Dec 22, 2020
- From their early days working together at the Elizabeth Glaser Pediatric AIDS Foundation (EGPAF).
- Shannon & Stephanie's philanthropic work with non-profits such as The World Central Kitchen
- There's a holiday fundraising effort for EGPAF, where your donations will be matched through December 24th! Donate Here
Episode Timeline:
- 2:00 - The Origin of the "Witches" name and how they discovered the magic of the Power Platform
- 6:40 - The Vampiric manner that the Witches are converting others to Power BI
- 12:25 - Power BI is vital in data optimization in the Non-profit space
- 17:20 - The Data Community is an amazing community-those that have the Data Itch are a different and helpful type of human
- 30:00 - Gender and the data world, Rob and Stephanie debate the usefulness of Calculus, and the elation of besting one's academic rivals
- 39:50 - Data quality is important, especially so when lives are at stake...it's evident in the analytics
- 47:10 - How the Witches are using Power Apps and Power Automate
- 55:00 - Upstream and Downstream flows to Power BI- The concept of the Action Loop
- 1:01:30 - Stephanie's recent graduate degree and the highs and lows of going back to grad school
- 1:07:30 - The resistance to change and new technology-predictive modeling and machine learning are a tough sell
- 1:11:25 - Shannon's new gig at the not quite so non-profit Microsoft
- 1:16:15 - How COVID has impacted operations, and how non-profits view Microsoft and other large corporations
Episode Transcript:
Rob Collie (00:00:00):
Welcome, everyone. We all know that Power BI is good. We all know
that the Power platform is good. But it's not often, that we get to
talk about it actually directly helping to save lives. But in a
very real sense, our guest this week, the Data Witches, that's what
they've been doing. They've been using the Power platform to help
save lives. The data witches, Stephanie and Shannon met while
working at the Elizabeth Glaser Pediatric AIDS Foundation. So
clearly, they're good witches, but they're data witches. And after
seeing the promise of the Power platform and Power BI specifically,
they've successfully advocated for the adoption of these tools
within that organization, to great impact. Years ago, on separate
occasions, I actually had the pleasure of having each of them in
one of my classes. And so I played a small part in their
introduction to this world. But that's just the beginning of the
story. Because these two have really spread the religion far and
wide, both within the nonprofit sphere and elsewhere. This was a
really great conversation with two great people. And I hope you
enjoy it. So let's get after it.
Announcer (00:01:20):
Ladies and gentlemen, may I have your attention please.
Announcer (00:01:24):
This is the Raw Data by P3 podcast with your host, Rob Collie. 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 (00:01:39):
Welcome to the show, Data witches, Shannon and Stephanie. Alright,
so let's start there. You started a blog. You call yourselves the
Data Witches. Did you have any particular witches from fiction in
mind that you sort of identify with? I'm assuming it's not the evil
witches. I know both of you, you're pretty nice. So was there any
particular inspiration? Is this a reference to the TV show, Charmed
or anything like that?
Stephanie Bruno (00:02:12):
It was not. No. First of all, thanks for having us here, Rob. Data
Witches is new, like you said. The name came because Shannon and I
were doing a presentation together at the... What was it
Shannon?
Shannon Lindsay (00:02:27):
The Power Platform World Tour, DC.
Stephanie Bruno (00:02:31):
That's right. Right before everything shut down. It was in March.
And we were doing a presentation on optimizing your data models in
Power BI and showing some tools and someone in the, one of the
participants said, "What is this witchcraft you're showing us?" So
we decided to just go with that. So, no. No specific witch.
Rob Collie (00:02:51):
Okay. Witchcraft, though. That's good. I like that.
Stephanie Bruno (00:02:53):
It's witchcraft. Yeah.
Rob Collie (00:02:55):
A talk on optimizing your data models.
Stephanie Bruno (00:02:58):
Indeed.
Rob Collie (00:02:59):
I went to School of Computer Science, and all of that. And I still
wouldn't have predicted that my future self might be on stage
giving a talk about optimizing data models. Did either of you sort
of see that in your career future 10 years ago. We're going to be
on stage talking about optimizing data models.
Stephanie Bruno (00:03:15):
No, never, not at all. I don't even like getting on stage. I never
thought I would want to do that ever. Which is one of the reasons I
like to present with Shannon a lot. Presenting with a buddy, I
think is a, makes it a lot more fun. And Shannon's a great
presenter. So I started presenting at user groups. But once I got
to present with Shannon, it just became a lot more fun.
Shannon Lindsay (00:03:36):
I never in a million years, saw myself talking about data models,
or even knowing what they are. I like to be on stage to sing and
dance. But the tech came later.
Rob Collie (00:03:48):
Do you sing or dance during the tech presentations?
Shannon Lindsay (00:03:51):
Occasionally.
Rob Collie (00:03:52):
Really?
Shannon Lindsay (00:03:53):
Interpretive dance works really well, for data model
optimization.
Rob Collie (00:03:57):
This is a many to many relationship. Let me show you what it looks
like. That's awesome. Yeah, being able to blend that sort of
personal touch into talk, to me, even though people just really
don't expect it. I love that sort of thing when you can, when it's
done right. So what are your backgrounds? How did you discover
Power BI and the road from singing and dancing right on stage,
right? The road from that to optimizing data models on stage. It's
an interesting road. I find this kind of stuff to be the absolute
best stuff. If we didn't set out to be techies to begin with. But
we discover something that helps us. That's the best thing. If
you're doing it for tech sake, you're almost certainly not even
doing it well. People might think of this, "Oh, the accidental path
to getting somewhere." No, it's the right path. It's really the
only one that matters. And so, I know bits and pieces of your
stories, but I don't know all of it. And I'm sure people listening
would love to hear it. So let's talk about you.
Shannon Lindsay (00:04:58):
I'll start with this one. I'm Mrs. Shannon. I had a non traditional
path into this, which I think a lot of people in this field did.
Which I find so fascinating. My background is really, as an
analyst, if you put it in data speak. Many moons ago, after
college, I joined the Peace Corps because I wanted to experience
other parts of the world. And I thought it was kind of my duty
being so fortunate to be able to work with people that were less
fortunate and understand what the rest of the world lives like. And
it was actually during that experience that I got to work with a
health facility. I was doing health education, and I got to work at
a health facility and help them implement their first ever
electronic medical records.
Shannon Lindsay (00:05:45):
And then it was from there that I decided to go on and get a degree
in public health. And from there, I specialized in something called
monitoring and evaluation, which is like a little niche part of
development and public health organizations. But essentially, all
it is, it's an analyst. And I was using a lot of Excel, that was
basically everything I was doing. And once I met Stephanie, after
years at different organizations, we landed at the same
organization. And once I met her, she introduced me to the literal
magic, that was Power Query. And it was honestly from there that my
whole journey with Power Query, Power BI really started. So it was
probably about I don't know, a year after Stephanie and I met that
I came to my first Power Pivot Pro training. And then there my
journey just continued.
Rob Collie (00:06:41):
Awesome. Okay, so that means Stephanie was like patient zero in
your life, right? She was the one that was infected first. Let's,
so Stephanie how many fellow... We can use the witch metaphor, we
start using the vampire metaphor, right? Because vampires make
other vampires. Stephanie you just going around in the world
minting other Power BI people or, how's this working?
Stephanie Bruno (00:07:05):
I'm doing my best. Definitely trying to, but I would say that
Shannon is the most infected of anybody that I've had the
opportunity to infect. Yeah, she's really taken it on. And once I
saw that in her, that she liked it. I didn't let up. It's like,
"Oh, Shannon, you like Power Query? Look at DAX, let's go there
next." Yes. So my opportunities to teach more people they are
generally more within our organization, is where I started with
them. But then also the community, I would say, that's the thing
that really was the big changer for me, in my path getting here. I
do have a IT background and a left brain background, I guess. My
degree is in math and physics. And I started off in IT as a
software developer. But I always liked that I could just hide in
the corner and do everything by myself. And I thought that was just
fine. Until I had an opportunity to go to a past summit in
2013.
Stephanie Bruno (00:08:06):
And I do remember thinking, "Okay, this is awesome." I have two
little kids at home. I'm busy all the time. I don't get any sleep.
I'm just going to go to these sessions and learn a bunch of stuff.
And then sleep. Just get a lot of sleep and watch movies and stuff
in my hotel room. You're chuckling because you know that's not what
happens at past Summit, I think. So-
Rob Collie (00:08:23):
I've been a couple times.
Stephanie Bruno (00:08:25):
... Yeah. So that it turned out, no, I was completely wrong. And a
bunch of nice people took me under their wing. And they said, "Oh,
you do this great work for HIV in Africa. And we want to help you
with that. This is amazing." So needless to say, I got very little
sleep. But I met a lot of people and my eyes were open to, "Wow,
this thing called community, I think that this is actually probably
way better than just hiding in a corner and doing my IT work by
myself." So everything just exploded from there. And I realized
that there's a lot to learn and a lot of people who are really nice
and want to help and share with each other. So that changed
everything for me.
Rob Collie (00:09:06):
Yeah. That's really cool. I didn't know that you had that math and
physics background. I didn't know. Or maybe I did, just don't
remember which is also-
Stephanie Bruno (00:09:14):
What? You don't remember everything about everybody who attends
your classes, Rob?
Rob Collie (00:09:17):
No, I don't remember a lot. I wouldn't have guessed from knowing
you that you came from an IT background. That's really just because
of the IT stereotype of, IT isn't necessarily known for their
communication skills, or their warmth. One of the things we're
always talking about on this show, is the notion that the hybrid is
where it's at. Whatever the sort of the day to day, subject matter
or business domain is. You have to be firmly rooted in that subject
out in the trenches, so to speak, while at the same time able to
execute technically. And if you have both of those sort of in one
brain that's infinitely more valuable than having it in two
separate brains.
Rob Collie (00:10:08):
It's kind of unfair to IT in a way, if I go, "Oh, you're just way
too personable to have come from IT." But that's something else I
want to talk about, which is that so many people who find
themselves to be really drawn to this stuff, whatever this stuff is
data, right? And using it, to help using it for an advantage. So
many of us, this group, this community, so many of us did not grow
up as the STEM people. The math and science people. Stephanie it
sounds like you did. You didn't go from a liberal arts and
humanities focus in high school directly to math and physics in
college, I don't think.
Stephanie Bruno (00:10:50):
That true.
Rob Collie (00:10:50):
You probably are already on that path a little bit. Did you go to
math competitions in high school?
Stephanie Bruno (00:10:55):
I did not do that.
Rob Collie (00:10:55):
Oh, see?
Stephanie Bruno (00:10:58):
No.
Rob Collie (00:10:59):
Yeah.
Stephanie Bruno (00:10:59):
Yeah. Not full on. I'm sorry to disappoint you here.
Rob Collie (00:11:04):
They were debating giving us math nerds a letter, like a varsity
letter to put on our jackets for math team. And I was like,
"Oh."
Stephanie Bruno (00:11:12):
So you did? You did go to math competitions?
Rob Collie (00:11:15):
Oh, yeah. I was a [mathlete 00:11:20] and physics competitions, and
all that kind of stuff. But I had the sense even then to know that
putting a letter on my jacket for math team was just going to make
me a magnet for ridicule and bullying. So I just said, "No, we
don't need to do that." But Shannon, did you grow up as a Math
Science nerd?
Shannon Lindsay (00:11:38):
No, not at all. My undergraduate degrees in biology, but I think I
picked it because it was the easiest science. I was never good at
Math, I would still say that I'm not great at Math. But through
Stephanie's coaching, and the work that I've done, I realized that
I am now capable of doing Math at least. But no, I was not a huge
academic. I really started getting serious about caring about the
data once I saw it, and I was working in the field. And I think
that your reference to having both brains is really interesting.
Because when I first started working with Stephanie, I had the
field experience, and she had all of the technical knowledge. And I
think over the last five or so years, we've both learned so much
from each other. In both arenas, it's been really cool.
Rob Collie (00:12:27):
You met working for a nonprofit, in the nonprofit space. And P3,
we've done a reasonable amount of work for nonprofits. My belief is
that really the kinds of problems that with data that you tend to
be solving aren't really structurally much different at all,
working in the nonprofit space versus the for profit space. The
biggest difference would be if we were going to advertise our
services to the nonprofit sector, we wouldn't say things like,
bring the impact to the bottom line. We probably just wouldn't talk
a certain way. That the work would probably be similar. For
example, we did some work for FEMA. FEMA is a government agency,
not a nonprofit, but it basically is. And one of the metrics that
they had that I thought was really cool, was essentially, how much
does it cost us to spend $1.
Rob Collie (00:13:23):
We've got all this money that we need to deploy in response to a
disaster. A hurricane wipes out Florida for the 10th time in the
decade and gets declared a disaster zone. And so FEMA deploys and
tries to put a bunch of money in the field in various forms,
insurance and things like that. But there's also an overhead cost
of doing that. So it's an efficiency metric. What are the sorts of
metrics and things that you've find yourselves optimizing for? I
know that you can't speak for the entire nonprofit space. What's
the name of the organization again, where you met?
Stephanie Bruno (00:13:58):
It's the Elizabeth Glaser Pediatric AIDS Foundation.
Rob Collie (00:14:02):
See, I was pretty sure that's what I was going to say. I didn't
want to mess up a word here. There's a lot of words in that-
Stephanie Bruno (00:14:09):
It's very long, if you want to say it again, you can say EGPAF.
Rob Collie (00:14:14):
EGPAF. All right. I figured that there's got to be a cool
abbreviation for it. What are the sorts of things that you find
yourselves optimizing for at that organization? It's not
profitability, right?
Stephanie Bruno (00:14:25):
Right. Well, I actually jumped over from our IT department a few
years ago to the department, Shannon was mentioning, monitoring and
evaluation. Because I wanted to work closer to the data about the
people, and less the operational data. So since I did that, then I
am focused on that. We call it program data. So that's the data
about all the facilities in Africa that we support. How many people
are retesting for HIV, and of those people, how many people are we
getting on treatment? And how many people stay on treatment? So
it's really about it impact, it's less about profitability. So we
try to show not how much profit we make, but how many people we're
serving, basically.
Stephanie Bruno (00:15:09):
But in addition to that, you mentioned that how much does it cost
to spend $1? For us a really hard number to get accurate is, how
much does it cost to test a person? And you can imagine that's
because data silos, right? So because we've got our finance system,
we've got a whole bunch of systems, and then we've got the program
data system, and they just aren't at the same level. And so it's
just years and years and years. And it's kind of the holy grail
that we're trying to get to is blinking all this up. So we can just
answer a question easily like that.
Rob Collie (00:15:43):
And that really is one of Power BI's top two strengths, as a data
tool, relative to its other BI tool competition. Is its ability to
splice across those silos, and start to allocate fixed costs that
you're measuring in one place. You still have an HR system, right?
You're still cutting paychecks. So that cost needs to be allocated.
Somehow, across the per test. It's just so many inputs. If you're
just saying, "Hey, look, the cost of the test kit... What
percentage of the cost of testing is just the test kit." It's
probably really tiny, right? There's the facility, all that kind of
stuff, right? Silos are everything, aren't they?
Stephanie Bruno (00:16:33):
Yeah. They're painful.
Shannon Lindsay (00:16:34):
And having worked in nonprofits, and also with governments for the
last 10 or so years, I think that it's pretty obvious that their
progress in new technology is a little bit behind other industries.
So what was really neat about implementing Power BI and working
with Stephanie was that we were one of the first nonprofits to
start to use this technology. We really got our leadership on
board, it really grew very fast within the organization. And then
we were able to preach the gospel, or I don't know how to say it.
But encourage other nonprofits to use this technology as well. So
there is a whole little nonprofit community that is interested in
using this technology too. And I think that Stephanie had a big
hand in that.
Rob Collie (00:17:22):
Is there a water cooler where all the nonprofit people get together
and hang out? And how do you connect like that?
Shannon Lindsay (00:17:27):
You know there isn't, really. Part of what's interesting about the
nonprofit industry, the part that I've been involved in, is that
we're all trying to accomplish the same thing. But at the same
time, we're all fighting for the same tiny pot of money. So there's
a lot less collaboration than you would imagine. And I didn't
really understand the benefit and the value of community until I
became a part of this data community and the sequel community and
the Power BI community. I didn't realize that people just helped
each other and just shared information, I did not know that was a
thing. So now I understand. And, I think that's another part of why
we do Data Witches and why we run user groups. We want to make sure
that other people know these tools are out there, and that they
will save you hundreds of hours of time and simplify your life. And
just from the analyst perspective, it completely changed my day to
day work.
Rob Collie (00:18:26):
You've probably both heard me say this, at least one time,
somewhere in some venue, the desire, the compelling, overwhelming
need to tell other people about a piece of software. That is very
strange. That doesn't happen. Software just sucks, it just wall to
wall. So much software, it's just so much of a source of pain. And
it's like, a lot of software's only slightly better than not having
it. I'm just really cynical about most software, even though that's
my industry of origin. It's probably because it's my industry of
origin. But this stuff is different. There is almost a compelling,
almost moral duty to tell other people about it and to get them
introduced to it. It's weird, isn't it? Even just in the way that
the two of you describe what you do? I can hear it. I can hear
that. This is a moral imperative. We have to share. We can't keep
this to ourselves.
Shannon Lindsay (00:19:22):
If I can save one other person from copying and pasting... I mean,
the reality is we were supporting 5000 health facilities when I
first started. And what we were doing was manually compiling data
into Excel sheets. It just sounds like madness, even saying it out
loud. Now that I know what's possible. I also used to joke that I
used to enter a lot of zeros into spreadsheets. That was my job.
Data entry, copying, pasting, entering zeros and none of that is
necessary, but I did not know that at the time. And so if I can
save someone else that pain, by all means.
Rob Collie (00:19:58):
Yeah. I think that in the course of any career, any sort of office
career, and I know that you're not always in the office, right?
Sometimes you're out in the field. But in any sort of knowledge
worker role, sooner or later, everyone is going to cross paths with
Excel. And the majority of people, when they cross paths with
Excel, they just bounce right off of it. It's like, "Ooh, gross,
I'm out." And that's at least 14 out of 15 people. But there's like
one in 15, one in 16, that you get this itch like, "Oh, I just
typed all those formula, that those numbers in, I was able to get
this plus this, plus this. And I was able to change the input
without rewriting the formula on the calculator plus this, plus
that. And I get the answer out again." This twitch starts to happen
behind your eyes and I go, "I was kind of cool."
Rob Collie (00:20:49):
And next thing you know, it's like, wall to wall VLOOKUP in your
life. So it's weird. I think that the reason why certain people,
and again, you don't have to have been someone who liked math in
high school. And I love this. Because frankly, most of the math we
learned in high school isn't useful. It's just most of it isn't all
that useful. The thing that draws you to it is the desire to not
repeat work. That's the thing that feels good about Excel the first
time to say, "Oh, I could really get into this." But then, of
course, Excel turns around and betrays you, five minutes later. And
it's just nothing but manual work.
Rob Collie (00:21:28):
Even though you're writing formulas and whatever, it's still awful.
And so it's really ironic that the crowd that is least able to
tolerate tedium, is the crowd that's drawn here. And then Excel
turns around, and it just tortures you. And I do think that has a
lot to do with the feeling of moral imperative to help others
because you know what it feels like. It's not just saving time,
it's people are literally suffering.
Shannon Lindsay (00:21:58):
Yeah, if we're spending all those hours compiling data, when we
could be analyzing is this having an impact? Is this saving lives?
It's a no brainer.
Rob Collie (00:22:08):
Yeah, when you start to think of the end result in particular,
right? When you're working in the for profit space, you don't
typically think that way. It's a little harder to get sort of
emotionally ramped up about some huge corporation saving more
money. But still, the people that you're working with at that
company, is more than enough to keep you charged up. It's their
success. It's their life getting better.
Stephanie Bruno (00:22:32):
So one of the things that Shannon and I did together at EGPAF with
our colleague, [Elisa Lei Laurie 00:22:39], because we wanted to
share more. Actually, I should back up and mentioned, when we say
we work at this nonprofit, people get the impression that, oh, it's
small, just a few do gooders trying to do their best. But we're
actually pretty big. I think we have a staff of close to 4000
throughout the world, honestly. Most of whom are in Africa. Our HQ
staff in DC is maybe around 150. But so mostly everybody that we
work with, is in Africa. Getting all of these people that we work
with to start using a new tool, Power BI instead of Excel, it was a
lot harder than I expected. But like you said, we knew that we
would save people so much time, if we did.
Stephanie Bruno (00:23:21):
I remember one time Shannon and I were on a trip together in
Tanzania. And I was in the office trying to see how their data
systems work together. And I looked up, and I just looked at this
sea of people on computers out in a room. And my jaw dropped
because I realized, "Oh my gosh, they're all just doing copy and
paste." That's what I'm seeing here. It's like 50 people doing copy
and paste. We have to do better with this. So we really wanted to
not be the only few people building Power BI data models and
reports for people. We knew that if this is going to work, we have
to get a lot of people doing it. So we started an internal user
group at EGPAF. We started it three years ago. And we have
quarterly meetings, we use Teams and people are sharing resources.
And it's people from our different country offices helping each
other, which is amazing.
Stephanie Bruno (00:24:14):
We've had some celebrity guests, I think, come on for us. We had
Patrick LeBlanc come on, probably [Comassani 00:24:21] came on and
presented. And so it's just been this great thing. It's really been
successful. You know, we can check out the metrics in Power BI to
see who's building reports and who's getting the most usage. And it
turns out, it's actually not our DC office anymore. It's our Malawi
office has the most reports being used and built. So this is kind
of been the best thing for us. Is this sharing with other people
and getting other people to do the work and build reports and get
excited about it. Yeah, so that's the sharing that's really been
the most impactful. I think has been this internal group and just
knowing all of our colleagues in 12 countries in Africa are now
Power BI experts.
Rob Collie (00:25:03):
That's great. Do you see the same sort of like ratio of author and
consumer that I'm talking about. That 1 out of 16? Not everyone is
going to, of those 4000 employees, for instance. Let's say if all
of them were using it, most of them would be using it, consuming
reports and dashboards that have been built for them. Maybe
interacting with them filtering, drilling down, whatever, but I
wouldn't expect that everyone is data modeling. It's not really
even a question about the nonprofit space. It's just more like,
it's your experience with the other offices. Because my experience
with all this is overwhelmingly, USA based. I would say that this
discovery of 1 in 16. And there's actually a couple of different
ways that we've come to that number that both agree actually that's
about 1 in 16.
Rob Collie (00:25:51):
I just wonder if in a different country, it's the same. Are you
minting authors at a greater or-
Stephanie Bruno (00:26:00):
No.
Rob Collie (00:26:00):
... faster pace?
Stephanie Bruno (00:26:01):
I think it's the same. We've done a few trainings in Africa.
Shannon did a couple. I did a couple. We partnered with Microsoft,
actually, some of the people from the CAT team came and did
trainings for our office. Casper, I know you guys are-
Rob Collie (00:26:15):
Oh, yeah.
Stephanie Bruno (00:26:16):
... good friends. Casper and Patrick, and then Maggie Sparkman from
the documentation team. They all came and they did these trainings.
And it was fantastic. And I would say I think probably the
countries, maybe had more people sign up for the trainings and
people who were actually ready to do it.
Rob Collie (00:26:33):
It's exciting.
Stephanie Bruno (00:26:34):
Yeah, it's exciting. But I do think that's just how it goes. And
then people come to these trainings. And not everybody that comes
to it is going to become a data modeler. But you're going to find
some people that maybe you didn't expect, that really grab on and
just love it.
Rob Collie (00:26:49):
Yeah, I think that suffering that we're talking about, is sort of a
crucial ingredient in deciding whether or not it's going to take
with you. The same person, the same brain is a lot less interested
or even understand the value of adopting these tools until they've
worked in the spreadsheets sweatshop, for a couple months minimum.
Even people who came to those classes that you're talking about,
some of them it didn't take with them. But some of them, those
people that didn't take with, will now go off and not use them. For
some reason, their job takes a turn, and now they're using Excel
all the time.
Rob Collie (00:27:33):
And they're getting ground down. And then the next time you run
that class, someone that you remember being checked out will be
like, "Oh, my God." You'll see the spark, you'll see the light bulb
go on. I've actually had that happen.
Stephanie Bruno (00:27:48):
You've had someone come more than once?
Rob Collie (00:27:51):
I did a training one time for a group, it was a company. Here,
there's sort of two parts of story. One part is, I did go back one
year later and train the same people on the same things, which is
really silly, right? And this is where I learned a very important
principle, that when you first learn this stuff, you have to put it
into use almost immediately. Not almost immediately, you better put
it into use immediately. And by the time, either of you showed up
in any of my classes, I'm sure I was already saying this. You spend
a couple days learning this stuff with us. You can't go back to
work and do it the old way. Until you get caught up. That's a very
seductive trap that you can fall into. And that's what this crew
did.
Rob Collie (00:28:36):
They spent like three or four days with me. And then of course,
their job piled up during those three or four days. They weren't
doing their job, they were already way behind. And so they got
further behind. And so the next week of work, they just threw
themselves back into the old way. Just try to dig out. Mistake, it
was over. Just get becomes less and less fresh. And there's never a
good time. And so a year later now they were really in trouble. And
they brought me back. And that's how I learned the lesson to tell
everybody like no, on Monday, you're not allowed to do it the old
way. But there have been someone assigned to this class. We call
them hostages. We teach a class for a company. A manager will tell
everyone that they have to be there. Some people are excited about
it, most people aren't.
Rob Collie (00:29:22):
There was a woman in the room who could not have been more
disinterested. I just had to tell. A year later the thing is she
was new in that job. She'd never had to do the the Excel analyst
job like she did the equivalent of sitting in the back filing her
nails. It was so hard to take as the teacher. But the next time I
came back she was on the front row asking the absolute best
questions. She parlayed that into a, she's a full time BI developer
professional now. And if I had based my impression on her on what
she showed me on that first time, I would have thought she was
zero.
Rob Collie (00:30:04):
It's so cool. I love that. So yeah. Sooner or later, you'll have a
similar experience, I think. You'll find someone that you wouldn't
have bet on but suddenly they light up.
Shannon Lindsay (00:30:14):
I think that's been part of the real, I don't know, joy of being a
part of building something from the ground up at EGPAF, we saw the
most unexpected people become the champions of the new technology.
And I can't say this for the whole world. But in the countries that
we worked in, I'm saying worked past tense, I do not work there
anymore. But in the countries that we worked in, a lot of the
monitoring and evaluation teams and the finance teams were
predominantly men. And a lot of our shining stars and people who
became the power users or the people who spoke up and presented at
our user group, were women, who were often overlooked. And they
took this technology, and they ran with it. And now they run those
departments. I mean, so not only is it making their day to day life
easier, but it's really giving them some visibility in the
organization. It's changed a lot of people's careers, myself
included.
Rob Collie (00:31:12):
Yeah, I'm really glad you brought that up. Because this is one of
my favorite things. And when I bring it up, I run the risk of
somehow looking like the bad guy, even though everything in my
heart is 100% good about this, I know what you're talking about. I
went to computer science school, I grew up in the Math Science
World. And I know what the finance world looks like. Oh, my gosh,
these things couldn't be more male dominated in terms of numbers of
people. I had one or two women in my computer science classes with
me. Stephanie was there. I see the same thing. We even talked about
this with Wayne Winston, when he was on the show. We're sort of
coming around to this idea that the analytical half of our brain.
This is a broad oversimplification, but the analytical curious
parts of our brain is probably the more female part of our
brain.
Rob Collie (00:32:06):
The stuff that, I don't know, calculus. There's something that's
super duper abstract about calculus that I often wonder. As an
older person now, I often wonder, why is it we're sitting around
trying to encourage more girls to study calculus? Because I think
calculus sucks. Girls might have it right staying away from
calculus. I mean, it's like, why is calculus considered such a an
objectively good thing that everyone should be doing? We're putting
it on a pedestal.
Stephanie Bruno (00:32:39):
Because it's awesome.
Rob Collie (00:32:40):
Oh, really? You still think calculus is awesome?
Stephanie Bruno (00:32:43):
I do. I do think calculus is awesome.
Rob Collie (00:32:47):
All right, give us your bullet point defense of why calculus is
awesome. Keep it in mind, I went to the Final Four in Florida state
competition for calculus. I mean, I was on a team that went to the
Final Four. I was the weak link on that team. We would have placed
first without me. So I've definitely been up to my eyeballs in
calculus before. And I'm still on the calculus side.
Stephanie Bruno (00:33:10):
So I mean, why do you think it's, that useful or unfun?
Rob Collie (00:33:17):
Well, it definitely gets to the point where it's unfun. By the time
I got to differential equations in college, I was like, "Oh, cool."
I want out of this.
Stephanie Bruno (00:33:29):
Rob, that was my favorite class in college.
Rob Collie (00:33:31):
Was it? Ah, God.
Stephanie Bruno (00:33:33):
But you know what I do remember, there was a guy in my math
classes, and he was just so smug all the time about how he was the
best one in our math group. And there was something about that year
that I took differential equations, I think I broke up with some
boyfriend and I was down. And I was like, "You know what, this
year, I'm going to beat that guy and differential equations." And I
did, and it was one of the most satisfying feelings. And it was
funny, it actually became known in our class. And I even had people
rooting for me, "Come on Steph, you could beat him."
Stephanie Bruno (00:34:12):
So I've really warm feelings compared to [crosstalk 00:34:15] not
based on anything altruistic in any way.
Rob Collie (00:34:19):
Just vanquishing-
Stephanie Bruno (00:34:21):
Just vanquishing.
Rob Collie (00:34:22):
... of someone who had it coming.
Stephanie Bruno (00:34:23):
That smug guy.
Rob Collie (00:34:25):
Good come upings.
Stephanie Bruno (00:34:26):
That's right. So as a result, I got to an A in differential
equations. It turns out, I actually learned something with my
competitive motivation there. So maybe I just have positive
feelings about calculus or differential equations.
Rob Collie (00:34:41):
So I have something similar with me, and that the scouting report
on me going into high school amongst my family was, "Yeah, bright
kid, but no ambition. Not interested in anything. Never applies
himself." I was still getting decent grades. I wasn't the
troublemaker. I was a really good boy. I just was not into things.
I got to a C in history one year in middle school. It was just
like, then I ran into a guy that was sort of, my twin, my mirror
image essentially. And a lot of ways, we grew into the same things.
It really mess with my sense of identity.
Rob Collie (00:35:18):
And in ninth grade, it became, I think it was both of our missions
for a while. Just nothing short of total character assassination of
the kid, was going to be acceptable. And we were friends, and we
were trying to end each other academically. That powered me.
Suddenly, I was really driven, really ambitious. And it was just
like, destroy. I carried that for four years. And for me, it's
almost a spirit of atoning. I threw myself for so long into
something so relatively arbitrary and meaningless in terms of human
life, and made it a defining trait. I think I probably took it a
lot farther in terms of, my own self identity, than what you were
in that diff EQ class, right? That was a maybe like a one semester
one class thing for you. That was a lifestyle for me. I think I
need to live three lifetimes to sort of atone for the bad things I
did to people close to me.
Stephanie Bruno (00:36:16):
I think its time to forgive yourself, Rob.
Rob Collie (00:36:20):
Maybe.
Stephanie Bruno (00:36:23):
Yeah. I think it's been long enough, you can probably let it go.
And who knows, think of all the good things that came from that.
That's another way to look at it.
Rob Collie (00:36:31):
Yeah. Got me to Microsoft, where I proceeded to not use a single
thing that I ever learned in college, or in high school, none of it
was relevant.
Stephanie Bruno (00:36:42):
I get it. I wanted, I was thinking about what you were just saying
about this competition feeling. And the same thing with that smug
guy in my class. But it did remind me that I think part of the
reason that I did go into math and physics and all of that stuff
was it was kind of just to prove something. To prove that I could
do it. I even had a physics professor in college suggest, "I don't
know, if you want to go to grad school in physics, you you might
want to just find a nice guy and get married. That might be
easier." Like really? You're my physics professor.
Rob Collie (00:37:19):
You said that out loud.
Stephanie Bruno (00:37:19):
Yeah. He said that out loud to me. I mean, I wasn't exactly a
stellar physics student. Not like diff eq. But that's the kind of
thing that pushed me I think, huge into going in this direction. It
was just feeling like underestimated. And wanting to prove that I
could do it. Which isn't, again, great motivation. But it got me
here. So I'm happy to be where I am.
Rob Collie (00:37:43):
When was the last time you use the differential equation?
Stephanie Bruno (00:37:45):
Okay, I haven't in a long time. But calculus, I did, because I went
back to school, Rob.
Rob Collie (00:37:52):
Oh, so you used calculus again?
Stephanie Bruno (00:37:55):
I did.
Rob Collie (00:37:56):
At school?
Stephanie Bruno (00:37:58):
At school.
Rob Collie (00:37:59):
You used it at school? Yes. You take calculus so that you can take
more calculus? Okay.
Stephanie Bruno (00:38:06):
Okay. So you are mean, still, I guess.
Rob Collie (00:38:07):
I am a little bit mean. I just remember the kids that were sitting
there, while I was the person you're describing in your Diff Eq
class, that was me in high school. Except no one ever vanquished
me. I mean, I did lose in certain classes, but then I had beat that
same person in another class. And that was all I needed to tell
myself, "I'm still better." But like those kids that were sitting
next to me going, "When are we ever going to use this?" And I would
make just viciously make fun of them. It says, "You're just saying
that because you can't do it." They were right. I'm never going to
use this stuff. I mean, some people do, if you're building bridges,
you're gonna be doing some calculus. Even there, I sneak in and go,
now you're probably going to be using modeling software that will
just take care of the calculus for you. But, I don't know.
Rob Collie (00:39:02):
It'd be like if they kept teaching slide rulers in a way. Tech
moves on, calculus is a little bit commoditize now, unless you are
a physics researcher. There are will huntings running around doing
that kind of stuff on a chalkboard. Today, it's just that most of
us aren't doing that.
Stephanie Bruno (00:39:19):
Well and machine learning.
Rob Collie (00:39:20):
Yeah. So statistics, linear algebra and all this, I had no idea
which of those things were actually going to be relevant. There
were some things that I was taking back then that I kind of wish
I'd either paid more attention or, more importantly retained
longer. Because I couldn't look someone in the eye today and say I
was a data scientist. That's not a title that would hold up to
scrutiny. And if I were to go back to school, it would be to do
stuff like that. If either of you considered that route, have you
pursued it?
Shannon Lindsay (00:39:52):
Well, speaking of statistics, jumping back, just a little bit,
EGPAF as an organization, is a research organization as well. And
so they have statisticians and bio statisticians on staff,
something that I think was really neat for me to see and be a part
of was while Stephanie was doing her graduate work. We got to take
some of the manual, redo the stats, every single time new data
comes in, we got to take some of those processes, and put them into
Power BI. And it was really cool to see. So Steph, do you want to
talk about any of the stats stuff that you did in Power BI?
Stephanie Bruno (00:40:29):
The one I was thinking of that you were talking about with grad
school was when we tried to work on a machine learning project
together for one of my classes. I dragged Shannon in because, A,
she's brilliant, and B, she's really the expert on the data that I
was working on. The project was, I was trying to develop an
algorithm to help us predict which patients were going to be likely
to be lost to follow up. What that means is, HIV positive people
need to take medication regularly, it's critical. If they don't,
then they can infect other people, they get really sick, it's
really important. They can also, I think start to be resistant to
their drugs if they don't take them regularly. So making sure
people come in every month and get their drugs every month is just
critical. So what we needed to do was try to find... We were
looking at a patient level database, which is actually hard to find
in the work that we do. But we're looking at that and we were
trying to find predictors of people who maybe we're going to be
likely than not come back.
Stephanie Bruno (00:41:33):
And so we used our for it, we use power query for it, we found that
most of the data was just in bad shape. So the biggest part of that
project, which was for a machine learning class, didn't turn out to
be the algorithm, it turned out to be the data cleaning. And I
would actually say that was my biggest complaint with grad school
and analytics was that, all of the datasets that they had us work
on were clean. They were just perfect. And so I remember feeling
like I was the oldest person there, first of all. And feeling like
the young people there were just not really getting to see what
real world data was like. And that they weren't getting to see that
what, okay, when they go out and they get their job. They're gonna
have to spend 90% of their time dealing with this garbage data.
That's ultimately what we had to do. So it was it was really cool.
It was exciting.
Stephanie Bruno (00:42:26):
But at the end of the day, it didn't turn out to actually be a very
beneficial project. Because the data was so bad, there wasn't
really anything that we can rely on. So as much as I wanted to do
all the stats, and machine learning and exciting stuff, 90% of that
project was janitorial.
Shannon Lindsay (00:42:43):
Well. And speaking of janitorial, I'm going back now to Power BI.
But I think one of the bigger impacts that we've seen with Power BI
and correct me if I'm wrong here Steph, just simply getting more
eyes on the data made a world of difference for the data quality.
In the donor environment, they require more data than you could
ever imagine at such granular levels. And the data is required in a
really rapid turnaround period. And it's oftentimes collected in
less than ideal systems. Most of the time, it starts on paper. So
that's where we're coming from. But just getting people to look at
the data and put their eyes on it and then say, oh, no, that
couldn't possibly be true. There's no way we only saw men at this
clinic or, whatever. But I think that the trajectory of Power BI at
the organization was really amplified by the fact that people saw
the problems with the data.
Rob Collie (00:43:40):
This is a universal, almost principle, the saying the buck stops
here, it stops at analytics. That's when all of the sins of the
organization, all of the misexecution, all of the sloppiness, all
of that stuff just comes home to roost, on your analytics. In the
for profit world, and I'm sure this is really probably the same
thing in the nonprofit world. But in the for profit world, if
there's some sort of failure or organizational failure, or
shortcoming that interrupts the flow of money that will get noticed
and taken care of. For sure. Now, it can interrupt a small
percentage of the money, and no one will notice. That's weird.
There's a lot of things like that going on. But if 100% of the
money in a particular segment stops flowing somewhere, someone's
going to complain because that's someone's income. Someone is going
to speak up in their own self interest. They're going to notice
when their money doesn't come in. Every other problem that the
organization experiences, the vast majority of them never get
addressed.
Rob Collie (00:44:50):
And then you get to analytics. And that's when all of that becomes
apparent. And it's oftentimes difficult to get people to care about
the data quality, even. In some cases, it's like, people just are
just so flabbergasted like, "Oh my God, I can't believe that that's
wrong." And they just want to go make it right. And you're lucky in
those cases. But in other cases, they're just like, "Nah, we really
can't be bothered to do a better job of maintaining the product
catalog are. Make sure things are categorized properly. I'm not
paid to do that." Yes, you are. But that's not what they're
actually thinking. It's hard to push data quality back upstream to
where it came from. But you absolutely notice it, it does. It just
jumps off the page, especially with Power BI, for better and
worse.
Shannon Lindsay (00:45:37):
I think where we were really lucky. And where a lot of people that
work with data about people is, those numbers represent human
lives. And if we miss a person that we know, is positive, and they
need to be on treatment, but we miss them because of a
transcription error, or because of the piece of paper didn't make
it from the facility all the way to the head office. I mean, that's
almost unforgivable. I know, it's not because we're human, and we
all make mistakes, but it raises the stakes for sure.
Rob Collie (00:46:05):
Well, that does it. We're gonna completely refocus all of our
client work on the nonprofit space. Especially where people's lives
are at risk, because they'll be the only time we can get proper
traction on data quality.
Stephanie Bruno (00:46:18):
Maybe you'd like to come help us with those silo problems we have,
Rob?
Rob Collie (00:46:22):
Yeah. I don't think you want me doing it. Turns out that one of the
things I like to say, because it's the truth, and it's awesome is
that, everyone that works at our company is better at this stuff
than I ever was. I'm quite a bit off peak, even with my own
capabilities. I like to think that I did my part sort of getting
some of this going, not just the company, but the awareness of
these tools. And as I started to discover all these people coming
at me that were saying, "Hey, I got into this because of your book
or whatever." And then I start looking at what they're doing. I'm
like, "Well, you're so much better than I am." That was actually
really gratifying, whereas high school me would have been
mortified. I don't want to let this linger. Earlier, you said maybe
you are still a little bit mean, come on. I'm so nice. I just like
having fun, though. So what about the broader power platform? You
getting into that at all, not just Power BI?
Shannon Lindsay (00:47:16):
I can speak to this. A couple years ago, maybe three years ago,
when Power Apps was brand new, I was working on a global project at
EGPAF. And I had just been hired on to Stephanie's team. So I came
from this analyst role I had been brought into the informatics
team, I was responsible for collecting data in nine different
countries, on when every single patient comes into the clinic, they
have to be screened for tuberculosis. Particularly kids are left
behind in TB screening and so we needed to make sure that every
single kid was screened. So when I came onto the project, I was
told, okay, we're going to put these little wooden boxes in every
single facility, and every kid's going to have a form. And we're
going to drop those forms into the boxes. And then once a week,
somebody is going to come pick them up, and then someone's going to
enter them in aggregate into Excel. So they're going to count up
how many girls, how many boys, whatever. Put it in Excel, and then
we're going to do this, whole Excel based reporting.
Shannon Lindsay (00:48:16):
And I knew enough by then that Power Apps existed. I knew enough by
them that that was what we were going to use. And I think the main
thing I learned from working with Stephanie, and just from learning
about data in general, is that Excel is not a data collection tool.
It should never be used that way. It is trash. Even when you're
doing it for good things, obviously. But so I like to say that I
took away their toys, and I made them use a Power App. And rather
than letting them enter the data, in aggregate, I required that the
data was entered at the patient level for just to make sure that we
had the right data. And as far as I know, that Power App is still
being used globally. And it was a huge success. So it was a Power
App that was feeding into a, don't cringe when I say this, but it
was feeding into a SharePoint list. And then that SharePoint list
was populated in Power BI. I know it's like, undergone some
iterations since then. But I really, by doing that project I got on
board with the Power platform.
Shannon Lindsay (00:49:15):
I can say that at the time, Power Apps was really incredibly
difficult for me to learn. I don't have any sort of a background as
a developer, I had no clue what I was doing. It took me six months
to build the simplest app, but I did it and it's being used and
it's in the rearview now. I believe that there is a lot of room for
Power Apps and Power Automate. But I haven't had any experience
with it since then. Steph, are you guys doing any new projects with
Power Apps and Power Automate?
Stephanie Bruno (00:49:49):
Yes, we are. And I can confirm that your Power App is still in use
indeed. But I struggle with it. I love the rest of the Power
platform but, A, I guess I struggle with how Power BI fits into it?
I guess from a community perspective, it's interesting because it's
still very much the Power BI community and the Power Apps and Flow
communities. They don't have a lot of overlap, which I think is
interesting. I don't think like, there's a lot of people that are
really in both. We went to one of those hackathons at the, let's
not call it the Data Insight Summit anymore-
Shannon Lindsay (00:50:25):
Business Applications Summit.
Stephanie Bruno (00:50:27):
That's right. And we were so excited. We're like, "Oh, good, we're
going to work with all the Power BI, Power Apps and Flow people.
And we were like, the only Power BI people there, and everybody
else was Power Apps. I think there's some work to do to kind of
bridge those communities together better, first of all. And then
second of all, I want to love Power Apps, but I struggle with the
licensing with it a lot. I think how things went with Power BI how
they made it free for a long time. I think a lot of people still do
complain about Power BI pricing, especially with premium. But now
that there's per user premium, blah, blah, blah. But because we're
nonprofit, we get a really good discount. So we actually jumped on
board with premium immediately. And everything just was seamless
and it went along really well. But the Power Apps licensing, I
still can't wrap my head around. So I did push Power Apps pretty
hard at first. But now I'm pulling back a little bit because I'm
worried about how much it's going to cost us.
Rob Collie (00:51:24):
Do you not get the same nonprofit awesome discount?
Stephanie Bruno (00:51:28):
Well we do, but I haven't seen an equivalent of the Power BI
premium. Because, for instance, we have a lot of data collectors.
So if we want to build a Power App and have 1000 data collectors
using it, no way.
Rob Collie (00:51:43):
Yeah.
Stephanie Bruno (00:51:43):
We have to do something way cheaper than that.
Rob Collie (00:51:46):
Yeah, that makes sense for a 1000 person organization. Without an
all you can eat premium version for something, that could be really
difficult. Hey, Microsoft, you're, we'll making it like a dramatic
joke, why are you killing people? We need to change this licensing
model pronto. We're talking about human lives here. It's not just
profitability anymore.
Stephanie Bruno (00:52:13):
Yeah, I know, I feel like a traitor saying this about it, because I
do want to love it. And I want us to use it. But I'm finding that
bit difficult.
Rob Collie (00:52:22):
Well, I often say that Microsoft makes like the best software,
they're really good at it. But they're mediocre at best at every
other thing. Pricing, marketing, documentation, everything.
Everything is... But they make the best software, they really do.
So the Power Apps experience. It's build as no code or low code.
But we get into it, and it's like, oh, this is code. This is
development. I can imagine Power Automate, having a lot of overlap,
potential overlap with the Power BI crowd. Because hey, part of it
is M part of it is Power Query. It's, "Hey, great. That's half of
our story." Power Apps is a different animal. And so it doesn't
surprise me, Shannon. I was going to ask you, before you volunteer.
I was going to ask you, okay, so you're talking about it. And from
the we point of view. We had a Power App, and we did this and I was
like, "Okay, so who... " I was going to ask you who built this
thing? So we get to this next question, right? And you said it was
you and it took you six months to get far, it gets hard.
Rob Collie (00:53:22):
It's not the same as Power BI. It's not just a third technology.
It's like a separate universe next door. So you said you took it
about six months to learn enough to build this app. That's still in
progress, right? So that's not, we wouldn't call that rapid app
development, right? Six months sounds like a normal app development
timeline. But if you had to go back and build a very similar app
today, how much faster do you think it would be?
Shannon Lindsay (00:53:49):
Significantly. I think the main reason for that is the amount of
time that has passed and the community that has formed around Power
Apps, the community is pretty robust now. I think at the time, I
did have help from the community. So I did not build that thing all
by myself. So it took six months with help. But I think that they
have made a lot more of the connectors native now. And there's a
lot more stuff that Power Apps can do that it could not do before.
I remember I had the hardest time with the right back and getting
things to be properly stored in the collections and then right back
to the SharePoint list. And it was probably just because I was
doing it wrong. But there wasn't enough information out there
about, step by step tutorials on how to do things.
Shannon Lindsay (00:54:38):
At the time, there was the Power App in a day training. I don't
want to call it that. The day long thing that existed and it was
cool. You built this whole app where you could pick things you
wanted to buy in a day and it was really awesome. But then, just
like working with clean data versus dirty data. You go into the
wild and you try to put it in your own environment. And it was
really hard.
Rob Collie (00:54:59):
There's a one of our episodes of this podcast is with a guy named
Kevin Overstreet. Sooner or later, if you're going to be doing
Power App stuff, you need to meet this guy. He got into Power BI
came to one of our classes, one of my classes many years ago. But
then with Power Apps, boys he found his calling. He's almost
disposed of Power BI. Like, I'm so over that. I was a Power App
person all along, I just didn't know it. And I think it has a lot,
I'm not going to say potential. I'm going to say it's got a lot of
value. It's in two places. Number one, you were giving an example
of upstream from Power BI. It's part of the collection process.
Yes, I don't have any problem at all with SharePoint being the
place where the data was stored. I'm really the opposite of
snobbish with this stuff.
Rob Collie (00:55:53):
In fact, even if you were like putting it into Excel Online, I
wouldn't have a problem with that either. Because really, the
problem with Excel is data collection is validation and things like
that, that the Power App will enforce for you. And I know that the
query speed against an Excel spreadsheet isn't so great, but surely
not super great against SharePoint either. So it's like six or one
half dozen or another. But there's also downstream from Power
BI.
Rob Collie (00:56:20):
And I think there's every bit as much value in potential there as
there is upstream. And this is going to be sort of almost one of
our companies messaging and marketing focuses next year, is that BI
doesn't make a bit of difference until someone takes an action,
that's a better action than they did the day before. I can inform
you to death, and actually not make a difference. If you've been in
one of my classes, I even talked about this a little bit, depending
on how much time we had. It's all about the action. So when you
think about it that way, this has been dawning on me over the last
six months. It's like, why does the dashboard, the report, why does
it get to take a total pass? It just gets to give up, when it's
time to actually make an improvement. You can just look at the
report and go the reports just looking at you smugly saying, "It's
not my problem. It's your problem."
Rob Collie (00:57:16):
Really simple example, trivial. And I use the words dashboard and
report interchangeably. I don't mean Power BI, capital D,
dashboard. Like to me a dashboard is a report, is a dashboard, the
same thing. If I'm looking at a dashboard, and I see some, I don't
know, a customer, or a prospect, someone who's thinking about
hiring us. And I want to change their status, I want to say we have
this person marked as, probably not going to do business with us.
But then I look at it and I go, "That's wrong." This is a valuable
person that we need to be spending more time with. Well, the
dashboard could say, "Okay, well, that person's name Jimmy Smith,
go get them. Have fun." But it doesn't take much effort to, in that
table visual or whatever, to put together a URL field, that is a
link to Salesforce or CRM or whatever that links to that person's
record.
Rob Collie (00:58:13):
Maybe it doesn't like change that person's record. But it takes you
there. It knows who this person is, right? The other example I'm
always using, and it's just doesn't really apply, maybe it does to
EGPAF. If you notice on a report that warehouse six is low on
product or low on test kits or something and they're going to run
out before you expect it. Imagine having buttons or something in
that report that allows you to go transfer some inventory from
warehouse five, or order more of that particular product from your
supplier with a rush order on it or something like that. There
could be a Power App, could be, it's just not a universal thing. It
can't be everywhere, because you can't anticipate what every action
is going to be. But why not embed a Power App in your dashboard or
elements like that, for directly taking action?
Shannon Lindsay (00:59:05):
Mm-hmm (affirmative). We actually did some of that with a Power App
inside of a report.
Rob Collie (00:59:10):
Cool.
Shannon Lindsay (00:59:10):
And this was the health facilities change their names all the time,
even though it's the same facility. And people get upset if things
aren't exactly correct. And then of course, we need a master list
to be able to understand what's what. So at the start of that
previous project that I was talking about where we're collecting
data with Power Apps, we also put a Power App right inside of a
report where people could confirm that things were correct. If they
weren't correct, they could make changes and then that all wrote
back to the database. So we have done a little bit of that, but
probably not as much as we should have.
Rob Collie (00:59:44):
Well, no one does. We always judge our own efforts more harshly
than we would if we could sort of see the whole universe. At the
beginning of this conversation, I forget, someone said the
nonprofit space was slower to take this stuff up than others. I'm
like, no. I mean, I am shocked. It's been more than 10 years now,
literally, more than 10 years since I realized that this
platform... Which didn't even really exist under the current name.
Power BI wasn't a thing. All we had was Power Pivot. We didn't even
have Power Query yet. More than 10 years ago, I was like, "Oh, my
God, this is going to change everything."
Rob Collie (01:00:25):
And it's going to happen overnight. It's just so obvious.
Everyone's just going to be like hoovering it up. It's like two
years from now, everything's going to be different. No, still
pretty early. 10 years later, you say like, "No, we probably didn't
do enough of that." Oh, that's a good mentality. That's a good
mindset, we can always do better. I think that Microsoft sort of
organizing these technologies together in the same division, and
renaming the Insight Summit to be Business Applications. It's like
they were a few years early on it, or maybe I'm just really saying
I'm just a few years slow on realizing what we're really doing. But
like, it's the action loop. It's the improvement loop. And so
there's upstream collecting, like we talked about the data
collectors. The tuberculosis screens, whatever. And then there's
the downstream, actually taking action. So we've got these sort of
two frontiers upstream and downstream from Power BI. With Power BI
still in the middle.
Rob Collie (01:01:24):
I think it's a very promising time. We're not even dead ending or
pigeonholing, as professionals in this space. Stephanie, you
mentioned that you went back to grad school for a while. What was
it that you studied? And when was that?
Stephanie Bruno (01:01:39):
I went back in 2016. And I just graduated last year. So that was a
relief. It was a lot more work than I expected. I went back and I
got an MSIT with a focus on business intelligence and data
analytics.
Rob Collie (01:01:55):
Awesome. Okay, so you got a master's? Right. That's what the M
means.
Stephanie Bruno (01:01:59):
Correct.
Rob Collie (01:02:00):
Okay. So what are sort of the highs and lows of the things you took
out? Because and I'm asking this, even from a personal perspective.
I'm not doing this to set you up for the, and how much have you
used it? I'll probably do that, too. But every now and then, I
guess, it's funny, that'd be like driving around somewhere. And
I'll see a billboard for like a Master's of Data Science. And I'm
like, I could try that. Maybe that would be a good thing for me to
do, even just for awareness sake. What is it that I'm missing? But
I'm also at the same time, skeptical of it, because I'm sure that
two thirds of what some academic would be trying to tell me would
be like, "Oh, come on the world doesn't work that way." You
mentioned at least a little bit of that.
Stephanie Bruno (01:02:40):
Yeah. You mean with the clean data?
Rob Collie (01:02:43):
Yeah. There's just something funny about listening to someone. And
maybe this isn't the way the professors are. But some of them I'm
sure are this way they live their whole lives in an ivory tower.
Then telling a citizen of the world what the world is like, when
they've never been there. I would have a real problem with that
mismatch of mindset, I think, but what were the most valuable
things that you've taken out of that program?
Stephanie Bruno (01:03:07):
It was valuable. And I'm really glad I did it. It was hard. I
didn't plan on doing it, honestly. I just, I applied, because the
application was free. And I thought, I'll never get in. Because it
was Carnegie Mellon University. But it was free. And I think I was
having a bad day. So I was like, "I need a little boost, if I get
in, that'll be great". But it's also going to be way too expensive.
So this is never going to be a thing I actually do. And I was with
Shannon, when I actually submitted the application. It was on that
same trip to Tanzania. But then I got in. And not only did I get
in, but they gave me a huge scholarship. It was a combo women in
tech and nonprofit scholarship.
Stephanie Bruno (01:03:51):
So it turned out I hardly had to pay anything for it, which was a
real shocker. So when I saw that, I was like, "Oh, boy, I guess
this means I am actually going to grad school now." So this was not
my plan, but I'm doing it. I guess you know, that's life, right?
It's a roundabout way of the way things happen.
Rob Collie (01:04:09):
Well, it does sound like having a bad day is a really important
driving force in your life, right [crosstalk 01:04:14]?
Stephanie Bruno (01:04:14):
Apparently.
Rob Collie (01:04:16):
The diff eq guy he had no idea what was about to hit him. You just
know just caught me on the wrong day. I'm going to [crosstalk
01:04:23] out-
Stephanie Bruno (01:04:22):
I really have to examine my motivations for why I do the things I
do.
Rob Collie (01:04:28):
Just be careful not to have too many bad days. You'll end up taking
on too much. Four years of grad school after one bad day. That's a
big commitment.
Stephanie Bruno (01:04:36):
Oh, yeah. My husband keeps saying he's afraid I'm going to have a
bad day and go try to get a PhD next. But no, that's not
happening.
Rob Collie (01:04:44):
Four years is a long time even part time.
Stephanie Bruno (01:04:46):
Yeah.
Rob Collie (01:04:46):
Well, Carnegie Mellon is in Pittsburgh, right?
Stephanie Bruno (01:04:49):
That's why I did it. Yep, that's where I live.
Rob Collie (01:04:52):
It wasn't a correspondence learning.
Stephanie Bruno (01:04:55):
Yeah, they actually do have it. You can do it fully online or you
can do it fully in person or you can do a combo. So I opted to do a
combo because I wanted to see what it's like to be a student again,
that'll be so cool. But it actually, the in person classes, it
really didn't turn out to be what I expected. I thought I'll meet
some more people that are also data nerds. But outside of the
Microsoft ecosystem, which is one of the things I did get out of it
was that I wanted to be exposed to non Microsoft tools. And just
data in general, not Microsoft data tools. So that is something I
got out of it.
Stephanie Bruno (01:05:34):
But it turned out, it was pretty intense. And so the kids that were
there full time, they had already sort of formed their little study
groups and their friendships. And so it wasn't really easy to get
to know people and the occasional in person classes that I went to.
So that was a bit of a disappointment. I thought there was going to
be this whole new community, I'd be a part of in addition to the
Power BI community, but that didn't happen so much. So that was
something I didn't get out of it that I hoped to get out of it.
Rob Collie (01:06:04):
If I went back like those people would all be half my age.
Stephanie Bruno (01:06:07):
Oh, yeah, that's exactly what they were.
Rob Collie (01:06:09):
Just crazy, right?
Stephanie Bruno (01:06:10):
I'm pretty sure I could have been their mom. And that was weird.
And I was definitely older than some of my teachers, too. That was
also weird.
Rob Collie (01:06:18):
Yeah. I would be too, wouldn't I? Don't even think about that.
Stephanie Bruno (01:06:23):
It's true. And I took one class, the one that Shannon and I worked
on that project for with the loss to follow up people. And the
teacher was so young Shannon, LinkedIn stalked him for me. Because
I didn't want him to see that I was LinkedIn stalking him, you
know. So Shannon did it for me. And we saw that he was like, I
don't know, 24, 25. So we refer to him as baby duck after that.
Rob Collie (01:06:47):
That's great.
Stephanie Bruno (01:06:51):
Man, that was weird. But he was brilliant, and it was a really hard
class. But I did get a lot out of it. And that was actually the one
class where he focused on dirty data. And what do you do with
missing data? And what are all your options for how you can deal
with it. It was really good. I think it is fair to say that I'm
probably not using a lot of what I learned, which is a
disappointment. But I would still do it again. And it was worth it.
And I learned a lot. And I was exposed to a lot more ideas and
methods that I'm not using. But it was good to learn. And I do
think it, it has helped me with how I look at things, and maybe how
I attack problems.
Rob Collie (01:07:33):
It's like the question, I asked Shannon, six months to build the
Power App, and how long would it take you today to do something
similar? It might be that the reason why you're not using so many
of those techniques, or whatever that you learned is that you're
just not being hit with a necessity for it.
Stephanie Bruno (01:07:50):
I think that's exactly true.
Rob Collie (01:07:53):
The model where you're talking about the machine learning project
that you attempted, that is exactly the customer attrition problem
that like every business in the world is like, we've got a good
regular customer, when are they starting to show us the warning
signs that they're about to defect?
Stephanie Bruno (01:08:13):
Right.
Rob Collie (01:08:13):
I know, it's a different real world problem. But as far as the
machine learning, it's the same thing but you're feeding it
different factors. I would expect that you'd run into something
like that, again, pretty quickly. That would be something that you
would cross paths with, again, pretty soon, if you haven't
already.
Stephanie Bruno (01:08:31):
Yeah, I hope so. But my job is, it's not exactly play with all the
cool new toys and try all these exciting new methods. It's really
deal with this data and make some, help us make sense of it and
help us make good decisions with it. So-
Rob Collie (01:08:48):
A little too much humanity in there, not enough room for machine
learning.
Stephanie Bruno (01:08:52):
I think there is. It's just to get to that point, we have a lot of
hurdles to overcome first.
Rob Collie (01:08:59):
That's something that you do discover pretty quickly about machine
learning is that the same data sources that might be okay, for
aggregate analysis. The things that Power BI is particularly really
good at, what's our percentage turnover, right? Or whatever, that
same data set might not power a machine learning model well, at
all.
Rob Collie (01:09:19):
Shannon, that thing you were talking about with the where they were
gonna tally it and lose all the individuality. Hey, maybe that
would have still been fine for certain kinds of aggregate analysis.
But it would have completely precluded any possibility of machine
learning in the future. It's like what would you do with it? It's
gone. All the variation is gone. All the richness is gone when you
collaborate together like that.
Shannon Lindsay (01:09:47):
I do you think there is a lot of opportunity for machine learning
and new technologies and all of that. But another thing that a lot
of these nonprofits particularly the health related nonprofits are
contending with is, a lot of their key staff are researchers and
academics. And they're very well established and pretty resistant
to change. And we had great success, pushing a new technology and
rolling out a whole new way of looking at data and doing things. I
don't know what the limit is for the uptake of new technology. But
I think looking at your data and analyzing it is one thing, but
then predictive modeling and machine learning. And all of that is a
lot scarier to that group of individuals. And maybe misreading that
stuff. But I think it's a harder hurdle to pass with that
group.
Stephanie Bruno (01:10:41):
No, I would agree.
Rob Collie (01:10:43):
Yeah. If you're saying that human beings are resistant to change,
you're going to get nothing but agreement. You got to be motivated,
right? Suffering. If you're saying that academics are even more
resistant to change, then again, that's the nightmare. I've been
trying to wake up from for the second half of my life is that,
there's such a snobbery associated with a lot of all of that. That
wasn't warranted and wasn't actually helpful. Again, I had to wake
up from that bad dream and go, "Oh, okay. Part of the reason why I
feel compelled to beat up on calculus, right? It's truly, it's that
trend in my own life, really. Shannon, you have changed jobs. What
are you doing these days?
Shannon Lindsay (01:11:28):
I have changed jobs a couple of times since I was last working with
Stephanie. I did a stint in consulting, where I worked solely with
nonprofit and government clients. I did some teaching, which I
really enjoyed. And now I just recently moved over to the worldwide
learning team at Microsoft.
Rob Collie (01:11:49):
So from one nonprofit to another?
Shannon Lindsay (01:11:54):
I mean, I do carry a heavy load of guilt. Both because of how I was
raised, and also because it's really my first time working in a big
for profit corporation. And it's really too soon to say how I feel
about the organization. But just within the first couple of weeks,
the sense that I'm getting about the culture, and the team that I'm
on, makes me feel like I did not, in fact, sell my soul to the
devil. So...
Rob Collie (01:12:21):
People at Microsoft are generally speaking really good people. I
really enjoyed my peers, my colleagues. I was there during the
antitrust trial back in 2000. And at that moment, we could see like
that our leaders, were the bad guys. They really did not come off
looking well there. And we didn't feel great about it. But so
Microsoft, as an organization, especially back then, played very
rough in the marketplace. Not necessarily always with the cleanest
of reputations. But inside the company was a very different story.
And I think today's Microsoft is quite a bit better, sort of
citizen of the world than the one I sort of grew up in. Worldwide
learning, can you remind me what that is?
Shannon Lindsay (01:13:07):
Eloquently? No. But I can try.
Rob Collie (01:13:11):
We don't do eloquent here anyway, really. So it's not our
business.
Shannon Lindsay (01:13:15):
Well, so I'm on the global technical learning team of worldwide
learning. And so our goal is just to help people learn this
technology. I mean, ultimately, it's a course, to drive sales. But
I feel like, I am in a really neat position where I get to develop
content to help people learn Power BI and Azure. And, specifically,
I'm focused on the data analyst role, which I think suits where
I've come from. I don't come from a technology background. And so I
speak plain English, and I can try to put things into the mindset
of, okay, coming from big flat Excel sheets. And the types of
things that analysts would have to try to deal with. So our team is
really just responsible for providing the content for folks to
learn and then subsequently get certified on those
technologies.
Rob Collie (01:14:06):
Got it. So it's external. The people doing the learning are
external to Microsoft. And it's about Microsoft technologies. So
what you're really doing is trying to put our training business out
of work.
Shannon Lindsay (01:14:21):
Absolutely not.
Rob Collie (01:14:21):
Yeah. Oh, come on. Look, it's all fair. But training isn't, even
though that's how the two of you know us. Training is now a small
fraction of our business. Most of our businesses in actual project
execution consulting. At breakneck pace, that's our
differentiators. That we move a lot faster than really, I think
anyone else. But we still do training. In a way it's good for the
world, even for us. If the training becomes your sort of
increasingly commoditized and increasingly easier to get. Just
don't make it too good. Because we still have to compete on
quality.
Shannon Lindsay (01:14:55):
Well, I'm on the team. So it's going to be good, right?
Rob Collie (01:14:59):
I know. But, maybe you could just occasionally leave something out.
Just... I know you won't. I'm not really asking you to. Obviously,
it's just a funny thing to say. Well, that's awesome. I think
that's a great spot to be right. Take your experiences. And you
know the things you're saying, when you were laughing about not
being eloquent, but that's great. I'm often told, what genius it
was to have written my books in the voice that I did instead of the
technical book voice. And I'm like, it wasn't genius. It was just
what I had to do. I couldn't get through it writing in the
technical voice. I tried, it wore me out in 10 pages. I did it for
me, not for the world. But it turned out that was the way it should
be done. So be that thing that you're describing, right? That non
eloquent, but sometimes eloquent.
Rob Collie (01:15:53):
But the human voice that walked in those shoes. That's what the
world needs. The vast majority of people who need to be using this
stuff are not coming from a traditional IT background. Even when
they are, when they're successful at it. It's to the extent that
they're also that hybrid. They are the Stephanies of the world that
you wouldn't necessarily guess were The IT Crowd.
Rob Collie (01:16:18):
So how has COVID impacted your organization, Stephanie? Shannon
mentioned earlier, screening for tuberculosis? Is there a Power App
now that's like involved in the screening for COVID? Are you even
able to screen for COVID? How is that impacting your
operations?
Stephanie Bruno (01:16:34):
Yeah, pretty significantly. Number one, we have so many data
collectors. So the way we work is we have office in each country
that we work in. And I think there are 10 countries that we work
in, in Africa. But then they do work to support 1000s of health
facilities are around them. And part of the work they do, is to go
out and collect data from these paper forms that Shannon mentioned.
So obviously, that has changed, because they can't go to the health
facilities now and collect data from the paper form. So that's a
huge problem.
Stephanie Bruno (01:17:05):
It's also impacted us because a lot of our work involves so much
travel. So many of our HQ people go to Africa so frequently, so
nobody's doing that anymore. But on the plus side, that's created
that need for being better about collaboration virtually, which I
think is fantastic. So they are doing better with that. It's
speeding up the need for electronic data collection tools, so that
people don't have to physically go and get the data off the paper.
We are not doing a lot, we haven't added a lot of COVID related
work onto our plate, because we don't have funding for that. So
it's the opposite of a for profit company where you do the work,
and then you get paid for it. We have to get donors to give us
grants, obviously, to do any work. So we don't have any grants for
COVID. We don't have any grants to do that work. Our mission is
pretty much HIV. I feel like I'm bouncing all around with this
answer. Because it's so big, but...
Stephanie Bruno (01:18:11):
The other thing about artwork is that we're so, it's so important
for us to meet donor set targets. Because the big donors give us
money, but they they give us the targets. They say, "Okay, we're
going to give you this money, but we expect you to test X number of
people in this region. And we expect 90% of those people to get on
treatment." And so we have to meet these targets, or we risk losing
funding. Then potentially having to close down a whole country
office and all those people lose jobs. So it's really important
that we meet these targets. Sometimes the targets seem
unreasonable, which is hard. But the targets didn't change because
of COVID. So the targets are still there. So we are-
Rob Collie (01:18:56):
Oh, that's so hard.
Stephanie Bruno (01:18:57):
It's really hard. So then it's important for us to try to be able
to explain the discrepancies. And so we've had to add COVID related
Power BI reports on to our work, to help us explain the impact of
how COVID is changing our results, really.
Rob Collie (01:19:16):
Is the Gates Foundation one of your founders.
Stephanie Bruno (01:19:19):
No.
Rob Collie (01:19:20):
That's actually the best answer, because I was going to say, those
sorts of targets, meet these targets or you miss them. That sounds
like Bill Gates.
Stephanie Bruno (01:19:29):
Well, right now, it's actually Deborah Birx, the person on the
COVID team. You may have seen her when Trump was suggesting we put
bleach in our lungs. And she was shaking her head or maybe looking
down at her feet.
Rob Collie (01:19:42):
That's right.
Stephanie Bruno (01:19:44):
I can't remember. No, she's the one who her team sets our targets,
really.
Rob Collie (01:19:48):
Okay.
Stephanie Bruno (01:19:49):
Yeah.
Rob Collie (01:19:50):
Interesting. Okay.
Stephanie Bruno (01:19:51):
They're very data oriented, but sometimes those targets are
unrealistic. And that's been hard because if we don't meet the
targets, they might say, "Okay, well, we're going to still fund you
at path. But you don't get to support this facility anymore."
Because they don't have enough HIV positive people that they need
any money. That's what the really scary thing is. Is if we don't
meet targets, we can lose funding. And that really does mean that
health facilities lose funding.
Rob Collie (01:20:19):
You mentioned that a lot of data driven people in celebrity
positions in this space, do you think that has anything to do with
the back to the Hans Rosling TED talk? Do you know what I'm talking
about? Where you talking about mortality rates-
Stephanie Bruno (01:20:34):
The one with a bubble chart-
Rob Collie (01:20:35):
The bubble charts, that's racing up and he's narrating like it's a
horse race.
Stephanie Bruno (01:20:39):
Yeah.
Rob Collie (01:20:40):
And now his son is carrying on that effort. Years ago, I had a
Twitter exchange with him. He was like, "Wow, I can't believe how
much Power BI can help us." We tended to avoid the Microsoft tools
in the nonprofit space, because of the specter of a big
corporation. It's way cooler, and sort of considered more like
egalitarian to use Linux tools or something. There's actually a
stigma, especially in the academic world, that lingers about
companies like Microsoft. And for a long time, companies like
Google, were spared that bias. It's sort of a full cycle now,
Microsoft is starting to look like a relatively benign entity
compared to some of these other tech giants.
Stephanie Bruno (01:21:25):
Yeah. You did ask at one point, maybe an email I don't know about
if we encountered any roadblocks. And I think that was what is
that. We were pushing Microsoft tools. And in this space,
traditionally, it's not so much about like the cool kids being
Linux, it's more about open source. Because one of the biggest
things that we have to do coming from the US and bringing solutions
into Africa is we have to make sure that when we leave, the tools
can still be used, it's still can be managed. And so oftentimes,
it's just like, right out of the gate requirement that it's open
source. Because, we don't want to have to pay for anything once you
leave. So it's got to be open source.
Stephanie Bruno (01:22:08):
I mean, I remember being pretty nervous that I'm pushing this
saying, and it's not open source, it's not free, it is a tool that
has to be paid for. Since we started using it, so many other NGOs
have also started using it, the CDC has started using it. So I
mean, I breathed a huge breath of relief about this. Because I
think that, I didn't take us down a path that was going to
ultimately get us in trouble. But I think part of that is because I
have to preface this by saying, I really have drunk the Microsoft
Kool Aid. And I love Microsoft. I think partly, hugely, because of
how much they've done for our organization. It's been amazing. And
also how much they do for nonprofits in general, with the really
big discounts they provide. And the training and the resources they
offer. It's impressive, it's really good.
Stephanie Bruno (01:23:00):
I would definitely argue that even if we left tomorrow, I think
that it's still sustainable. It's not something that people can't
afford to use. And it's not too hard to use, it's just all going to
fall apart, if we leave.
Rob Collie (01:23:13):
And most people don't realize this, but Microsoft software is
relatively affordable, relative to the competition, even at full
price. So then you add in the the discounts and stuff. Microsoft
model is the bulk model, their goal is to have the whole world
using it. And if the whole world is using it, they don't have to
charge any one particular organization, necessarily all that much.
Yeah, so that's cool. I hadn't really thought about that really the
headwind that you might get, but thinking about that conversation
with Ola Rosling brought it back.
Stephanie Bruno (01:23:43):
Yeah, it was scary. But I think we're cool now. I think we're
okay.
Rob Collie (01:23:47):
No one ever got fired for going open source in the nonprofit
industry, right? You had to stick your neck out and suggest
something really risky like Microsoft?
Stephanie Bruno (01:23:58):
Yes. Something risky like Microsoft. Exactly.
Rob Collie (01:24:01):
That kind of came up with Austin when he was on the show, talking
about the museum space, the conservation space that he works in,
and the inherent distrust of private enterprise in that space,
right? If you're a for profit company, you're to be watched
closely, which I think is completely right. It's a good thing to
keep an eye on. So it's hard running a tech startup type business
that isn't predatory, in a way. Because you're still going to be
viewed as suspiciously as if you are, but you're not getting the
benefits of being predatory either.
Rob Collie (01:24:40):
So well, I've certainly enjoyed this. I'm really glad that we were
able to do this. I'm glad that the two of you took the time. I
really appreciate it. Thanks for joining us.
Stephanie Bruno (01:24:48):
Thanks so much, Rob. I'm really glad you reached out to this has
been fun.
Rob Collie (01:24:51):
Maybe we'll do it again sometime. We just got to wait a little
while for the, to manufacture more things to talk about.
Stephanie Bruno (01:25:00):
Oh, we got plenty of stuff to talk about.
Rob Collie (01:25:03):
Indeed. 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, Luke P-L-U-K-E-P at powerpivotpro.com. Have a day to
day!