[MUSIC]
[MUSIC]
>> Hi. Welcome to
another session of our customers
Power BI sessions in the Microsoft
Business Application Summit.
My name is Lauren Faber
and I am a member of
the Power BI Customer
Advisory Team and I'm
super excited to have Melissa,
Sean, and Dr. Renda with
us today from Humana.
Dr. Renda, if you
could start us off by
telling us a little bit more about
what we'll be discussing today,
more about Humana, and
then introduce yourselves,
that will be fantastic.
>> Well, thanks, Lauren.
yes, today we're going to
give a brief overview of Humana.
We'll talk about our roles,
we'll talk about some
opportunities and
solutions with using Power BI tool,
we'll give a great demo of a
new tool that we've developed,
and then we'll
summarize at the end of
some key drivers around adoption
of these type of tools.
So to get us started,
Humana is a health and
well-being organization.
We're committed to driving
best health outcomes for millions
of medical and special team members.
We do that with three
principles in mind.
Number 1, we want to deliver
an easy and seamless
customer experience.
Number 2, we went to help members
achieve their best health.
We do that through five points
of influence; primary care,
social determinants of health,
pharmacy, home, and
behavioral health.
We power those two things with
our third principle and
that is the power with
integrated technology.
We have a digital health
and analytics team,
we have an enterprise
clinical operating model,
and of course, we have a strong
partnership with Microsoft.
Let me tell you a little
bit about my role.
I'm the Associate Vice President
of Population Health and now
I'm going to pass on to
Sean to describe his role.
>> Hi everyone. My
name is Sean Chandler.
I'm a Senior Business
Intelligence Engineer
working in the Enterprise
Analytics space.
Specifically, I build
Power BI tools for
teams just like Dr. Renda's
and I consult on a variety
of Power BI applications
throughout Humana as
part of the Analytics Platform
as a Service team. Melissa?
>> Yes. Thanks, Sean.
So Melissa Hooper,
I'm a Product Implementation
Lead and my role
in Enterprise Analytics
and Analytics Platform as
a Service is to bring
together business teams like
Dr. Renda's and our BI engineers,
like Sean, to create wonderful
user-centric designed BI report.
>> Thanks for that
introduction. That was perfect.
When it comes to BI and data,
I know that Humana faces
a situation that's very
common in this industry
and in many industries.
Melissa, could you
tell us a little bit
more about that BI opportunity?
>> I would be happy
to, Lauren. Thank you.
In 2018, Humana undertook our
beginning of our digital journey.
We really formalized it
by creating the vertical
that Sean and I set in called
digital health and analytics.
As part of the business case for
digital health and analytics,
we did a survey with an
outside consulting firm.
What we found is that our
analytics professionals are using
47 disparate data
sources and BI tools.
Just to let that sink
in for a moment,
you can imagine that it's very
difficult to become a
master of anything if you
have 47 different tools
that you should be a
little bit familiar with.
Since 2018, our team,
along with the rest of
digital health and analytics,
has been working to
standardize our data sets,
standardize our reporting tools,
and we're really doing that,
moving toward this
idea of the data in
our Gen2 data like in Azure and
using Power BI specifically,
but also other elements of
the Power Platform so that
our analytics professionals
can really become
experts in the Power
Platform and not have
to spread their learning journey
across multiple sources.
What we're really working toward
is this data-driven culture,
and that's where our
team analytics platform
as a service comes in.
>> What solution were
you able to come up
with to take advantage
of this BI opportunity?
>> Our current architecture,
and we're still in
the process of making
this transition is we have
very siloed data sets.
As you can imagine, finance data
about a member or clinical data,
all of this is important and it's
all part of the single
view of the member,
but right now, it exists in silos.
So data across Humana is incredibly
siloed and it takes our
analytics professionals weeks,
sometimes months to be able
to track down the tables,
the sources, get connections and
access to all of the
data that you need.
All the data is currently On-Prem in
EDW warehouses and we have
other on-site assets,
we're using a Taas, we're using SaaS,
all of these things to perform
the data transformation.
What we're working on
right now is migrating
our data assets into our Gen2
Microsoft Azure Data Lake,
creating common data sets,
getting access through Azure
Active Directory groups,
and really using the power
of the Power Platform
and Azure to create these
citizen analysts where we're
delivering the right information
at the right time to
the right-hand associates
to really help
our members achieve
their best health.
>> Maybe I could take a minute to
talk about the subject matter that
we're designing our specific tool on.
Social determinants of health are
the conditions and the
environment in which people live,
learn, work, play, worship, and age.
Essentially, they're the upstream
influencers of poor health.
Another term that we use that's
health-related social needs and
those focus on individual
needs in the shorter term.
Examples are things
like financial strain,
social support, food
insecurity, housing quality.
We measure the impact
of social needs or
social determinants on health
related quality of life.
We use the CDC healthy-based tool.
Now, Sean, I think we have
a demo ready to share.
>> Yes, we do. I'll go
ahead and take over.
From a design perspective
for this Power BI report,
the real big challenge in
front of us, of course,
was taking survey data related
to social determinants of
health from multiple vendors
and tying it all together.
In Power BI, we could tell
a cohesive story about these
different domains and about
the screenings related to them as
a whole rather than
in these data silos.
So to execute this,
it was really important that
we find a way to compile
those multiple data sets from
these different
screenings providers into
a single master table that
Power BI could consume
for insights and in a way that
was attractive to our users.
Currently, we have about nine sources
for our screenings data and we know
that this number is
going to continue to
grow as we keep iterating this tool.
One of my colleagues, Derek Childers,
who helps me on the
data side of things,
uses SaaS to take
those disparate data sources
from IT for these screenings,
bring them together into
one single table of
members and screenings,
and then he puts that into
a data warehouse environment
that I can grab in
Power BI just using a
simple ODBC connection.
In Power BI, my goal then is
to take this master data set,
which is a very wide
table that you can see
here has a bunch of
different columns,
and you can see how many of
these are just numeric counts.
What I need to do is I need
to figure out a way to
transform that table
into something that
Power BI can more easily consume.
I think the first time that I
started building this tool,
I noticed that some of
my visuals were taking
45 or even 60 seconds to load,
which of course isn't
a very good experience
for the users or the
consumers of this report.
Fortunately, one of the first
places that I started in
this Power BI report is just in
the query editor where I can
take all of these numeric
fields that contain
my counts of screenings for
the different domains and
also my counts of members.
I can just pivot this data set
with a simple walk-through of
these processes so that I end up with
only one column that contains
my different metrics,
I can identify those as
screenings or member attributes,
and then it just makes
it so much easier
for me to sum up those values
into something that Power BI
visual can consume very easily.
While that might not sound
interesting to everybody on the call,
it really is a powerful feature in
Power BI when you're
dealing with larger,
wider data sets that you need
to create a whole bunch of
different visuals for your users.
>> Sean, I understood every fair
word of what you just said.
But what I will say is
what you're getting
there is really important
to our strategy.
As a customer of this tool,
and as I try to grow
our social determinant
strategy in our business,
we are screening more of our members,
we're collecting more data from
different sources within
our organization,
and so it becomes really
critical to collect,
to standardize, and to aggregate
this data in a seamless way so
we can get it into a
tool and then create
the cuts and the
visualizations that we need.
>> I definitely
appreciate that point.
To that end, the collaboration
that our team under Analytics
Platform as a service has had
with Dr. Renda and his team to
have everyone involved in a
weekly in the iterations of
this report has really
made it so much easier for
our team to iterate
on this tool and add
new dimensions to it
as we need them to
add supporting metrics to
this as they've requested.
You can see up here on the
top right corner that this
is the 19th version of this tool.
I'm already hard at work
on the 20th version of it,
which is actually maybe our
largest update yet to this tool.
Because of the close
collaboration between
our teams and the pipeline that
we've set up for this data,
we're able to turn around
very new versions of
this tool in a matter of days
or weeks instead of months.
I know that sometimes in the past,
adding or changing some of
these tools can be very
cumbersome for teams.
>> Yeah, I think
that's a great point.
I know every element of
the dashboard in these tools
is carefully chosen through
our weekly meetings but I love
the ability to iterate on
the fly as well because
previous to this
and other tools that I've
used that it's taken,
we have to wait for IT cycles and
data cycles and can sometimes
take months even to make
a small change to meet
a new business need.
I'd love that adaptability
within this tool set.
>> Fantastic. From a
visual standpoint,
I also wanted to highlight
just some of the
built-in Power BI features that we're
using in this tool to make for a
better experience for the users.
Because I know we have a lot
of people who use this tool,
who this may be their very first time
even dipping their feet
into the Power BI waters.
So one of the first things
that we wanted to make
sure that we did with
this tool was take all of
these different surveys,
each of which may have their
own questions related to
these different domains and create
a very seamless way for users to,
for example, just highlight
food and security
screenings or just
highlight social support.
What I've done up here is
I've created a list of
icons that I can apply
in the Chiclet Slicer,
which I've downloaded from
the Power BI marketplace.
These icons were selected in
collaboration with the Bold Goal team
to represent these different domains.
By using them and the Chiclet Slicer,
users have a very intuitive way to
immediately just filter to the
food insecurity screenings,
or if they're interested
in social support,
they can do this on
the fly and the tool
updates and that you get the
information that they've come for.
This is something that in past tools
was just not as easy to do as
Power BI has made it for us.
>> Yeah, Sean. I think
that's really important.
Another one of a
parallel exercise that
we're taking in the
organization is trying to go
through a data certification process
for social determinant data.
We're taking data from all
these disparate data sources,
we're trying to standardize it,
and really map it to
a specific domains.
Then we want to see it
visualized in a tool like this.
Exactly what you just described
is really helpful to us
because we can take the totality
of social determinant data,
and then we can slice down into
one single social determinant domain,
and understand how that
impacts utilization,
and how it looks from a
prevalent standpoint.
It's a really important
functionality for us.
>> Continuing on the
subject of standardizing
the different dimensions and
the different views that
are in this report,
one of the early
exercises that our teams
participated in was taking the time
to carefully curate this list of
filters that we have in the filter
pane on the right-hand side.
We knew that these were
the dimensions that we really
wanted to highlight in the report,
and give users the ability to come
in and filter this data by market,
for example, or by region,
or by state, or by line of business.
But something that we
were able to do to take
that concept even further in Power BI
was to create a summary table that
contains the different values
within these dimensions,
and put them into their own table
that we can then connect
back to the source dataset.
For example, if someone wants
to take this bar chart,
and if they don't want to filter
by every single state to figure
out which state had the most
screenings for food insecurity,
they have the ability to actually
take any of these filters,
and select them from
this slicer here,
and update that visual on the fly.
If instead of seeing
this visual by state,
they wanted to see it
by division because
that's more relevant to
them, they can do that.
You can see how quickly
it replaced the values
that were in that visual,
likewise for region.
Just to give you a couple
of examples for something
that Power BI made really
easy and really gave
our users the ability to come in
and not have to spend so much
time going through the filters,
even though they're important,
they could literally
just apply those to a
visual within the tool itself,
and after one-click, be done and get
to the insights that
they were looking for.
>> Yeah, Sean. I think that
functionality is really important.
When we think about
end-users for this,
we have regional leaders,
we have market leaders,
all have different
perspectives depending
on their scope and
their responsibilities.
Being able to filter down by
those different geographies and
different lines of business
is really critical.
>> Something else that we started
running into as we were
developing this tool,
and I think this is
something that happens for
many developers in any kind of
business intelligence platform,
is the more stakeholders
you introduce to the tool,
the more users you
introduce to the tool,
the more views those people want.
We started running up
against the situation where
we could conceivably
have had more visuals,
and then could comfortably fit
on one tab of a Power BI report.
Another Power BI feature that I was
eager to implement here
that I think is made
for a really cool use case is
the integration of Power
BI's bookmark features.
Up here, we give the user
the ability to just toggle
quickly between some different
commonly requested
views on screenings.
Obviously, it starts
with just screenings
being a KPI that we wanted
to look at month-to-month.
The default option for this is
just a simple overtime trend of how
screenings are doing
and year-over-year.
But if somebody came to this,
and they specifically wanted
to know how many screenings
we were getting from each of
the different source
or different vendors,
they can just toggle this
bookmark to that view instead,
or if they want to
see if the screenings
broken down by the different domains,
that's also easily done.
I'll unselect "Food Insecurity,"
let the charts update here,
then they'll be able to actually see
the different screenings
broken down by the domains,
and what percentage
of those screenings
were positive versus negative.
I'll wrap up by also
just highlighting that
the tooltips feature
was also something
that the more I've
used it in Power BI,
the cooler I think it is,
and the more endless the
different applications.
We expected that a lot of
users would come to use
this tool to break down
screenings information by geographies
or by the different
states, for example.
By using the ShakeMap visual,
and integrating it with tooltips,
I was able to take inspiration from
the different views that I created
up here using the bookmarks,
and just create a tooltip
view for each state that
a user can easily just
hover over Texas,
and see a similar
breakdown or just like
an executive summary
level view that they
would see if they had toggled
through all the different bookmarks.
As I hover over any given state,
you can see how
seamlessly it updates.
Again, that's just another step
that we worked through that
make the overall experience
better for our users.
>> Sean, to me, this
is the easy button.
It creates an executive summary
for business leaders so
that just at a glance,
they can go and see some key metrics
for the geography that
they're responsible for.
I think that's really helpful
if they only have a short
period of time to go in,
and pick an insight,
and then make it,
and keep its vision.
>> Another cool thing that
we were able to do with
the tooltip feature in
Power BI that I thought
was something very unique to this
report that I had never done
before was I actually
used the TreeMap visual,
which is literally
over here just with
one single metric that I think
contains the number zero to
populate a block over here that
I then assigned a tooltip for so that
users can hover over these little
blocks that are very unassuming,
and actually see the definitions for
the different social
determinants of health domains.
Or if they want to see
the definitions for
the different metrics that are
visualized in this report,
or even the filters that are on
the filter pane to the right.
Users have the ability
to come in here and find
information about what
is the underlying data.
Because again, we have
users that come in that may
broadly understand what social
determinants of health are,
but they may have questions
about the different
fields that they're
seeing or some of the different
metrics that are being calculated.
We wanted to give users a subtle way
to educate themselves or find
information if they needed it.
I believe that that
concludes our demo.
Again, it's been a fantastic
experience building out this tool,
and it's really been exciting
getting to share it with you.
>> Thank you so much for that, Sean,
showing us the report that
you've been able to build.
It really showed me
the thought and the specific
design that was put into this.
I really love how clean
the report looks,
and how I can see that if someone
was coming to this
report not really ever
having used Power BI before,
because of the things
that you've implemented
like the bookmarks,
like the tooltips,
like the specific icons,
it would still be pretty easy to
navigate the report and to
understand what's going on.
You can really tell
that there was a lot of
good thought put into
everything that was built,
so thank you so much for that.
But we all know that coming up and
building the solution is
really only half the battle.
The other half is getting
users to adopt it.
What are some of the things
that you've been able to do at
Humana to drive adoption
throughout the organization?
>> Absolutely. Lauren, I'm
happy to take that one on.
It's part of what my job is
here on the analytics platform.
Some of our key adoption factors.
You can read this list,
but I would say in anyone's
starting this journey,
what was key to us in the beginning
was that executive
sponsorship and alignment,
and really getting the
leaders of Humana RC suite,
and our top-level executives
familiar with and
exposed to Power BI,
and what it could do to transform
the way they run their
business segments.
I think education and
communication has been key for us.
We talked about the disparate tools
that people were using before.
It was really hard in the beginning
for people to understand,
this isn't just another tool
and that can have 48 tools now.
This is one that you can put in
your tool bar that will replace
a myriad of other tools,
and it's not the same as all
of the other analytics tools,
but it does work in concert with
our Azure Data Lake with our
certified datasets to create
this holistic environment
ecosystem that is actually
more beneficial for our
analytics associates
and for that citizen analyst.
Our common measurement criteria,
you can't manage what
you can't measure.
We said it may use what we were
tracking some monthly active users
for the Power Platform as our way
of tracking enterprise adoption.
In the current times, catalyzing
events really gave Humana
an opportunity to rethink
how we were using our tools,
and that's where we get into some of
the digital transformation that we
actually saw as a result of COVID-19.
Our monthly active users since
the crisis really picked
up has increased from by 97 percent.
It's interesting because
of that 97 percent,
only about 55 percent is
directly attributable
to the COVID dashboard,
so what people are seeing is how
quickly Humana stood up in a crisis.
All of this various reporting data,
they're seeing that,
and they're thinking,
"If we could do that
in a time of crisis,
how can I move past the crisis mode,
and think about how I
can use Power BI and
the Power Platform in
my daily operations?"
I know Dr. Renda was part
of our leadership team,
who was deeply involved in
those early standing up,
the efforts around
measurement for COVID.
If there's anything you
wanted to add here.
>> Well, sure. None of us expected
the COVID pandemic to occur,
and when it did, very quickly,
we had a lot of different
things we needed to understand.
Tests that were available,
antibody tests,
who of our members
were testing positive,
and make some key decisions
about how to respond.
Having access to Power BI tools
very quickly build dashboards
to visualize all of
that information and to update
it in real time was absolutely
critical to both forming our
strategy and then executing on it.
>> Definitely no one wants to be
in the middle of the pandemic,
but it really is amazing that
we live in a time in the world
where employees are able to get
information using the right tools,
and be able to work from home.
Just as you demonstrated,
Humana has done a phenomenal job of
keeping their employees connected
throughout all of this.
As we wrap up, is there anything
that any of you would like
to add before we finish?
>> Sure, Lauren. I'll add two points
to summarize what our
work is around this.
Number 1, our goal is to create
citizen data scientists
to use these tool,
to derive actionable insight,
and to make key business decisions,
and that's really what the suite of
Power BI tools allows us to do.
The second point is, well,
we never expected COVID to happen.
When it did, having access
to tools like Power BI
really allowed us to quickly build
the dashboards that we need,
derive the insight that we had,
form our strategy and our response to
the situation and just got ability
to quickly adapt on the fly,
create things, get insights,
and then take action has been really,
really important to
our response efforts.
>> Awesome. Thank you so much.
I just want to say a huge
thank you, Melissa, Sean,
and Dr. Renda for all the effort
you put into this presentation,
and for your time today.
I really appreciate it.
>> Thank you.
>> Thank you very much.