in Fact the webinar
we just did a moment ago whichyou can hop online to Inside Analysis dot
com to find It was all aboutthe associative engine and Click, which is
very interesting stuff. We're going totalk about that and what it means and
why it's special. The title forour show is Inside by Association, Exploration

(02:14):
without Constraints. And then, last, but not least, our good buddy
David Lintikam is online. He's beendoing a lot of where this guy's been
around, I mean he needs towork for NASA. He's doing agentic AI
work these days. That refers toAI agents. We may even talk about
that on our next show, orwe're going to be featuring kindy a very
interesting large language model tooled. Ofcourse, these large language models CHATGBT,

(02:36):
Gemini, Claude. They've taken themarket by storm, but they have certain
use cases. There are times whenyou want to use those to do interesting
things like code, like code generationor text generation, but they're not really
analytic engines in the traditional sense.So we're going to talk about what that
all means and basically help you figureout what does your business need, well,

(02:57):
what kind of solution will make sensefor you. So I'll just throw
out a few comments. We didtalk about the associative engine and click can.
I remember getting a first briefing onthat, gosh, probably fifteen years
ago by good buddy mind Donald Farmer. And you want to talk about when
something does clicks? Man, Iwatched it. I was like, wow,
that's cool. So what it doesand this goes all the way back
to the genesis of the tool tothe kernel is it will automatically show you

(03:21):
visually relationships between entities, between conceptslike products that you sell and customers you
sell them to, for example,pretty important information. You're not really going
to see that as readily if you'relooking at ros and columns. I mean,
you can look and excel and builddifferent graphics and different things to visualize
the data. But how nice isit to just see the relationships right out

(03:44):
of the box. It's very veryimportant stuff because the whole process of discovery
requires thinking through and understanding what you'relooking at. You know, in a
previous show we did in this serieswith a lask and also from Athena Solutions,
had this great quote where she said, in order for data to be
an asset, it must be understood. I was like, good point.

(04:08):
If you don't understand the data,it's not an asset, and in fact,
it might be a liability if youdon't understand what that data is telling
you. This gets back to aconcept called data literacy, which is very
much in conversation these days for goodreason, because now we have all this
data. You know, twenty yearsago, really only the fortune two thousand

(04:28):
companies could afford to build an enterprisedata warehouse. It took six months,
It costs ten or twenty or thirtymillion dollars. It was a huge effort.
It took a lot of time.You want to talk about time to
value, We're talking years to getto value. Well, that's just not
even acceptable anymore. You cannot.No one is going to approve a three
year project in data that will onlyprovide value a year or two later.

(04:50):
There's no chance of that happening.And luckily it doesn't have to happen.
The new technologies these days and theold technologies that are modernized, they allow
you to get value very very quicklyfrom your data, and that's what you
need. It's a very fast movingenvironment these days. Just think about the
Internet and all the stuff that youcould buy on the internet. Think about
all the data that's out there.Think about data science. There's this whole

(05:13):
data science industry. And we've talkedon this show before. I've marveled in
fact that the data science teams inlarge organizations often don't interact with the data
warehousing teams, which in my opinionmakes exactly use zero sense. You want
these folks talking to each other,but it's very common that they don't.
So on this show we're going totry to hash through some of these issues

(05:34):
and really explain why this associative engineis so important for analysis, because,
again, anytime you're doing what's calleddecision support, that's one of the older
terms, you're trying to get informationthat helps you make better decisions about where
to spend your money, where tospend your time, how to hire,
whom to hire, where that personwill fit in an organization. All these

(05:57):
questions can be aided and should beaided with data and analysis. And the
data, again from an analytical perspective, has no value until you've analyzed it
right. And just very quickly,last comment, One very cool thing about
all these large language models hit inthe markets today is that most of the
analytics world deals with structured data,so data that's in relational databases, data

(06:17):
that's in tables, rolls, andcolumns. Being able to analyze that and
get some value. One cool thingabout large language models, we'll pick this
up on our final show in acouple of weeks on the twenty ninth,
is that it analyzes unstructured data too, text, documents, word, documents,
PowerPoint presentations, some really really coolstuff in there to get context to

(06:38):
help understand the numbers. And sowith that, let's bring in our first
guest, Aaron Wilson of Athena Solutions. Welcome back to the show. You're
in the industry with us, andyou've done some pretty serious work in financial
services. You got to get thosenumbers right. What do you think about
the importance of an associative engine forbeing able to analyze data? Yeah,

(07:01):
I mean, I think it can'tbe overstated. And I think one of
the the the interesting thing is,and there are people at Click who I'm
sure would say this, is thatit's not it's not that complicated a concept,
but many people don't really understand it. And I think it goes to
Jim's point about how usually the emphasisis on you know, okay, how

(07:25):
many cool visuals can we can wegenerate from this thing, or you know,
oh wow, there's the pop thatcomes out of like, you know,
a visual that maybe nobody's ever seenbefore, and which is all that's
great, but it's very important inthis transition in terms of, you know,
making it not just a visualization tool, making it a tool for analysis
because the associative engine is key tothat. It's you know, query based

(07:51):
tools are very good, just likeI said, if you know where you're
going, but to be able toexplore the I think the associative engine is
I'm I mean, it's huge becauseit shows you all the data in front
of you. It shows you contextualdata, things you might not be thinking
about it you didn't sit out tolook for in the first place. Yeah,
you know. And in fact,before I throw it over to Jim,
I'll just throw it back to youfor a comment on this. Aaron

(08:15):
Jim had a great quote in thewebinar we just did where he said,
this tool will give you answers toquestions you ask and answers to questions you
didn't ask, right, which isgreat because it helps shape the contours of
your understanding and that's important stuff.It's just like being able to visually assess
a situation. Like if you're asecurity guard at a rock concert or something.

(08:37):
You want to have a good viewso you can see everything that's happening.
Yeah, you may want to keepan eye on this person or that
person, but you also want tohave the ability to see the whole room
and I think that's what this does, is it gives you the ability to
see not just one answer to aquestion you're asking, but the broader context.
And then when you kind of movearound and select and decel, whether

(09:00):
it's products or services, or regionsor individuals or financial amounts or whatever it
is, everything changes in that discoveryprocess. It's a learning experience. I
mean, you've got to try hardto not learn something by engaging in that
process right completely. And I thinkyou know the idea really, you know,

(09:24):
SEQL queries aren't necessarily designed for thiskind of analysis. You're you're continually
either drilling down or moving back upthe ladder. But the idea of being
able to have all the data frontin front of you and explore, I
mean, I can tell you thatusers really like this because you know,
for a user, once they gettheir hands on the data and they get
their hands on the visualization tools,the next thing they want to do is

(09:48):
explore it. They want to doanalysis. And in that sense, you
know, if visualization maybe part oneover the last few years has been democratizing
these graphical capabilities along with that accessto the data. In a sense,
what Click's able to do is democratizedemocratize analysis, which is extremely powerful.

(10:09):
Yeah, that's a good point.We have a couple of good comments from
our audience members too. I'll bringthem in probably later on in this segment,
but let me throw it over toJim Smith from Click. I mean
you've talked about how this goes backto the kernel. This was an idea
someone how long time ago, andI'm always fascinated by the kernel of a
technology, right because you have someidea, you're trying to address a particular
issue, and I mean, Ihave to say, I think that this

(10:31):
associative engine is central to the capacityof really exploring data very quickly and getting
that value, that time to valueway way down. What do you think,
Jim? Yeah, First of all, Eric, thanks for having me
on. Absolutely. I mean twothings about this associative engine that really brings
valued organizations right away. The firstis the fact that you know when you're

(10:56):
looking at your visualizations, when thatdata is in memory and you start to
slice and dice it. As anend user, what you don't want to
do is you don't want to seea bunch of progress indicators saying oh,
you just asked for this very complexpiece of information, Why don't you just
wait five minutes as I go tothese different sources to get that information.

(11:16):
So that's the first thing that kindof click helps with is kind of that
immediate access to the information you askabout. And then the second thing,
and we kind of saw this inthe webinar if you were able to attend
that is this whole concept of green, white, and gray. That's kind
of the color scheme that click useswhen a business user is starting to filter
the information. Well, that clearlyallows a user to get smacked in the

(11:39):
face with what they're asking for andwhat they want to see and what they
don't want to see. And I'vedone a lot of demos of the technology
over the years, and it's amazing. When you show an organization this technology
with their data, people are alwayson the edge of the seat, you

(12:00):
know, when they come in andthey're sitting back, and then you can
kind of go and start clicking ontheir information and they see those gray values,
which I refer to as kind ofthe golden nuggets. Those are the
things you didn't ask for. Peopleare the users of the data just get
very excited. Sometimes it's happy excitement, sometimes it's not so happy. Excitement

(12:20):
because they're seeing things they don't wantto see, but they're always able to
see them above and beyond what theywould have in other solutions. And the
key is the context, right context, And this is actually one of the
real challenges with artificial intelligence is whatcontext does it have? What is the
context window people talk about. That'ssomething that has to do with both time

(12:43):
and dimensions. And if you playwith these large language models, you'll know
if you get a very very complicatedtask, it'll sit there and think for
a while, and then sometimes they'llgo, oh, I'm a language model.
I can't figure that out. That'swhere it just defaults. Now.
Sometimes that's a guardrail. Sometimes itwas just too complex and it doesn't want
it to do that. Yeah,although I heard something very strange that will
just pick up later maybe. Butsome guy told me that if you tell

(13:07):
chat GBT that gem and I cando something, it'll work harder. That's
true. That's crazy, But kindof back to you, Jim, the
context is so important. In thatwebinar we did earlier, you were showing
an example of snack items that acompany is selling, and you can see

(13:28):
one of the dimensions that's visible iswhat kinds of customers get that, like
grocery stores and schools and other things. And there was a gray area at
the bottom, which was hotels forexample. So the beauty from an analytical
perspective is that right away the usersees, wait a minute, we're not
selling these to hotels. How come? And that's a question you asd that's

(13:48):
that's a golden nugget, right,yeah, yeah, and we see that
all the time. The example Ialways I get excited about is when we
go in and you start showing thisto let's say a sales organization. You're
showing it to sales reps or salesmanagers. And you know a lot of
times when you're doing using that kindof scenario, you always say, oh,

(14:09):
you know, the company wants tosee the top reps or the top
products, so they use a toollike click to show that information. And
with click, what always comes outis, oh, well, here are
the reps that aren't selling. Hereare the products we're not selling at all,
and we're not selling them into thisregion. They never thought about that

(14:30):
information as far as you know,kind of the context they were looking at.
They were looking for top reps andtop products and what they're getting now
is reps who aren't selling products,that aren't selling, regions that aren't selling,
which they win that see with othersolutions, and that actually changes the
query or the question that they startedto ask initially, which always gets them

(14:50):
excited. Yeah, that's an excellentpoint, and I'll bring in David Linham
here to comment on this. Youknow, David, we're kind of talking
about the null set right where there'snothing, and you don't typically search for
that. I mean, I've heardlots and lots of analysts over the years
say, look, if you reallydon't want to explore your data, look
for the zero values, look forthe null values where is there nothing?

(15:11):
Because that could be interesting, Butthat's not terribly intuitive, right, you
want to look, as Jim wassuggesting, oh, who are the top
selling salespeople. That's good information,but it's also very good to know that
we're not selling any of these productsor in any of these regions, and
these guys aren't doing anything like waita minute, let's call a meeting.
Right, What do you think,David think's absolutely right? I mean,
you know, give you an example. You know, had a ceramics manufacturer

(15:33):
that was concerned about the quality ofthe ceramics going down at certain times,
and they couldn't figure it out.They looked at the quality of the suppliers,
they look at the quality of thegoods that went into it. Ultimately,
and you know, at the endof the day, it was related
to environmental factors whin the factory.When the humidity was up to a certain

(15:54):
amount of a certain amount, that'swhen the error started occur and they lost
lots of money in making these happen. So looking at ultimately how information relates
to other information without an intuitive understanding, and how they relate, in other
words, making these are previously unrelatedthings that come together and then they make
sense. And sometimes that's going tobe the ability to look at null sets

(16:14):
and other operations within certain data setsand how they relate to certain systems.
When I see data missing, that'sdata onto itself. That doesn't mean that
the data is missing. That thismeans that's a data point that I need
to explore why is the data missing, and ultimately what it means that the
data is missing. And so inthis case, they had a certain margin

(16:36):
of errors and a certain defect ratethat went up, and they were looking
for correlations between it and looking atthings that were unrelated, and ultimately they
left to a huge amount of value. They saved huge amounts of money.
In some instances they save the wholeproduct line. That's amazing. And what's
so interesting is it's atmospheric orright,Probably no one going in thought for a
second, hey, maybe it's thehumidity until you started looking at the data

(17:00):
and you're like, oh, waita minute. Every time this happens,
the humidity goes up. So that'sa classic aha moment that can change a
whole business and save, as yousuggest, the whole product line. It's
something I'm always fascinated by these revelationsthat are outside the sphere of what you
were considering, right, And Ithink that's one of the problems with structured
data is that we're trying to forcethe world into the structured model where it

(17:23):
doesn't always fit perfectly. Right,David, Right, and absolutely. And
also it comes down to the peopleare trying to look for AI to save
them in this area, and itwon't. Unless the AAI system is going
to be trained in the information,it can't make the correlation. So everything
dependent. All an AI system doesis a mirror of the data that's used
to train it. And so ifwe're not training it with all of the

(17:45):
correlated data points and the ability tokind of look at all these unrelated systems
because they're not trained, because thepeople who train the data don't know that
they're related, then you can't reallykind of uncover the value and that data.
So and you know, looking atthe webin we just went through,
and that was kind of an AHAmoment in me that the ability to leverage
data in new dynamic ways is theability to find value and information that previously

(18:07):
wasn't there. And that's really whatunderstanding data analytics is all about and how
it brings value back to the business. And I wish more businesses would see
this and kind of understand where thevalue is. Yeah, that's just an
excellent point. And seeing is believing, And one of the promotions I sent
out for this event was seeing isknowing. When you can see something,

(18:30):
the visual metaphor is a very powerfulone because you can see disparities, you
can see connections, and when youcan start to play with that, especially
moving things around, like I'm ahuge fan of slider bars. If you
slide over time back and forth,where does something happen. Well, it's
kind of important to know where somethinghappens, but folks don't touch up.
De will be right back. You'relistening to Inside Analysis. Expect you welcome

(18:59):
back to Inside Analysis. Here's yourhost, me, Eric Tavanaugh, to
show. Okay, folks back hereon Inside Analysis talking all things associative analytics.
We've got Aaron Wilson with us fromAthena Solutions, as well as Jim
Smith of Click and our good buddyDavid Linthikam, an industry analyst formerly of
Deloitte. Now he's on his owndoing all kinds of interesting things. And

(19:21):
this guy has got to answers forquestions I haven't even asked yet, so
we'll try to get to those atsome point in the show. But Erin,
I'm going to throw it back overto you to comment on to me.
And we've talked about this for manyyears. That DM radio or other
show is in year seventeen, sowe've been going a long long time.
And I'm always amazed by the importanceof the fluidity of your experience with data.

(19:42):
In other words, you can't justclick something and run a reportant come
back on Monday to see what Imean. You can do that, but
it's not a whole lot of valuein that. What you really want is
this experience where you can play aroundwith things, select, deselect, change,
maneuver, bring in different dimensions,and that experience, especially if it's
in memory and that's the way it'sdesigned and click, that fluidity is crucial

(20:06):
to sort of match the analytical processof the brain. What do you think
erin Yeah, I'd say that's definitelytrue. I mean, one of the
things that you know at Athena thatis kind of near and dear to our
hearts. I mean, we havea product that you I know you've heard
about called the data analysis Sandbox,right, So this idea of exploratory analysis,
and what our product does is basicallybrings in a semantic layer over top

(20:30):
of all different types of data,different sources, different formats, and allows
you to essentially do an exploratory processkind of in the same way as click
the click associated engine does. ButI definitely think that there's a demand out
there for it. I think thatyou know, it's really again, I
think that once you give people thepower to get their hands on data and

(20:53):
to produce visualizations is short hop fromthere to they really want to explore it
and they want to produce real analysis. It's like a video game. I
mean, there's this whole concept ofgamifying things to make it fun, to
make it interesting. And when youcan do that rapid fire analysis, whether
it's with slider bars or selecting andde selecting entities and characteristics to look for

(21:15):
whatever the case may be, aslong as it's fast, as long as
it's real snappy, that's gamification,right, Aaron, what do you think
A bit? I mean, andthat's a really interesting analogy. Of course,
I'm of a generation where the analogyisn't lost on me, but it's
mostly what I've seen from my kids. But people do like people do like

(21:37):
working that way. I think thatyou know, the idea of just having
the ability to go somewhere just atyour fingertips, you know, it's it's
extremely compelling, it's powerful, youknow, and if you can increase engagement
amongst your users, I mean,that's powerful in and of itself. I
think that's a great point. I'llthrow it over to Jim to comment on

(22:00):
that. Getting the user to usethe data, I mean, if you
don't use the data, that datais not being used and it's not generating
value, and it might just bea liability. What do you think,
Jim, Yeah, I definitely agreewith that. And that's one of the
things. I know, what we'vebeen talking about is kind of that that
analytics type user who goes in thereand maybe sees what someone has created for

(22:22):
them and then wants to start changingand slicing and dicing. But a click,
you mean there's a whole set ofusers that don't do that. I
mean you still have users who wantto come in in the morning and get
an email with a PDF attachment witha bunch of rows and columns, and
you've got executives who don't really doany slicing and dicing. They just want

(22:42):
to get their dashboard and they don'twant to have to go to a separate
tool. They want their dashboard tobe in the application of choice that they
want. So I think one ofthe things that we always talk about at
Click is visualizations are great, butyou got to get the right data to
the right person in the right rightand it's not always taking advantage of let's

(23:03):
say that associative engine. But Itake that back, it's always taking advantage
of the associated of associative engine,but it's not necessarily always interacting, you
know, slicing and dicing. Sometimesit's just, Hey, I need my
information when I need to make abusiness decision, and I need a tool
that can provide me that data inthe format that I want at the time

(23:25):
that I want. And again,I think that's something that at CLIP we
do a really good job at providingthose different avenues to get the data to
the user. Yeah, and youknow, Aaron brought up one of the
magic words in his commentary a momentago. I'll throw it over to you
because I think in this environment,with the associative engine and just what you've
described, you enable analysis of semantics, right, semantics are very important.

(23:48):
And you know, for example,you could see, as a user,
wait a second, why is thisarea down here? Gray? I know
we sell this product. Maybe thesemantic engine was wrong, maybe it wasn't
coded properly, or maybe when thedata was imported there was a column missing
for example. I mean, we'restarting to see that very adeptly addressed by

(24:08):
observability tools that show when something doesn'thappen, because yeah, hitherto it's like
you would load it and you justpresume it's in there. Is it?
I don't know, let's take alook, but you don't know it's not
there until you see that gray andyou're like, wait a minute, why
is that gray? It gets backto this golden nugget thing which I just
love. It's the null set,like why is this a zero that shouldn't

(24:30):
be a zero? I know thatit should be X y Z number.
But that's what helps you get tothe answer, but helps you figure out
what's wrong in the system, rightJim. Yeah, And that's one of
the things that I would never sellclick Sense, which is the tool from
click as a data quality tool.There are data quality tools out there well.
One of the things that you alwaysrun into with this associative engine.

(24:53):
When we do let's say a proofof concept or just get some sample data
from a customer. You can loadthat in and sometimes you get gray values
because they're gray values you didn't sella product in a particular region. But
a lot of times you'll start toget gray values because of data quality issues.
Now again we don't fix it.We kind of highlight it for you

(25:15):
and kind of smacking in the face. As I mentioned before, with it
so that you can go back andsay, all right, you know what,
I've got two hundred and fifty thousanddollars of missing revenue. It's not
missing revenue, it's missing data thatsales transaction is not tied to the proper
customer ID. And I see thateasily and click. So I think you're

(25:36):
right there. The semantics of themissing data sometimes is because the data isn't
there, and sometimes it's because thedata is wrong, and that's what an
associative engine can help you see.Yeah, I'll throw it over to David
linthencom. Figuring out what's wrong.That's pretty important because making decisions based on
bad data is a very, verybad idea. You can think that this

(26:00):
group of salespeople is doing a fantasticjob. In fact they're not, and
you give them all raise, andthen you wind up throwing good money after
bad as they say, so,understanding what's incorrect is a huge part of
this equation, right David, Yeah, it's everything. And most enterprises out
there aren't utilizing their data in thecorrect way where they're able to find insights
into what is incorrect. They can'tsee what's wrong with their business based on

(26:25):
the way that they're currently tracking information, so everything's transaction oriented. Everything is
basically entering things into an inventory databaseand a sales database, things like that.
They don't see the force through thetrees and understanding their data. And
of course we went through the wholedata warehousing stuff. We're supposed to have
analytics to get us into there sowe could see the force through the trees.
And now we're moving into AI.And you look at the utilization of

(26:48):
data by most enterprises out there,they really couldn't tell you where the single
source of truth is. They couldn'ttell you where their business is rising and
failing. They couldn't tell you wherewhere a certain product lines are becoming weakening
into the marketplace until another six monthsof data transactions, and so they're missing
a huge piece. And I thinkthat a lot of those businesses are just

(27:10):
going to fall by the wayside becausethey can't see where they're steering the ship
and they're gonna en up running intoicebergs. Yeah, no, that's an
excellent point, and Aaron, I'llbring you back in. Our good friend
Kate Stratchnia from Dedicated had a greatpost on LinkedIn the other day. She
said, some companies love a singlesource of truth so much that they have

(27:30):
many of them, which I threwback my favorite quote about standards is the
good thing about standards is there areso many of them, right, but
we do need to watch out forthese things. It kind of gets back
to semantics too, right, understanding, But the whole point is that data
literacy itself is an ongoing process,and especially for some midsized or large organization,

(27:52):
there's gonna be a lot of stuffthat you don't know and you don't
understand about how the business operates.Maybe it's in operations are manufacturing, like
the example that David gave. There'sa lot to be learned out there,
and you want to be learning it, so you have to be using the
information and then collaborating with people too. That's another big part of the equation,
right, is don't just use itfor your own personal consumption, but

(28:14):
use it to start conversations, toask people about things, to get that
collaboration going, because that fuels analysisand it also enables data governance. Right.
What do you think, Eric,I definitely think so. I mean,
it definitely ties in with the youknow, the other part of this
series, the series that we weredoing with you about data catalog, where
the idea of how data cavalog canbe so important to governance, the idea

(28:40):
of, you know, first ofall, the more people can get their
hands on the data and work withit, the more they can point things
out, the more they can find, like you said, problems with the
data, maybe problems with the semantics, and you know, offer their expertise
and maybe fixed problems. Click helpsin that regard, in the sense that

(29:03):
the more people get their hands onthe data, the more people get involved,
they get engaged, and you havethe potential to improve governance. Really
engagement, I mean, you jumpedon that a minute ago, Jim.
I'll throw it back over to you. Engagement collaboration, you know, and
I've seen this myself. When someonebecomes engaged, it's a very powerful thing
and they don't want to let itgo. I mean, it's like I've

(29:26):
been doing tracked email marketing now foroh my goodness, twenty four and a
half years or so, because Ihave a friend who built a solution in
nineteen ninety nine, so I wasusing it back then. Once you can
see who opens and who clicks,you can't go back into the darkness.
I mean, you can't go backinto just the spray and pray nonsense.
And it's like it's an iterative process. You get closer and closer to the

(29:48):
signal, and that's what you want, right as signal. You don't want
noise, you want signal. Andthe more people you get to collaborate,
that's engagement and that gets you somewhere. I guarantee if people aren't engage,
these good things are happening. Whatdo you think, Kim, Yeah.
Absolutely. One of the ways thatwe've always gone to market at Click is
this whole concept of land and expandand really really what that was all about

(30:11):
is Click going into a particular businessdivision and showing the power of the solution
and getting users in that particular businessdivision very excited about it. And guess
what, when a set of businessusers is excited about things, they'll talk
and they'll talk to you know,their friends of the company, and they'll
show kind of the reports that they'reusing. And then sure enough, another

(30:34):
business division says, hey, Iwant that. I want to be able
to do those things that this firstbusiness division has been able to do.
And I think that just kind ofmakes Click or any other bi tools.
Again, it's not just for clickspread like wildfire in an organization. When
users can start to consume data ina way that makes sense for them,

(30:56):
everybody wants to do that. Yeah, that's right, And it's like the
snowball going downhill. It gets biggerand bigger, it gets better, you
get more attention focused on it.I'm actually looking our live studio audience has
lots of good comments and quotes,so I'll share those in the break and
we'll tackle them in the final couplesegments of the show. But you know,
David, I'll bring you back in. You know else is very interesting

(31:18):
here is that like just like adata catalog. We're doing a separate series
on that, but obviously it's related. The analysis of data is a galvanizing
agent. And when you can lookat this stuff and then have meaningful conversations
with people, that's very compelling becauseyou're not just asking what's going on.
You can see the data, youcan see the relationships, and you can

(31:40):
call someone and say, hey,Bob, I it's realized we're not selling
any snacks to hotels. Do youknow who's responsible for that? Like,
oh, let me check, Wellthat person actually left the company last year
and we haven't fulfilled that position.There you go, that's the kind of
thing you're looking for when you lookat this data, right David, Yeah,
absolutely, I mean you got toeven like the readiness for AI systems

(32:01):
is your ability to understand the useof data, and I think as if
you can't do that, you don'thave these insights, these current even the
rudimentary insights, then you have nohopes of leveraging AI to any kind of
value purpose. And these are veryexpensive systems to implement, So people seem
to be trying to jump directly fromkind of core understanding data semantics and core

(32:22):
understanding data analytics into the ability toleverage AI systems to kind of amplify that.
And my thing the readiness there.If you don't have an understanding what
your data is, what it means, and also the power that it's able
to be leveraging your way to derivemeaning and derive insights in the data,
you have no possibility of moving intoAI. And kind of that's a core

(32:44):
metric to success. Yeah, andthis is something that I've been on a
soapbox talking about it. We havea couple good questions from our audience.
For example, one gentleman is asking, do analysts expect that data analysis tools
like Gemini Advanced and Chat GPT plusone accelerate data use? Well, David
is making a good point there.You need to make sure your data house

(33:06):
is in order and you can actuallyuse these tools. I mean, I've
already seen amazing use cases where yougive a chat GPT or a Gemini a
significant amount of data and ask itto summarize. The summarized function is fantastic,
by the way, it's really interestingstuff. But you can ask it
questions. And I think that we'regoing to head down a road where this

(33:27):
nexus of GENI and traditional structured analysiswill be very, very compelling and very
powerful. Not quite there yet,but to David's point, you've got to
get your house in order. Andmaybe we've got a minute left in here
in this segment, Aaron Wilson,I'll throw it over to you. The
key ingredient there to fuse these worldsis data governance, right, and data
quality and understanding your data before yougo point in years, before you go

(33:51):
training a model, for example,on all this stuff that you found.
First, you want to sort throughthat stuff and make sure it's good,
real good. Thirty seconds go ahead, erin no question there. I mean,
you know, all kinds of thingscan happen when you put AI on
data that isn't ready for it.I mean even something I think a common
mistake that people do is is there'sold data floating around, right, you

(34:14):
know, there's data that's that mayhave been relevant at one time but no
longer is. And if you traina model on, you know, using
calculations and formulas that have been discarded, it's not going to be very helpful
to you. Yeah, that's exactlyright. Well, folks, don't touch
up that. We'll be right back. You're listening to Inside Analysis. Welcome

(34:40):
back to Inside Analysis. Here's yourhost, Eric Tabanac. All right,
folks, back here on Inside Analysis, talking to several experts. A wonderful
show today, Aaron Wilson of AthenaSolutions, Jim Smith of Click and David
Lentkam our industry analyst of Today andAaron, you a question for Jim,
so take it away. Yeah.Well, one of the things we haven't

(35:05):
talked about in terms of, youknow, the advantages of the associative engine,
but I know It's something that Clickhas well has pointed out in some
of their videos that I've seen,for example, is the advantages that you
have when you're trying to bring indifferent data sources, maybe a data source
that comes from outside of you know, maybe you have a data warehouse,
we have certain tables and so forth, and then somebody says, well,

(35:30):
you know, what would happen ifwe connected in this whole other you know
column, or this whole other dataset and look at those relationships. The
associative engine can really help you withthat, can it? It absolutely can.
So one of the things about theassociative Engine and what it's been able
to do since it was rolled outat Click back in the nineties is there
is a component of kind of dataintegration built into Click sense. Now again,

(35:53):
we have a data integration offering thatsays, hey, if you've got
one hundred of data sources and youwant to get your information into a data
warehouse to be used by a businessintelligence tool or machine learning tool, can
absolutely do that. But what's greatabout Click is if you're a smaller organization
and you just need to get lotsof sources put together, you can go

(36:15):
into Click and do it directly intothat in that solution. So you could
say, hey, I've got atable in Oracle on prem, I've got
a Snowflake table, and I've gotsomeone's personal Excel spreadsheet, and I need
to bring those things together. Notonly can you make those connections within click
Click's got kind of this capability thatwe'll look at kind of column definitions,

(36:38):
profile the data and say, yeah, even though this is Excel and this
is Oracle, we can put thosetogether based on what we see in the
data, and we'll do that foryou. And then you know, you've
got users who are able to puttogether their own reports. Even though it
hasn't built out the you know,the defined data warehouse, they're still allowing

(37:00):
users to kind of do that selfservice against lots of different sources. Does
that answer your question? Aron?That pretty much hits it right on the
head. Yes, that's where Iwas going. Cool. Well, and
this is such a good point becauseand I'll throw this over to David to
comment on when you're people still builddata warehouses, right, It's not going
away. They're going to be datawarehouses. I'm pretty sure forever and ever.

(37:22):
Llms are not going to supplant thedata warehouse against the different tools text
generation. It's great for summarizing things. It's great for getting a consensus about
what has been published on a topic. That's really what these lllms are,
their text generative consensus engines, andit's good for a big part of the
process. But the understanding the dataand the relationships it's a separate deal and

(37:45):
it's very very important. And David, when they were talking about this,
I'm thinking to myself, that isreally valuable to in the process of deciding
what's going to go into your warehouse. Explore that data with click, explore
the relationships between things, because that'sgoing to help you figure out what is
the optimal set of data that we'regoing to use to put into this warehouse

(38:06):
to fulfill business needs. It's avery important part of the process. What
do you think data It's everything,And I think that ultimately that's what's missing
in terms of the data analytics asto the data, what the data means
and what it can be used for, and the kinds of insights that we're
looking to get out of the information. So your ability to understand the usage
of data and your ability to findvalue within the data before you start building

(38:30):
these things, before you build theseanalytic tools and data warehouses and things like
that is something that many enterprises aremissing. I think ninety percent of the
enterprises out there are grossly under utilizingtheir data and the other ten percent are
almost are minorly under utilizing their data. So everybody's under utilizing their data.
And ultimately, the companies that areable to utilize their data for strategic purposes

(38:52):
are able to gain insights, They'regoing to provide innovative differentiators for them to
accelerate themselves in the marketplace. Itruly think in ten years we're going to
see lots of businesses that have goneunder and lots of businesses that have succeeded
into in a meteoric success, andlook at the differentiators there. How do
they do it. They're able tolabage data the age labor data in the

(39:12):
context of analytics and the context ofAI, and something that provides them of
the core force multiplier for their abilityto take the business. So it's everything.
Yeah, Jim, go ahead.Yeah. The one thing I wanted
to add the David's comment, andwe see it a lot at click,
is you know, there are someorganizations we go out to and they say,
okay, yeah, we don't wantto look at a data analytics tool
yet we've got this twelve to eighteenmonth project to build the data warehouse,

(39:37):
and then we'll come back and lookat a business intelligence tool. And I
think what we tell customers, Ithink it's similar to what David said,
is use a tool upfront that iseasy to get started with, so that
you know what data your users actuallywant to consume in your analytics environment.
And if you can do that quicklyand easily, you can go back and

(39:59):
make that some of the requirements foryour data warehouse projects. So in a
sense, it's kind of and Idon't mean to use an old term there,
but rapid application development. You know, start with kind of some of
the reports, see if they're successful, work that into your data warehouse projects,
so that when your data warehouse project'sdone, you're actually giving information that
users want to see as opposed to, you know, try to do that

(40:22):
all up front in twelve months andthen realize you failed and have to go
back and change that data warehouse.Right, that's exactly right. And just
to put some meat in the boneshere, I'll throw this one over to
Aaron. We've got about four minutesleft in this segment. If you want
to know where the value accrues,where is generated, Look at companies that
come out with new promotions like threewill sell three for a discount of thirty

(40:45):
percent, for example, or buyone now, get one free. Most
of those deals are data driven.Someone has crunched the numbers and figured out,
aha, if we sell this manyat this price point, we'll get
this many new customers. And thenthey test that stuff. I mean they
make sure that working. So youget your idea, you put into the
market, then you test it,and I mean those cycle times are coming

(41:05):
down and down and down. ToJim's point about wraplet application development, new
data products are what people are creating. Think insurance companies. I mean,
you've done a lot of work infinancial services understanding that you can test these
theories out, see how they work, then double down, triple down,
offer it in more regions for example. That's where you actually see the innovations
occurring because of the data, right, Aaron, go ahead, Yeah,

(41:28):
I definitely think so. And Imean that gets to the point of exploratory
analysis, right, because you knowyou've you've mentioned really sometimes you're testing a
specific you know, idea specific hypothesis. You've mentioned a couple of use cases
there. But you also have situationswhere you don't know ahead of time,

(41:49):
you know, I mean the inthe promotions case, right, you may
have a whole lot of factor thatyou could throw into the promotion that could
make a big difference in terms of, you know, sales. But it
could be sitting in front of you. But it could be sitting in the
data. And that's the thing wherethe associated engine might help you to bring
in that you know, that datapoint that comes you know, you know
you weren't looking for it, buthere it is. Yeah, And it's

(42:12):
all part of this process. It'spart of discovery. Never stops. Discovery
is ongoing because market conditions change,you get new products, new services,
You're having to adjust pricing, Imean, pricing is one of these things
that is really under scrutiny right now. I've been paying close attention to inflationary
forces like most Americans are. AndI saw staff the other day that just

(42:34):
jumped off the screen at me,which is that juices like orange juice and
drinks are up forty percent forty percent. And what that tells me is people
running grocery stores have figured out,hey, we can inch that stuff up
because everyone wants their orange juice,right. It's also why they put it
way in the back of the store, so you got to go through everything
to get to the orange juice.But Jim, what do you think The

(42:57):
one thing that you made me thinkabout we're talking about that Eric, was
just the fact that, you know, when you're using a data analytics tool,
it's great to actually see historical performance. You know, you can kind
of see all the visuals to sayhow we've done. What we're seeing a
lot of people are moving towards isbasically the machine learning capabilities within our product

(43:19):
that says, hey, not onlydo we need to know what happened in
the past, we need to knowwhat values are the biggest influencers. And
it'd be nice to be able totake those influencers and kind of propagate out
what's going to happen in the future. And I don't want to just call
it forecasting, because forecasting is hey, I've got a line I'm going to
draw through the data points. Whereit becomes really helpful is when you can

(43:40):
kind of look at that historical informationand then say these are the influencers,
and let me change those influencers tosee what might happen in the future.
And I think with the inflation exampleyou had, that's what we're seeing a
lot of people wanting to do withthe data analytics tool. Yeah, that's
right, to understand what are thevectors impacting our business right now. And

(44:01):
again, the fluidity of that experienceis just crucial because as soon as it
stops, as soon as you haveto go to it to get some other
query or to get access to somedata set or whatever the case may be,
that analytical process is dead. Likeit's just over. Maybe you wrote
it down, maybe you'll think aboutit next week when you come back in

(44:21):
the office. But the point isthat you want it to always enable that
fluidity of interaction with data and understandingof the data, and that's going to
get you somewhere. Well, folks, we got one more segment coming up,
and we have some fantastic questions fromthe audience today. I'll give our
guests a bit of a teaser sothey're ready. Some folks are asking about
Apache, Iceberg and Delta Lake andHoodie and does Click support all these things?

(44:44):
I mean, this is this wholemovement now, a patche Iceberg in
particular, which really took this themarket by storm and generated tremendous traction.
Everyone agreed upon it as a standard. And then Data Bricks announced that they're
buying Tabular, the company that sitson top of it, during the Snowflake
con So yeah, I don't thinkthat was a coincidence. We'll be right
back. You're listening to Inside Analysis, all right, folks, Tom for

(45:08):
the podcast bonus segment here and afantastic inside analysis. We've been talking to
Aaron Wilson of Athena Solutions, JimSmith of Click, and David lnthiccom Ore
Entry analyst of the day. Wehad some fantastic questions today, folks.
We'll be sure to pass these alongto our presenters if we did not get
around to your question. But thereare questions about AI, and there are
questions also about these new table formatslike Iceberg, Apache Iceberg, who do

(45:32):
he is another one? And DeltaLake that's the data break specific one.
And then of course Data Bricks boughtTabular, which sits on top of Iceberg.
So first I'll throw it over toyou, Jim, How does that
fit into the clickworld. Yeah,I mean a click we've got. You
know, the important thing about ananalytics tool or a data integration tool,
and Click has both of those isbeing able to connect to any source and

(45:55):
pretty much deal with any target.And because of that, there are hundreds
of data sources that we can support, both on the analytics side and the
integration side. I don't know thatI can address those in particular, but
I can tell you that we workon adding data sources on a monthly basis.
So we're I believe, rolling outfifteen more data connectors on our data

(46:17):
integration than data analytics side within themonth. So we are really good partners
with Hyperscaler databases, Snowflake data bricks. So as those vendors start providing different
ways to get access to information,I think you'll find Click following along with
that support. Yeah, and David, I'll bring you in. This open

(46:37):
table format stuff is very interesting becausewhat we were talking about earlier context and
being able to leverage new data sources, new types of data. For example,
I mean, one of the challengesis that from an analytical perspective,
SQL queries it's a structured query language. It doesn't deal so well with unstructured
data or some other unwieldy sources likej ON in different formats like that.

(47:01):
But now in these open table formats, you're going to be able to bring
in lots of external data that's goingto improve context and really help kind of
see the big picture. Of course, you have to know how to do
it. But what are your thoughtson all that data. It's extremely valuable,
I mean data unto itself without thecontext of where it exists and what
it means to other data. Imean, like we went in to the

(47:21):
previous example, in other words,the data of erroneous the errors that occur
in production, and how it meansin the context of other things, other
environmental factors that have to come intoplay. And so your ability to find
problems, your ability to understand whatthe data actually means to the business other
than the data as it exists fundamentallyinto itself. Everybody likes to do the

(47:43):
analysis to what sales data means tosales data, that's meaningless to me.
What does sales data mean to demographics? What does sales data mean to the
environment, What does sales data meaninto social media means and are we able
to make the correlations which allows usto adjust the business to actually get the
growth that we're looking looking for.And that's core to what businesses need data
to do. Yeah, that's brilliant, Aaron. I'll throw it over to

(48:05):
you for some final thoughts. Imean, it's getting very exciting in the
data world. It's a lot lessexpensive to play around with the stuff that
it used to be. There aremany more data sources, it's a lot
more fluid. I mean, really, it's it's kind of a golden age
for data. What do you thinkerin it is? I mean, we're
it's an interesting time for so manyreasons. I mean the question about open

(48:28):
table format, I mean it showsyou that, you know, we're kind
of looking at integration challenges all upand down the spectrum. I mean,
there's still a lot of legacy onTREMD that you know, as an implementing
station analyst, you have to beaware of that, but you also have
to be aware of these new formatsas well. But it is a very
exciting time for both AI and analysis. And I think what we've kind of

(48:50):
shown here in this show is ananalysis you know, having a human being
and actually get curious about the dataand go down that exploitatory path. You're
not going to be able. You'renot going to do effective AI unless you've
got people doing that. Yeah,that's a really good point. And I
will use the last minute or sohere to tease our next show and our
past show is also online. Youcan hop online to Insideanalysis dot com to

(49:14):
see past events that we've done there, radio shows, you can watch the
podcast, you can listen to it. We have several new stations that have
picked us up recently in Saint Louis, in Sarasota and Tampa and in Iowa.
I got a bunch of new stationsout there carrying us, so big
shout out to our friends out there. And if you want to be in
the show, send me an emailinfo at inside Analysis dot com. But
our first show is about data fabric, this one is about visualization and click.

(49:36):
Next show is going to be aboutAI. And to points made by
each one of these guests, ifyou want to leverage the power of AI,
you need to make sure that yourdata house is in order and you
want to and click is great forthis just for the reasons we talked about
for knowing the associations between things,for being able to assess which data fits
with which other data set, understandingcovariance, which is the key to analytics,

(49:59):
and to understand any relationships between things. These are all important, they're
all part and parcel. But youdo want to take your time right.
Don't rush into this stuff. Definitelydon't rush into using AI or jenai.
And frankly, one of the bestbits of advice I've heard yet about JENI
is use it internally first. Becareful about using it externally. There are

(50:20):
some great use cases around customer support, for example, but just wait,
be careful, make sure that it'sreally, really good. You'll hear all
about RAG models, retrieval augmented generationvery important. I think that most enterprise
data companies are going to have tofigure out where they fit in the RAG
model. It's going to be huge. I mean, your RAG model is
almost like your operating system for GENAI. It's very important to understand what your

(50:45):
anchors of truth are going to be, what your embeddings will be, how
you use these tools. But aslong as it's for decision support, as
long as it's inward focused and inwardin nature. You're going to be pretty
safe because and the last thing Imentioned. A great friend of mine pointed
this out the other day. MichaelBarras will be on a show sometime soon
around data governance through data catalogs.He said, Remember, you don't have

(51:07):
to be one hundred percent accurate.The systems we have today are not one
hundred percent accurate, but you canfind the mistakes. Now you want to
be at least eighty to ninety percentaccurate, but you can find the mistakes
and then go back and address that. And I've got to tell you,
man, addressing root causes is very, very important. It's going to do
a lot of benefit to your organizationif you understand where the data quality problems

(51:28):
are, how did they get there, how did they get there in the
first place, what's the data onboardingprocess? All this stuff can be addressed
in data governance programs, and that'sreally important before you use AI. With
that, we're going to bid youfarewell, folks, Thanks so much for
your time and attention. Send mean email info at inside analysis dot com.
We'll talk to you next time.By bye. Wishing for a little
bit more information, you'll get it. All here on KCAA Radio. Redland's

(51:54):
Ranch Market is a unique, fullservice international or grocery store that specializes authentic
food items from Mexico, India,and from many Mediterranean and Asian countries,
including popular items from the US.They offer fresh baked items from their in
house bakery, housemade tortillas from theirTortillirea, a delicious array of prepared Mexican

(52:14):
foods, a terrific fresh food andjuice bar, and a large selection of
meats, seafoods and deli sandwiches,salads and hellal meats. Their produce department
is stock full with fresh, localand hard to find international fruits and vegetables
that you cannot find anywhere else.Don't forget to step into the massive beer
cave and experience the largest selection ofdomestic, artisan and imported beers in the

(52:37):
IE. They can also cater yournext event with one of the delicious takeout
catering trades of food. Visit themat Redlands ranch Market dot com. That's
Redlands ranch Market dot com. RedlandsRanch Market a unique and fun shopping destination
to Heblotea Club's original pure powdy Arcosuper Tea helps build red corpuscles in the

(52:57):
blood which carry oxygen into our organand cells. Our organs them cells need
oxygen to regenerate themselves. The immunesystem needs oxygen to develop, and cancer
dies in oxygen. So the Tis great for healthy people because it helps
build the immune system, and itcan truly be miraculous for someone fighting a
potentially life threatening disease due to aninfection, diabetes, or cancer. The

(53:19):
T is also organic and naturally caffeinefree. A one pound package of T
is forty nine ninety five, whichincludes shipping. To order, please visit
to Hebot club dot com. Tohebo is spelled T like tom, a
h ee b like boy. Oh, then continue with the word T and
then the word club. The completewebsite is to Hebot club dot com or
call us at eight one eight sixtyone zero eight zero eight eight Monday through

(53:45):
Saturday nine am to five pm,California time. That's eight one eight sixty
one zero eight zero eight eight toEbot club dot com. With sixty years
of fascinating facts, This is theman from Yesterday and back in time.
We go to this time in nineteenseventy seven on NBC's Midnight Special this week.

(54:06):
Here in nineteen seventy seven, it'sCasey and the Sunshine Band as they
host with Cat Stevens and the BayCity Roller. And from this time in
nineteen sixty five, Nancy Sinatra andTommy Sands divorce. This happens just as

(54:29):
Nancy Sinatra's recording career takes off.These boots on a fuck and that's just
what they'll do. One of theseDaisy Bootstock gonna walk all over. And
from this time in nineteen ninety nine, it's the second Coming of Woodstock.

(54:52):
Performing at Woodstock ninety nine. Onthe second day, the Dave Matthews Band,
Counting Crows, Alanis Morissett and Metallicalost. You soon told you all,
You'll come crash with more at manfrom yesterday dot Com. Hey,

(55:19):
Southern California, you know what's goingon. It's my new hitch show,
The Scene with Dorian and I'm yourhost, Dorian Taylor. Join me every
week as I chat with my celebrityfriends from TV, movies, music,
theater, sports, and everything inbetween. We explore the lives and careers
of some of the entertainment industries thebiggest names, with some fun surprises too.
Tune into the scene with Dorian Tuesdays, it's seven am on KCAA Radio

(55:44):
ten fifty am. I always hearfrom our clients who hired another firm that
they wish they'd hire DNA Financial First. Don't have regrets about your IRS tax
case. Just hire the best inthe first place. One owed one hundred
fifty thousand to the IRS and it'sspent thousands on another firm. We stopped
the lovey's, negotiated a payment planand had their penalties forgiven. And while
every case is different, we guaranteethat we'll find your perfect resolution and get

(56:06):
it done right. For a freeconsultation, call us at eight six six
two zero one zero one five six. That's eight sixty six two zero one
zero one five six. Then youcan say DNA did right buy me?
TV journalist Jane Veles Mitchell here excitedto tell you about my show Unchained TV.
Animals people Planet. Most of ussay we love animals. When we
truly respect them, we benefit witha happier outlook, better health, even

(56:31):
nicer weather. I'll explain when youjoin me Mondays at one pm on KCAA
ten fifty am one oh six pointfive FM. The stations that leave no
listener behind and check out UNCHAINEDTV dotcom. KCAA Radio has openings for one
hour talk shows. If you wantto host a radio show, now is

(56:52):
the time. Make KCAA your flagshipstation. Our rates are affordable and our
services are second to none. Webroadcast to a population of five million people
plus. We stream and podcast onall major online audio and video systems.
If you've been thinking about broadcasting aweekly radio program on real radio plus the
internet, contact our CEO at twoeight one five nine nine ninety eight hundred.

(57:15):
Two eight one five nine nine ninetyeight hundred. You can skype your
show from your home to our Redlands, California studio, where our live producers
and engineers are ready to work withyou personally. A radio program on KCAA
is the perfect work from home avocationin these stressful times. Just time KCAA
radio dot Com into your browser tolearn more about hosting a show on the

(57:37):
best station in the nation, orcall our CEO for details. Two eight
one five nine nine ninety eight hundred. Listening to KCAA lo Melinda at one
O six point five FM K twoninety three cf Brito Valley NBC News Radio.
I'm Chris Gragio. President Biden isdropping out of the twenty twenty four

(57:58):
presidential race after inten pressure from hisown party. Democratic leaders for weeks have
expressed concerns about the eighty one yearold's mental fitness and his path to victory
over Donald Trump. Democratic National Committeechair Jamie Harrison gave his reaction to the
news. I am emotional about thePresident's decision because this President, Joe Biden,

(58:19):
has been a transformational president. He'sbeen a great leader. He's a
good man, a decent man.In a letter posted on x Biden said
it was time for him to stepdown and focus on the rest of his
presidency. Biden also announced he's endorsinghis VP, Kamal Harris, who said
she's honored by that endorsement and saidshe'll do everything in her power to unite
the Democratic Party to defeat Donald Trump, but she isn't automatically the official Democratic

(58:43):
nominee. That'll be decided over thenext few weeks. The Clintons are endorsing
Vice President Harris. In a post, the couple said they'll do whatever they
can to support her. The Clintonswent on to thank President Biden for his
service. Recent polls are showing aclose race between Donald Trump and Vice President
Harris should she be the nominee.A CNN poll of poll's average show Trump
has forty eight percent support, Harrisholds forty seven percent. Third party or

(59:07):
independent candidates were not included in thepolling average. This is the first time
a sitting president hasn't sought re electionin more than half a century. Democrat
Lyndon B. Johnson dropped out innineteen sixty eight, but his circumstances were
much different. He was in atight primary and the DNC was months away.
Biden drops out with just four weeksto go until the DNC. This

(59:28):
week kicks off the release of anew postage stamp. Tomorrow, the United
States Postal Service is honoring longtime Jeopardyhost Alex Trebek on what would have been
his eighty fourth birthday, the daytimeEmmy Award winner died in twenty twenty from
pancreatic cancer. The new first ClassForever stamps design is in the form of
a Jeopardy style question. I'm ChrisCaragio, NBC News Radio, NBC News

(59:51):
on CACAA Lowel sponsored by Teamsters Localnineteen thirty two. Protecting the future of
working families, Teamsters nineteen thirty two. ALCOH thank you for tuning in for
this addition of