Nov 10, 2020
Episode Transcript:
Rob Collie (00:00):
Hello, again, data people. Today, we welcome Kevin Overstreet to
the show. Now, Lon Chaney was known as the man of a thousand faces.
Well, Kevin Overstreet, I call him the man of a thousand jobs
because he has had almost every job you can imagine from horse
trainer, to analytical chemists, to dynamiting mining, making roads
construction guy, and all points in between. But for the past six
years, Kevin has been a Power BI and Power Platform professional.
He works at Eli Lilly, and he provides a tremendous amount of
essentially like anything glue, to make everything run.
Rob Collie (00:37):
We crossed paths a long time ago when he came to one of our Power
BI classes, one that I taught in Cleveland, but he's not a one
trick pony. He is really up to his eyeballs in the entire Microsoft
stack, especially in the Power Platform, and he is a really
entertaining conversation, and I hope you enjoy it. Let's get after
it.
Announcer (00:58):
Ladies and gentlemen, may I have your attention please?
Announcer (00:59):
This is the Raw Data by P3 Podcast with your host, Rob Collie and
your co-host, Thomas LaRock. Find out what the experts at P3 can do
for your business. Go to powerpivotpro.com. Raw Data by P3 is data
with the human element.
Rob Collie (01:17):
All right. Welcome to the show. Kevin Overstreet.
Kevin Overstreet (01:21):
Hello.
Rob Collie (01:22):
How are you doing, Kevin?
Kevin Overstreet (01:23):
Good.
Rob Collie (01:24):
As usual, you're joined by myself and Tom, or Thomas LaRock,
depending upon, whether you use the professional version or not.
Well, Kevin, let's get right to it. I'm really excited to have you
on the podcast. I think it's going to be a real treat. Let's start
here. Why do we know each other? What do you do these days that is
relevant to data and the Microsoft data platform and business
application platform? Just give us a quick summary there.
Kevin Overstreet (01:53):
I work at Eli Lilly. Although I work at Lilly, all my opinions are
my own, but one of the things that we do is we deal with massive
amounts of data every day. Previously, years ago I was doing that
in Excel and then one day I was searching for how to figure out a
distinct count of items in a pivot table, and that was where I
stumbled upon the Power Pivot Pro blog.
Rob Collie (02:19):
Oh yeah. What year roughly was that?
Kevin Overstreet (02:23):
I think that was around 2010, 2011. It wasn't very long after Power
Pivot was first released and you had started the blog that we kind
of gravitated and found each other. At least I found you at that
point.
Rob Collie (02:37):
That's been a long road, hasn't it? We like to joke about, and you
like to joke about, you being patient zero at Eli Lilly for this
stuff. There's a lot of people who started out in Excel and found
their way to this stuff. That's no accident. You're not the usual
Excel story. What's your trained discipline. If you're at a
pharmaceutical conference, how are you going to introduce yourself?
What's your background?
Kevin Overstreet (03:04):
At one point, I was an analytical chemist. Now, before that, I was
actually in the department of immunology at the University of
Kentucky and we were doing brain tumor research there. Technically,
I'm a biologist/immunologist.
Rob Collie (03:18):
Yeah. Some really lightweight stuff going from immunology and
cancer research to analytical chemistry, which I don't even really
know how that differs from chemistry, but it's got a cool adjective
on the front of it, doesn't it? Yeah.
Kevin Overstreet (03:34):
It does.
Rob Collie (03:35):
But as a personal matter of interest, I call you the man of
thousand jobs.
Kevin Overstreet (03:40):
That's true.
Rob Collie (03:41):
You've already dropped some serious interest there, but everyone's
got a different path to this stuff. You probably have a unique
path. What other jobs have you had Kevin? Just like a softball of a
question.
Kevin Overstreet (03:51):
I did not have an IT background, so I grew up on a horse farm in
Kentucky where we had about 300 horses when I was growing up. I
think my first job was probably hauling hay and cleaning out horse
stalls for my dad at 50 cents a day, which he docked us on if we
didn't think he had done enough. That started at about
12-years-old. A friend of ours had a pizza parlor. So, I was in
charge of the big, giant, deep fryer, this high pressure deep
fryers, which you probably couldn't do today.
Rob Collie (04:25):
Every pizza parlor needs a gigantic deep fryer, yes?
Kevin Overstreet (04:29):
That's true.
Rob Collie (04:30):
Because you need what? Fried pizza.
Kevin Overstreet (04:32):
Whether anybody remembers this is definitely me dating me, it was
the shaky pizza parlor. Not only did they have pizza, but they also
had fried chicken.
Rob Collie (04:41):
Oh, okay.
Kevin Overstreet (04:42):
So, I dipped my hands down into the hot grease to fried chicken at
12-years-old. Then around 14, we built a racetrack on the farm and
started training horses. So, I was training horses there for
several years. We've had about three world champions come off of
that farm. There, I went, while I was in college, I was a security
guard at Rupp Arena. We did security for all the concerts and
everything there. I hung out with a Rat one night, if anybody
remembers Rat. During time off from college, I worked for a
communications company that was installing all new phone system at
Auburn University.
Kevin Overstreet (05:23):
I worked for a construction company where we were installing a
water main in the middle of winter, which was a little chilly. The
great part about that was it was over an old strip mine, so we had
to dynamite out holes for the pipe, and probably not quite as many
safety measures back then. Our safety measure was taken off running
and as fast as possible and waving our hand because we were setting
off the dynamite with telephone wire on a car battery.
Rob Collie (05:52):
Really just your average usual, like in Austin power, it's just
your typical usual stuff really.
Kevin Overstreet (05:58):
That's it. Yeah. Nothing to see here.
Rob Collie (06:00):
Nothing to see here. And despite all of that, the last decade,
basically.
Kevin Overstreet (06:05):
Yep.
Rob Collie (06:05):
You've been in our world, and you left all the dynamite and
analytical chemistry behind. What does an average work week look
like for you? What are the things that you're doing with the data
platform?
Kevin Overstreet (06:16):
We have a massive SharePoint installation. I guess one of the great
things was, was just about the time I was discovering Power Pivot,
I was also discovering the power of SharePoint. It had kind of
always been this black box at Lilly of this is our sharing platform
of the future, and then they put out about three PowerPoint slides
for how to use it, which is probably not a whole lot different, but
I'm a huge SharePoint fan. At this point, we manage a lot of one
off processes through SharePoint.
Kevin Overstreet (06:50):
Not only the data management, but the data input, sort of whatever
that user facing front end is. We're doing that with Power apps.
We're doing huge amounts of things with Flow. Flow has been
incredibly powerful. We've taken processes that used to take us 30,
40, 50 hours a month and now have completely automated that using
Microsoft Flow. Those have been incredible opportunities, things
that we found. But on a daily basis, probably 95% of my day is in
one of the Power Platform tools. I'm either in Power apps, Flow or
Power BI. And we sort of see all three of those things going
together in any solution that we deliver.
Rob Collie (07:33):
How does that workload, the things that you're working on, that
95%, how does that contrast with your sort of official position,
your official job duties? It'd be hard for me to imagine that your
official job duties are defined as that 95% that you're currently
doing.
Kevin Overstreet (07:52):
That's always been a little weird for me no matter where I'm at. My
job description, when I was an analytical chemist, I was still
doing a lot of data management there and building tools in Excel.
At that time, of course, it was all VBA and pivot tables and moving
information around. But I guess I've always had an interest in
that. When I was, technically, I was still an analytical chemist.
At that point, I had already moved into a data role about 50% of
the time.
Kevin Overstreet (08:22):
Yeah, now, for the most part, I'm known as the SharePoint guy, or
the Power BI guy, or the Flow guy, whatever it is, typically, if
there's something going on or somebody has a problem, eventually
those things find their way to me. Because I have been using it,
like you said, for about 10 years now.
Thomas LaRock (08:39):
Kevin, the question I have, because I'm always fascinated by data,
because first of all, I think you have a fascinating background.
Love the accent. Before I forget, I wanted to mention, since you're
Kentucky and horse racing, all that, I was really happy to see Bob
Baffert win the Kentucky Derby, because if there's ever a guy in
this world that needs a lucky break and something good to happen to
him, it's Bob Baffert, so I just thought you'd appreciate that.
Thomas LaRock (09:07):
Anyway, tell me something about the challenges you have with data
over Eli Lilly, without getting really too. But I feel that you're
going to share something, myself, Rob and our listeners will just
sort of be like, yep, that's exactly what I go through on a daily
basis as well. Tell me some of the challenges that you have.
Kevin Overstreet (09:24):
Sure. I mean, a lot of the challenges that we have are just
compiling data. People have a lot of requirements, and for the most
part, people are comfortable in Excel. I think in any company
you're at, you're still going to see massive Excel reports that are
out there that people are producing every day. Like most companies,
we are aggregating massive amounts of data on any given day. We're
compiling that into reports that are delivered either to specific
people or maybe to a common location where other people are picking
those up.
Kevin Overstreet (09:59):
And everybody talks about a golden data source, but I think that
becomes even more important now as these tools get stronger and end
users can begin to become more accustomed with them, they start to
take those reports and files and do lots of crazy things with them.
I think that's one of the things that the Power Platform gives us,
is that way to sort of standardize those processes and deliver kind
of a consistent product, no matter what that is, whether that's a
visual or whether that's a report.
Kevin Overstreet (10:32):
Now, the other thing that I'm sure a lot of people struggle with,
and especially in SharePoint, if you think of something as simple
as document metadata. So, everybody is tagging documents with
metadata and companies can have hundreds or thousands of SharePoint
sites across that, and of course everybody's tagging it
differently, and that can be project name, whatever it might be,
depending on where you're at.
Kevin Overstreet (11:00):
But one of the things that we've been able to do with Power
Automate is basically create one SharePoint site that we consider
our master metadata and then we populate that out to every other
SharePoint site sort of in our functional area there. Now,
everybody's using the same metadata. So, when it comes time to
compile that data, it's very easily to take Power BI connect into
those SharePoint sites and be able to pull together a really
consistent list of information. Because if not, over time, people
record things differently, they use different terms, but then at
some point somebody's going to come and say, "I want all of that
data." And then that's going to result in a two or three week
sprint of everybody trying to decipher all the different ways that
data's been stored over, possibly years.
Rob Collie (11:51):
So how? The naïve way to look at it would be to say, well, of
course, Kevin, you were in charge of all of these SharePoint sites,
and you came up with the Schema, that was standard. Then you just
rolled it out because everyone had to listen to you. I just don't
think that's how it happened. How do you heard the cats like that
to get everyone on the same page with such a thing?
Kevin Overstreet (12:12):
It's hard to get people on the same page, and I don't think by just
rolling something out and saying, here it is, you're not going to
get much uptake on that. People are locked into their old ways of
doing things, no matter how any efficient that is, you can go to
somebody and say, I'm going to save you 50% of your time. They go,
that's okay. I'll just stick with what I've got, because they're
uncomfortable doing something new. I think what you have to do is
you have to deliver something, and this does end up in a very
iterative approach sometimes, but you have to deliver something
that has a personal impact to people, where they can physically
sort of feel that change of, oh, this is going to make my life
easier.
Rob Collie (12:56):
Deliver, then explain, rather than explain than deliver.
Kevin Overstreet (13:02):
Yeah. I tend to talk in the air a lot and people can't necessarily
follow that when I'm drawing on my virtual whiteboard that's in
front of me. When people come to you with an idea, they're basing
that idea on what their experience is. They'll start with things
of, I need a form that does this, or I need an Excel file that does
X, whatever that might be. A lot of times they're just going based
on what experience they've had and what they know is out there and
may not even know what the possibilities are.
Rob Collie (13:37):
The reason I said that deliver then is it hearkens back to
questions that I've been asked over the years, which is ... I get
asked this less often lately, which is actually really encouraging.
But for a long time, I would be asked, hey, how do I get buy-in in
the broader organization or in my sister organization, the peer
organization, how do I get them to buy into the power of, let's say
Power BI? The advice I would give them is don't try to play, my
software can beat up your software. Don't try to explain it like
it's a conflict of two tools.
Rob Collie (14:13):
Just go and build something that was impossible. Go build something
that was impossible and then give it to them and say, see, this is
it. It sounds like your philosophy, even though I don't think you
and I have ever actually talked about this. It sounds like your
philosophy has found itself with that exact same method.
Kevin Overstreet (14:31):
Yeah, I think so. Over time, before this was kind of my official
role, the tools that I built were for things that annoyed me. I
don't want to do that anymore so I'm going to build a tool and
that's what I always tell people now. If I have to do it more than
twice, I'm going to automate it in some way. That's it. I can
remember we built just a scheduling tool and it was because I
didn't feel like walking around on the floor and writing my name on
one of 50 whiteboards for a reservation when we've got ... We're
all setting at our computer, at least half the day, why not build a
reservation system in SharePoint?
Rob Collie (15:08):
That's great. There's so many things that are interesting here. One
of them is, is that a lot of analyst types, I would count myself in
this, a lot of analyst types, we kind of want to join the story
after the data has been created. I gravitate towards, let's go look
at what you're already recording about your business and see what
that's telling us. Of course that's super valuable. That's, in a
lot of ways, central. That's what BI is.
Rob Collie (15:41):
One of the things that I really like about you is that you don't
perceive that artificial boundary. You're really willing and
excited about engaging upstream, even in like what I would think of
if I wanted to really paint it with a broad brush, I would say that
you're involved in the data collection process, the data creation
process, like you get into the workflows that are upstream from
analysis, and I'm going to going somewhere with this overall theme,
but I want to stop and see what your reaction is to that.
Kevin Overstreet (16:17):
I do agree with that. I think that the one thing that I've always
liked is solving problems, and I think as I go back, and we were
talking about all my jobs before, I think that, especially as I've
gotten older, the jobs that I really joyed, even if the sort of
subject matter was different, was the ability to solve a problem.
Whether that's developing a new analytical method, or in this case,
developing a new process or form, but sort of relating that to
your, I don't want to come in till the data is there.
Kevin Overstreet (16:52):
I really got started in collecting the data for ... Someone was
asking for me to analyze data that was in no shape for me to be
able to analyze. So, it took so much manual preparation. It was
like, I'm just going to go in and build something that's going to
collect the data the way that I want it and the way that I need it.
I think Power Pivot kind of revved that up a little bit more in
2010, when all of a sudden, now I can deal with millions of rows of
data and I can deal with millions of rows of data in SharePoint.
Okay, now I have no limitation to how much data I can record.
Kevin Overstreet (17:27):
I'm just going to record everything and I'm going to figure out
what I'm going to do with it later, because I think there's value
there.
Rob Collie (17:33):
Tom likes to talk about no one set out to be a data janitor.
Thomas LaRock (17:37):
That's right. No one went to school to become a data janitor.
Rob Collie (17:40):
That's right.
Thomas LaRock (17:41):
And you replied to me, and yet here we are.
Rob Collie (17:44):
Yeah, here we are. There's a little irony in that. At one point in
his life, Kevin was cleaning out horse stalls. I think the Power
Platform's got a lot to it, but I'm not sure that would really help
you with the horse stall cleaning.
Kevin Overstreet (17:59):
It probably would with the horse racing. I've gone back and I've
talked to my dad now about, oh, if we'd only had these tools back
then, the things that I could do. Because I can remember my first
computer was like a 386SX with a whapping one megabyte of Ram and a
40 megabyte hard drive, and I think I spent about $200 to update
that to two megabytes. I was using Quattro Pro and dBASE III Plus.
I was trying to do all these things with horse races.
Kevin Overstreet (18:32):
Even back then, I think I kind of had an interest, but in my
everyday life, I don't find anywhere that you can apply some of
these tools. We've had that conversation before about, anybody who
has any small business at all, now would just be crazy not to jump
into these tools and get a subscription because of everything that
it can do for you.
Thomas LaRock (18:54):
What I like hearing is you talking about a success story and you're
combining a couple of tools, Excel and SharePoint. Like you said,
back in 2010 was the, I'll call it the first of a game changer,
we'll say, right?
Kevin Overstreet (19:10):
Right.
Thomas LaRock (19:10):
Because before that, in my background being a production DBA for
seven years, if anything touched SQL server, all of a sudden it was
mine. SharePoint being one of those things. Back then, SharePoint
was ... It was ugly. It was a mess. It wasn't something that will
have the scarves. You say SharePoint, and I'm like, just get away
from me, man. Just [crosstalk 00:19:34]. When I hear those stories,
the thing is, is at the time, those stories didn't exist, or if
they did, they didn't really resonate with me. Hearing it now, I
have so much more appreciation for the collaboration that, that has
fostered.
Thomas LaRock (19:49):
Here we are 10 years later and we see these tools, and I see them
in a much more positive light. Tell me some of those success
stories for you where you are. Have you seen that same thing where
it really fosters more of a collaboration, there's no real stigma
with it. Nobody's saying, oh God, you're using Excel or you're
still writing code in Python, that type of gate keeping, like it
really has helped foster and made things better.
Kevin Overstreet (20:18):
I think so. I mean, especially in that kind of citizen developer
role. You, as the SQL guy, you had access to plenty of SQL servers
and everything, where to store your data. When I first start out,
SharePoint was just the only tool that was available to me. I had
zero IT background. Other than some fairly basic VBA, I wasn't
going to go in and write crazy SQL code or anything. I just started
using the tools that I knew how to use at the time, which started
to be SharePoint and then Excel, and then, as those things, then
Power BI came along, and then now the Power Platform.
Kevin Overstreet (20:59):
I think those things just build over time. I know that Rob is
constantly talking about people with the data gene. I think that's
what you find. I mean, I still think SharePoint has a lot of
detractors no matter where you're at. But I think part of that is,
SharePoint is still kind of the wild west. If you're going to let
people just create sites and do whatever they want, there's going
to be some crazy things that are going to happen in there that are
going to frustrate you for sure.
Rob Collie (21:27):
Well, now that SharePoint has moved largely to the cloud and online
and it's Microsoft servers that's running it for most people, all
those problems that you're talking about, Tom, they don't come to
you anymore. But yeah, I completely understand. As someone who was
sort of like thrust into managing a SharePoint firm 2010, because
there was no alternative, oh my gosh, this is something that you
would give to your enemies, was managing that.
Rob Collie (21:57):
From a SQL perspective, nothing could have been more unclean.
There's no abuser of SQL quite like SharePoint.
Kevin Overstreet (22:03):
Well ...
Rob Collie (22:03):
Okay. Okay, people. Second to people, is that other thing.
Thomas LaRock (22:13):
There's still some products. There's other products. We don't have
to name them.
Rob Collie (22:19):
Yeah. But really, SharePoint's like ... That's a product that's
really kind of come into its own, I think. The stuff from 10 years
ago is not remotely relevant anymore. But boy, that was not a lot
of fun back in the day.
Kevin Overstreet (22:32):
I think people get overwhelmed by SharePoint when they begin to
learn how to use it, because you search and there's all these
things. I'm just a basic guy. I've never gone, I've never written
one piece of custom code in SharePoint. Everything I've done has
been out of the box. I've never created a content type. Everybody
talks about content types and SharePoint. I've never created a
content type in SharePoint. Could not tell you how to do it.
Typically, whatever's there is what I use.
Rob Collie (23:02):
Kevin, it just now struck me, you remind me of Phil Hartman's
character on Saturday Night Live, the unfrozen caveman lawyer. I
don't understand this stuff. I'm just a simple guy. I've never
created a content type in SharePoint. I've never written any custom
SharePoint code. However, I've been a cancer researcher. I've been
an analytical chemist, and on top of that, I've blown things up
with dynamite. You didn't drop this earlier, but I'm going to drop
it for you. You've won every horse training award that there can be
won that's awarded by the state of Kentucky.
Kevin Overstreet (23:36):
That is true. Yeah.
Rob Collie (23:38):
You're just a simple man. You're really just an every man. We're
all basically you, is what you're really trying to tell us.
Right?
Kevin Overstreet (23:44):
That's it. I'm a basically A to B kind of guy.
Rob Collie (23:49):
Yeah, that's it. That's it. Simple. So simple. If only everyone
were as simple as you Kevin. These podcasts always end up with at
least one small, and not necessarily always small, but there's
always some moment of Rob monologue, and I'm aware of this. Here it
is. We're entering the monologue zone. One of the things that I
have been witnessing and watching as a professional over the last
10 years, especially recently, is this notion that BI, business
intelligence, analytics, whatever you want to call it, it's been
its own thing. It's been its own discipline for a very long time,
but that's only because it's been so bad.
Rob Collie (24:36):
When something is really bad and it hasn't delivered on its
promise, or even on its need, a lot of attention gets focused on
it, and it becomes a thing, almost like a goal in and of itself.
BI, the I stands for intelligence. Why? Why do we need to be
intelligent? We only need to be intelligent because we need to
improve. We need to improve our business. I'd like to mark today,
the small number of us, we try to start a movement where we rebrand
the eye from intelligence to improvement.
Rob Collie (25:13):
Being informed is only meant to power improvement, and this is
where I think that the story comes back around to you, Kevin.
Because I think you don't come into this story with this like, it's
BI or nothing mindset. It's just problem is problem. I talked
earlier about how a lot of your activities, a lot of things that
you do are about optimizing, even upstream of analysis. You're
saying, hey, in order to get a better picture for our business, we
need to improve the data collection process.
Rob Collie (25:45):
I think that's just incredibly noble. That is the right mindset. At
the same time, when you're looking at a dashboard or report,
whatever you want to call it, all the nouns, get confused, but
really people think of them as dashboards, when you're looking at a
dashboard and you see something that needs fixing, or you see an
opportunity that you want to go after, there's also a downstream
from the report, from the dashboard.
Rob Collie (26:11):
I like to talk about this as the action loop. There's upstream, and
then there's noticing an opportunity or a problem, and that's what
BI does. It helps you notice things, and then you take action on
it, and then you see if the action worked. Did it actually have the
desired impact? How much, if any, work have you've done with the
Power Platform that's extending sort of downstream from power BI.
We talked about upstream, the collection of data and things like
that. What about downstream? What about the taking action phase?
Have you done anything there? It's okay if you haven't, I'm just
actually curious.
Kevin Overstreet (26:43):
Yeah. I mean, I think, if what you're talking about, we've started
to implement more Power apps into dashboards so that people can
annotate data in real time, in some cases, depending on what your
application is, you're limited by your dashboard refresh rates. And
other times, you can show that information in real time right in
the Power app. I think, as we move forward and those tools continue
to mature, and we begin to learn more how to do those things, that
that's going to be the thing. Because no matter what, if you've got
a number in the box, there's always context around on that.
Kevin Overstreet (27:23):
You're never going to be able to provide all of the context that
everybody wants in that dashboard. So, giving people the ability to
add comments to it, and then make that available in a popup in the
dashboard, a popup tool tip where somebody can hover over something
and see comments that people have left over time or questions that
they may have had, or even to the point where we implement a Power
apps and SharePoint sort of solution in order to be able to correct
data that we're getting from other places.
Kevin Overstreet (27:57):
Basically, we're looking, if there's a value in our tool, we're
going to take that value over the value that we may be getting for
from another place. I think there's lots of opportunities to sort
of affect that in both ways, but yeah, particularly upstream, and I
think those opportunities are just getting bigger all the time.
Rob Collie (28:15):
You make that sound so easy.
Thomas LaRock (28:17):
He really does.
Rob Collie (28:19):
It's just like second nature. My grandfather, who you would've
liked, had a saying, which was like, it's very difficult to fence
in an earthworm. You, Kevin, are the earthworm.
Kevin Overstreet (28:31):
It's true. Could be.
Rob Collie (28:33):
Oh yeah, this is the box. No, no, this is not a container. I am
going to show you that we can solve anything. Yeah. You should be
the poster child for the entire Power Platform.
Kevin Overstreet (28:45):
A lot of times, everybody, they look at these things and go, oh my
gosh, that's so complex. How did they do that? The very first form
that I ever built that I integrated into Power BI was just a
customized list form from SharePoint where I just went in and said
customized form, and then just went in and figured out how to re
place the data sources with the data from Power BI. These don't
have to be complex solutions to sort of generate huge value and
change.
Rob Collie (29:16):
Kevin, I'm going to ask you this question very delicately.
Kevin Overstreet (29:20):
Okay.
Rob Collie (29:20):
Have you ever considered a career in consulting?
Kevin Overstreet (29:26):
I think I'm too old now. 10 years ago, 15 years ago, I may have
considered that. Now I'm getting older. I think more importantly,
my dad's getting older, so at some point, somebody's got to go back
and start hauling hay again, and I guess that's going to be me.
Rob Collie (29:41):
Ladies and gentlemen, Kevin is 37-years-old. Someone's got to haul
that hay. I mean the hay doesn't haul itself.
Kevin Overstreet (29:56):
Yeah. I always say, I'm 53 now, but I really feel 43 because
there's like 10 years in there that are a little iffy.
Rob Collie (30:03):
And who are you if you haven't had a 10 iffy?
Kevin Overstreet (30:07):
That's right.
Rob Collie (30:08):
You've got to have that.
Thomas LaRock (30:10):
What really just resonated with me just now is how Kevin, you
mentioned, it doesn't have to be complex. All you did was populate
that list, but that made a difference for members of your team and
your organization. It wasn't complex. It was just that nobody else
was going to do it. So, it's part of you being the person that, not
only knows what needs to be done, but is willing to roll up your
sleeves and go do it. I think a lot of that just goes back to who
you've described yourself as a person. Cleaning out the stables as
a kid, working the deep fryer, everything you've done.
Thomas LaRock (30:48):
These aren't glorious jobs, but it's work, and it's work that
somebody has to take on, and you're the guy that's just, well,
yeah, I'll see if I can do it. I think, for a lot of people, they
kind of come at things just a little bit differently. I don't know
what we need or I know what the problem is, but I don't know how to
solve it. I don't know where to go. And they're just not willing to
roll their sleeves and read a manual type thing.
Kevin Overstreet (31:11):
Yeah. I do, do a lot of reading. I've said before, I think I've got
every Power BI book that was written, probably not recently, but
from about four years ago, historically, I probably had every Power
BI book that had been written, and three of those were autographed
by the author. The chemist may be the biggest geek in the room
whenever I'm around. But yeah, I think that people get caught up in
those processes that they're in every day. It seems overwhelming to
try to change from that. Pat, one of my colleagues at work, he
always says, look, if you take 10 minutes today and learn
something, it's going to save you 30 minutes, and then take that 30
minutes tomorrow and learn something else.
Kevin Overstreet (31:58):
Then, eventually those things are going to kind of snowball, and
even when you were talking about problem earlier, you don't
necessarily have to solve the whole problem, but you can always
start by solving part of the problem. If you see that a solution to
a problem is emailing around an Excel sheet where 50 people are
making changes in it, then somehow you're supposed to take all
those versions and put them together in some coherent way, there
are ways to fix that. A lot of times, I think the thing that you
struggle with the most is probably just behavior change.
Rob Collie (32:33):
It's always those people. Those people, they just don't do what
they should, do they?
Kevin Overstreet (32:39):
They won't listen.
Thomas LaRock (32:40):
It's always a people problem.
Rob Collie (32:41):
It's always a people problem. I agree. We identify as technology
professionals at our own peril. What we are, are people who are
able to use technology in a human context. If you start identifying
yourself as a technology professional, you kind of are already
defining yourself into failure in a way.
Thomas LaRock (33:01):
Yep.
Rob Collie (33:02):
I did have one other thing I was thinking while you were talking
just in Kevin, the way you were talking about these problems and
how just sort of the way you approach them and everything, it's not
a risk to say that a large percentage of the people who are
listening to this suffer with imposter syndrome, this belief, this
uncertainty that, maybe I'm just not that good. Maybe I'm not all
those kinds of things. This comes from humility. It comes from
questioning yourself. You really can't stop doing those things.
Rob Collie (33:31):
You always need to be questioning yourself. Otherwise, you never
grow. You don't treat other people correctly. It is the way to be.
But it does, it results in imposter syndrome. What I've found over
the years, and what we found as a group, is that the only antidote
to imposter syndrome is understanding, I'll use Kevin as example,
understanding that people like Kevin, or yourself, whoever's
listening, that most people aren't like you.
Rob Collie (34:00):
The things that you're saying, Kevin, you do, you make them sound
so simple. Of course, it's just this kind of problem and we're
going to approach it this way and we're going to solve it. To you,
that doesn't sound that special. It's obvious. It's almost like
table stakes to you. I'm not going to say there's only one you,
even in your workplace, but there aren't that many. You stand out.
People come to you from all the way across the organization, people
who don't even work in the same building. Well, we don't work in
buildings anymore, do we?
Rob Collie (34:28):
But back when you worked in buildings, they would come from other
buildings to seek out your help and your advice on whatever their
project was. That's because you do stand out. Our team, our whole
company, we're populated with people who do experience imposter
syndrome despite being very, very good. You just got to understand
that there is something, even though you're questioning yourself
all the time, you are kind of special because no one else is really
doing those same things.
Kevin Overstreet (34:55):
Well, I appreciate that. I mean, we've had this conversation before
that, I don't feel like I'm been doing things that are that
complex. I've told you, when I get on your site or I get on other
sites and I look at the things that people were doing and the DAX
code that they're writing, even like Pat, my colleague and friend
that we talked about earlier, he has a blog post on your site where
he calculated like 10 billion nations for like Texas Holdem or
something. I'm not capable of doing those things.
Kevin Overstreet (35:28):
I kind of always have an interest in stuff, so when something new
comes out, I at least want to investigate it. Although I may not be
able to solve something immediately and I'm not going to write
these incredibly complex solutions, I think that over time, I have
found approaches that work to problems and you kind of try to
categorize those, and then use those proven approaches that you
know are going to work. I mean, to this day, and especially now
with Power Automate, I still do a lot of stuff with Excel and Power
Query, where I may need to pull information from one location and
transfer that to another while I just make a Power Query connection
to the data location.
Kevin Overstreet (36:13):
I can then drop that Excel file into a SharePoint and have Power
Automate, go and create those items as many places as I want
automatically. It's just sticking with those things that I know
have worked over time.
Rob Collie (36:28):
Do you have access to Dataflows? Have you had any exposure to
it?
Kevin Overstreet (36:33):
We do. Yeah.
Rob Collie (36:34):
It seems like this is your next place where you just catch
fire.
Kevin Overstreet (36:39):
Yeah. So, we've started exploring Dataflows a little bit. The other
thing that I've started to get interested in is basically creating
a Power BI data model that you can then publish into the Power BI
service, and then that can be used as a data source for your power
app. Now we're taking information that maybe is in all of these
different locations, and we only need a little bit of that, but we
need a little bit from every location. Now we can just use Power BI
to grab all of that, throw that together and basically create a
data source for a Power app or for Flow to operate against.
Thomas LaRock (37:21):
What you've just described to me, Kevin, is essentially nothing
short of a nightmare. Because now, now you've got data everywhere,
right? That's going to lead to other problems. Now you're at risk
of somebody takes that power BI, that project, and they bring it
home with them, or they put it on the USB and they leave it on the
bus. How do you handle that situation? Because I mean, you've done
a great job of curating your data, collecting it, moving it around,
getting it to the right people. But how does Lilly make sure it
doesn't walk out the door. And don't give away any trade secrets,
but obviously it must be a topic of discussion.
Kevin Overstreet (38:00):
I can't speak for Lilly. I can only speak for just sort of in
general, but I think that absolutely security has to be a top
concern, whatever that is. I mean, I know now that a lot of
companies, you can't even copy information to external drives. That
happens. I've been through this discussion multiple times, because
a lot of the times, I cannot necessarily get the access I need. I
may have access to a reporting portal where I can get all the
information I need, but I can't get direct access to the data. So,
I developed some process to get around that, to get to that data
that I need.
Thomas LaRock (38:41):
La, la, la, I'm not hearing that. I'm just going to put fingers in
my ears.
Kevin Overstreet (38:46):
Here's the thing I would say. I always say the security that we
need to worry about is not necessarily at the system levels, or
even at the Power BI level where somebody else is controlling
security. I mean, no matter what company you are at, there are
millions of Excel sheets floating around through email and
SharePoint and networks and wherever else it may be. I mean, I
always think that, that's probably a bigger security concern than
putting some in Power BI.
Rob Collie (39:19):
Oh yeah. Tom's reaction to what you were just saying. I know Tom
was being tongue in cheek, but at the same time, it is very much
the stuff of IT nightmares, the earthworm that can't be fenced in.
You're out there every day generating systems. You're generating,
whether you think of it as code, sometimes you do, sometimes you
don't, it's all code. You're generating solutions every day. I
think the real problem here is that, as soon as something becomes a
solution, as soon as it crosses a certain psychological threshold,
now it sounds like IT's problem.
Rob Collie (39:59):
Whereas the spreadsheets don't most of the time. They don't sound
like it's problem. Spreadsheets seem like they're the businesses
problem, until those spreadsheets integrate with some sort of
backend system in some sort of Rube Goldberg way that IT didn't
anticipate, those spreadsheets are the business problem. But you
start building things like automation workflows that go from
SharePoint to this, to that, and then you decide to go back to
bailing, hey, you go back to the farm. What's left behind is
something that someone on the business side created, you, and now
you're gone, and IT, they're not necessarily up on those
languages.
Rob Collie (40:40):
They don't have what was in your brain. Now it's their
responsibility. That's where the perceived trouble starts, or
that's where the fear creeps in. Even if you're still there, you
haven't gone back to bailing hay, you're not cleaning stalls.
You're still just an analytical chemist masquerading as a data
platform specialist. That's what you are today. What you've done
though, has reduced the total surface area of problems for the
broader organization. It's just done it in a way that now it starts
to feel like it's IT's problem. That's a really interesting thing
for us to come to terms with.
Kevin Overstreet (41:17):
I that's always going to be the case. Like I said, I'm sort of in a
weird position where anybody who's on the business side sees me as
IT, and anybody who's on the IT side sees me as a renegade or a
cowboy or something, depending on the solution. Eventually, you've
got to build those relationships with the people in IT who can help
you get that done, I think, if you want to move forward, and
especially, let's face it, eventually at some point, if you're
doing enough, you're going to need IT's help to get some of that
stuff done.
Kevin Overstreet (41:55):
Speaking from, on my standpoint, I'm not going to be able to go out
and build some SQL server solution, even though that may be what I
need. Now, I may be able to figure out how to populate that once
you've got it for me, but you have to build that relationship with
IT. I think part of that is, is just over time, I think, as I've
began to develop more complex solutions, they've become more
comfortable with me because I've been kind of doing it for so long
now.
Rob Collie (42:24):
There's an old blog post, it's several years old now on our site,
something about the Twilight Zone, where people who work for our
company talk about how they'd found themselves in a similar spot.
They didn't really belong to either world. The business viewed you
as IT, IT viewed you as renegades. What I view you as is the
future. This is where it happens. What you're talking about, where
you're gaining acceptance over time from IT. You're seeking and
increasingly winning their support. That's where things are
headed.
Rob Collie (43:04):
You, off on your own, building these solutions without any IT
visibility or any it support, that's bad. But the answer to that
isn't to have you not doing it. Having you not doing it is worse.
The right answer, in my experience, is that it needs to recognize
you, people like you as their greatest ally, the ally they've never
had. It's like having a bigger organization. There's more IT people
out there. They're not going to get headcount to go and like triple
the size of their IT org. But if they embrace people such as
yourself, you're not exactly dime a dozen, but there's enough of
you out there in the business.
Rob Collie (43:52):
If you were all embraced and sort of deputized and treated as
ambassadors between the business and IT, IT will have a much
greater capability, not of just to play defense, like they
typically are required to, they can play offense with your help.
They can be part of advancing the business goals, and not just
keeping the lights on. It is really encouraging to hear that, I
mean, you work for a really big company, and really big companies
are the places where this sort of change happens the slowest. So,
it's really encouraging to hear that it might not be like
overnight. I think that's a really positive thing.
Kevin Overstreet (44:27):
I think so. I mean, I think that one of the things that helped me
be successful is the fact that I was familiar with the processes
and data that I was trying to work on. You were talking earlier
about, had I ever considered in consulting work. I have to get in
and sort of understand the data and the processes and what they're
doing a little bit. I think there's a difference between building
something from just a simple set of requirements that maybe
somebody in IT has put together versus having an IT sort of
technology oriented business person in there to help translate that
into, okay, that's what this really means though.
Kevin Overstreet (45:14):
Because I think we've all been through those things where people go
off and they build a system or they build a dashboard, and they
come back and technically we've met all the requirements, but
that's just not what we had in mind, or whatever it is.
Rob Collie (45:27):
Well, I mean, even if the business were able to, so this is the
problem with the requirements gathering process. This is the
problem with waterfall oriented business intelligence. It's super
fundamental and it needs to go away. If the business were able to,
in a reasonable timeframe, collect and convey their requirements,
in a lossless form that was then transmitted to IT and understood
without noise, without misinterpretation, without loss, and then
implemented to that specification, first of all, that'd be the
first time in human history that, that had ever happened once.
Rob Collie (46:14):
It just has never, ever gone down like that. It always takes
forever. It's always littered with omissions. It's always littered
with misunderstandings and misrepresentations. But even if you got
there, as soon as the business saw exactly what they wanted, they
would go, oh, you know what? We need something different. We didn't
know better. This is why we've come around at P3 to the saying that
human beings do not know what they need until they have seen what
they've asked for.
Rob Collie (46:49):
The old tools were just so slow. It just didn't make sense. As a
business stakeholder, you would never sit in the same room as
someone who was developing like an old school ETL process to power
a data warehouse that someday might have an analytical model built
over it, that someday would put its first dot on a chart. You would
never sit there and watch it. But today's tools move so fast that
you can actually sit with the person who's building it and you can
develop it, and you can just get started.
Rob Collie (47:23):
You'll find yourself, even after like the first couple of days, in
a place that you ... It sounds like an exaggeration. It sounds too
good to be true, but you can find yourself, after a couple of days,
in a place that it would've taken you six or seven months to get
to. By the time six or seven months have elapsed, the original need
is already evolving, people are exhausted. They want to give up.
They don't want to keep on it. It's not even a matter of like a
couple of days versus six or seven months. It's a difference of we
got somewhere versus we didn't at all ever.
Rob Collie (47:57):
That's a huge change. And boy, is the world slow to wake up to this
change. You can deliver the technology all at once that makes this
possible. It's those people again, isn't it? People take a long
time to trust that the rules have changed.
Kevin Overstreet (48:14):
Yeah. I think that a lot of times people are locked into all of
these old tools that they're comfortable with, that they just don't
even know what's even possible, and that's going in and trying to
explain what's possible, doesn't really work, so I think you have
to give them something to react against. Like you said, give them
what they've asked for.
Rob Collie (48:37):
We'd be remiss if we didn't close with the story of how we first
cross paths. There's an old, and it's actually, it's just a truth,
which is that the things that are hardest one are the things that
we end up valuing the most. I think your journey down this path
started with an accidental psychological operation committed by the
universe against you.
Kevin Overstreet (49:02):
It did. Not long after I discovered Power Pivot, it was probably
three or four months, so when I discovered Power Pivot, I
discovered the blog. That was when I saw the training class, so I
was like, going to go to the training class. I went through all the
effort to convince everybody that this training was going to be
hugely beneficial to me. It was in February, and I believe it was
2012, and so I had had to go from Indianapolis to Cleveland at the
time. I left Indianapolis, and it was snowing, and by the time I
was five miles outside of Indianapolis, I was right in the middle
of what we now know as the Ground Hog day blizzard with 40 inches
of snow across 74 in Cincinnati.
Kevin Overstreet (49:55):
I was driving the most appropriate vehicle, which was a 2010 Honda
Civic Si hatchback. You can look that up. I basically was driving
like a snowmobile. That was it. I fought my way through the 40
inches of snow and was now headed north towards Cleveland, when I
started to notice that my lights appeared to be dimming. Then I
realized that I'd had trouble with my alternator, but that had kind
of cleared itself up.
Kevin Overstreet (50:29):
But at this point, I thought I was having alternator problems
again, so the first thing I did was turned off the lights, turned
off the heat. I got on the phone, called my wife and said, if we
haven't joined AAA, do that now. So, I drove about 60 miles in
complete darkness through what was left of the ice and snow to make
it to a quality inn outside of Cleveland. At the time, I was
coaching softball and I had like a training bat.
Kevin Overstreet (50:59):
It was about a half an inch around aluminum ride that was three
inches long. While the car was running, opened the hood, down in
the engine and start tapping the alternator to try to get it to
kick back in. And all of a sudden, I hear the car rev up,
alternator starts charging again. I jump in the car and turn on the
heat. Because it'd been zero the entire way there. Finally made it
to the hotel that night, went through the two days of Power Pivot
training. At the time, I talked to Rob and I said, I just don't
know. I don't know where I'm at.
Kevin Overstreet (51:37):
I asked him one question, and he said, "You're way down the rabbit
hole." That was kind of the start of my journey on Power Pivot and
everything. Yeah, what should have been about a three or four hour
trip ended up taking about nine hours through 40 inches of snow and
complete darkness with no alternator in the basically a little tiny
sled.
Rob Collie (52:04):
Yeah. I mean, I don't even know where the alternator is in my car.
I suppose I used to know. I mean, it's probably the thing that the
wires come out of, right? But I wouldn't think to tap it. This is
some real MacGyver stuff here.
Thomas LaRock (52:18):
Roll up your sleeves and get the job done.
Rob Collie (52:20):
Yeah. See a problem, fix a problem. What's the big deal? It's the
Overstreet way.
Kevin Overstreet (52:26):
It is. I mean, it's probably not the safest way. There was multiple
fan belts and fans and all the things spinning around that probably
could have flung the ... Became a flying death rod.
Rob Collie (52:41):
I'm imagining you timing the fan blade. I got to get the bat in
between the fan blades as it's spinning at 3,600 RPM. You're like,
one Mississippi, two Mississippi.
Kevin Overstreet (52:54):
Yeah. I was reminded of the times when I was a kid and my dad would
be, want you to put your hands somewhere and go, "Here, hold this."
One time we had the axle broke off of a trailer and he wanted me to
get underneath it and hold it up while he chained it up with a
piece of chain that we had gotten by calling a trucker on the CB
and he had stopped and given us a scrap piece of chain that he had.
So, we were going to haul the horse into the track on three wheels.
Well, we ended ... We did haul the horse in on three wheels with
the axle chained up with a scrap piece of chain from a trucker, and
the horse in that trailer would eventually become a world
champion.
Rob Collie (53:38):
I mean, it's really just a story, a story as old as time, isn't it?
It's a classic archetype. Let it be said that data people are not
all cut from a single mold and it's never boring, is it?
Kevin Overstreet (53:53):
No.
Rob Collie (53:54):
And Kevin, thank you so much for being on the show. I know I've
enjoyed it immensely. I hope everyone else does as well. We'll be
sure to catch up soon.
Kevin Overstreet (54:02):
Well, thank you guys for having me. I've enjoyed it a lot.
Rob Collie (54:05):
We might have to have you on again. I just don't think that this is
the last of Kevin Overstreet on Raw Data by P3.
Kevin Overstreet (54:11):
We could talk about all the different times I should have used data
to prevent near death experiences on my farm or something
maybe.
Rob Collie (54:19):
Yeah. The 11 times Kevin almost died if only he'd had data,
Kevin Overstreet (54:25):
That's it.
Rob Collie (54:25):
All right. Well, I'm glad you survived all of those occasions to
make it to this point.
Kevin Overstreet (54:31):
I am too.
Rob Collie (54:32):
All right. Well, thanks again.
Kevin Overstreet (54:33):
All right. Thanks. Cheers.
Announcer (54:35):
Thanks for listening to the Raw Data by P3 Podcast. Find out what
the experts at P3 can do for your business. Go to
powerpivotpro.com. Interested in becoming a guest on the show,
email lukep, L-U-K-E-P, @powerpivotpro.com. Have a data day!