►
Description
Speaker: Manish Sood, CEO & Founder at Reltio
Slides: http://www.slideshare.net/planetcassandra/3-manish-sood
The Life Sciences industry is undergoing significant changes in how companies do business due to recent legislative changes. The evolving landscape is forcing the Pharmaceutical companies to change how their entire Sales model and move from Prescriber based sales model to an Account based sales model, which has a downstream impact on Sales team organization, Field sales alignment, Incentive compensation and Marketing. This changing reality also requires that the Pharmaceutical companies understand and drive the changes to business strategy on insights driven by data about Prescription Sales, Medication Adherence, Claims, etc. to name a few categories of data sources. The required insights are derived from the convergence of data from multiple sources that include numerous internal applications, 3rd party data sources and social media. In this session, learn how Reltio is helping various Pharmaceutical companies cope with the evolving business landscape with a data driven strategy by leveraging the Reltio data science engine that runs on Cassandra.
A
Hi
everybody
thanks
for
coming
to
the
session,
given
the
size
of
the
crowd.
I
think
it
will
go
relatively
fast
and
you'll
be
off
to
lunch
pretty
soon.
So
you
know,
let
me
go
ahead
and
share
some
of
the
details
of
what
we
are
doing
and
how
we
are
using
Cassandra
and
especially
for
solutions
in
life
sciences.
A
The
capabilities
that
we
have
are
there
to
support
multi-tenancy,
make
sure
that
we
can
onboard
multiple
sources
of
data
and
provide
the
applications
on
top
of
that
the
customers
can
use.
At
the
same
time,
one
of
the
big
reasons
for
going
with
the
Cassandra
was
to
ensure
that
we
can
also
support
the
high
availability
type
of
scenarios
so
that
the
solution
is
always
up
and
running
in
terms
of
the
life
sciences
vertical
and
the
work
that
we
are
doing
here.
A
Some
of
the
drivers
for
that
in
terms
of
the
kinds
of
applications
that
you're
creating
as
well
as
what
the
life
sciences
of
the
pharmaceutical
industry
is
reacting
to
right
now
is
the
fact
that
the
sales
model
that
they
have
is
changing
and
earlier
what
they
used
to
do
was
they
used
to
engage
or
work
with
the
doctors
or
prescribers.
Who
would
be
so
and
so
responsible
for
writing
the
prescriptions
and
they
were
essentially
their
direct
contact
for
influencing
our
promoting
sales.
A
So
the
sales
team
alignment,
as
well
as
incentive
compensation,
was
based
on
that
direct
sales
results
or
performance
that
they
would
see
on
a
individual
doctor
or
prescriber
basis.
But
now,
given
some
of
the
recent
health
care
changes,
the
model
is
shifting
towards
account
based
sales
strategy
and
there
it
becomes
a
little
bit
more
complex
because
you
not
only
have
to
understand
who
the
prescribers
are,
who
write
these
different
kinds
of
prescriptions,
because
they're
still
going
to
be
involved
and
part
of
the
equation
from
a
sales
perspective
and
a
marketing
perspective.
A
But
at
the
same
time,
the
kaisers
or
Sutter
Health
type
of
groups
are
becoming
more
influential
and
powerful
and
negotiating
some
of
the
pricing
around
what
kinds
of
drugs
are
on
their
roster,
as
well
as
the
hospital's,
because
of
the
latest
changes
in
the
healthcare
space
are
forming
buying
groups.
So
you
know
a
term
that
you
might
have
heard
about
accountable
care
organizations.
They
are
necessarily
being
formed
to
drive
down
the
cost
of
procurement
of
these
drugs
and
the
services
that
are
being
provided
by
these
hospitals.
A
You
have
to
look
at
accounts.
You
have
to
look
at
the
value
of
the
count,
the
potential
sales
and
the
revenues
that
can
be
generated
out
of
it
and
also
competitive
insights
in
terms
of
what
are
other
pharmaceutical
companies
doing
with
the
same
set
of
hospitals
or
pairs.
Are
these
online
pharmacies
and
that's
becoming
increasingly
a
part
of
the
decision-making
process
in
terms
of
how
the
sales
and
the
marketing
organizations
within
these
pharma
companies
are
reacting
to
this
situation
and
in
order
to
you
know,
better,
illustrate
the
sort
of
the
complexity
of
the
problem?
A
A
But
at
the
same
time,
you
have
the
different
prescribers
who
are
responsible
for
writing
the
prescriptions
who
influence
and
refer
patients
to
each
other.
So
there
is
a
health
care
provider
to
a
health
care
provider,
type
of
affiliation
network.
That
needs
to
be
understood
to
and
better
understand,
the
influence
that
these
people
yield
in
how
they
will
be
able
to
help
promote
some
of
the
drugs
and,
at
the
same
time,
these
different
individuals
who
are
responsible
for
writing
prescriptions.
A
They
also
work
for
a
one
or
more
clinics
or
hospitals,
because
you
know
rarely
ever
do
you
see
a
doctor
who's
not
affiliated
or
working
with
more
than
one
clinic
or
hospital,
so
that
also
plays
a
big
part
into
figuring
out
how
we
should
go
ahead
and
engage
these
different
kinds
of
customers.
Although
the
drugs
are
being
sold
to
us,
but
we
are
the
indirect
consumers
of
that
from
a
sales
or
marketing
engagement.
A
Point
of
view,
the
prescribers,
the
people
responsible
for
writing
the
prescriptions
of
the
main
targets
in
this
whole
equation,
and
in
order
to
get
a
better
picture,
that
would
essentially
help
drive
a
lot
of
the
decision-making
on
the
sales
marketing
or
the
compliance
side.
Now
most
of
the
pharmaceutical
companies
are
collecting
data
about
these
prescribers
about
these
health
care
organizations
and
they
also
by
third-party
data
feeds.
A
So
there
are
third-party
data
vendors
that
provide
some
of
this
information,
but
none
of
these
sources
are
accurate
in
the
sense
that
there's
always
a
gap
in
terms
of
what
kind
of
information
is
available,
because
this
is
not
just
the
simple
name
and
address
type
of
demographic
information.
It
also
includes
which.
A
What
what
states
are
these
prescribers
licensed
to
practice
in
it
has
to
be
validated
information,
confirmed
information,
because
in
a
lot
of
these
cases,
when
drugs
are
being
dropped
as
samples,
you
cannot
give
a
sample
to
a
doctor
who's
not
licensed,
or
anybody
who's
not
licensed
to
be
able
to
handle
that
kind
of
a
drug.
So
in
any
case,
most
of
these
companies
by
data
from
third-party
data
vendors
and
on
an
average
they.
A
Typically,
when
you
look
at
small
to
large
pharmaceutical
enterprises,
they
typically
have
12
or
more
of
their
own
internal
sources,
which
are
multiple
sales,
CRM
type
of
applications,
fulfillment
applications,
any
other
type
of
essay,
p
type
of
applications.
You
know
they
have
data
about
these
types
of
entities
in
there
and
they
try
to
consolidate
and
combine
that
to
get
better
information
across
their
enterprise,
but
at
the
same
time
they
also
have
a
dozen
or
more
different
data
sources
that
they
buy
on
a
periodic
basis.
So
some
of
these
sources
provide
information
on
a
weekly
basis.
A
A
A
So
if
you
go
to
your
general
physician
and
you
get
referred
to,
you
know
a
specialist,
that's
referral
information,
that's
again
flowing
back
into
these
data
collection
type
of
systems,
and
this
is
where
we
are
helping
them
understand
the
overlap
and
overlay
of
this
information
across
the
different
data
sources
that
they
are
combining
together.
But
on
top
of
that,
what
it
helps
them
drive
is
to
measure
account
profitability,
so
understanding
what
are
the
sales
that
are
currently
being
driven
and
what
is
the
potential
size
of
the
account?
A
What
can
be
the
potential
revenue
that
can
be
generated
from
these
accounts
or
understanding
any
kind
of
price
likud,
because
when
you
look
at
Kaiser
or
Sutter,
these
are
large
organizations
or
some
of
the
insurance
companies.
These
are
large
organizations
and
having
multiple
contracts
can
lead
to
price
leakage
type
of
scenarios.
So
you
want
to
have
a
consistent
set
of
pricing
policies
or
discounts
or
debates
that
are
being
offered
across
the
board
to
these
large
organizations
and
any
mismanagement
of
that
leads
to.
A
So,
instead
of
trying
to
go
down
that
path,
which
was
more
of
a
one-to-one
sales
model,
now
it's
based
on
the
value
of
the
account
and
what
kind
of
what
kind
of
revenues
can
be
generated
out
of
it
and
what's
the
level
of
investment
they're
willing
to
make
from
the
point
of
view
of
if
this
is
a
large
account,
do
we
need
team
that
maintains
this
account,
or
if
this
is
a
large
account,
we
haven't
done
much
business.
Our
competitors
are
in
there
and
they're
doing
the
volume
of
business.
How
do
we
push
them
out?
A
Organizations
there
is
a
legal
entity,
type
of
structure
or
a
commercial
contracting
type
of
structure
that
needs
to
be
maintained
and
defined.
This
is
where
the
structure,
trees
and
hierarchies
come
into
play
and
then
essentially
creating
these
information
maps
that
allow
us
to
traverse
all
of
the
information
and
create
insights.
A
That
would
essentially
allow
us
to
understand
account
opportunity
from
hi
to
a
low
scale
or
the
degree
of
complexity
of
certain
accounts
and
do
the
distribution
so
that
now,
with
this
information,
we
can
decide
what
kind
and
what
size
of
a
team
is
going
to
focus
on
those
accounts
vs..
These
are
low
value
accounts
and
how
much
effort
are
we
going
to
put
into
these,
and
that's
that's
the
kind
of
information
that
we
are
driving
with
this
and,
at
the
same
time
as
we
are
collating
information
from
these
different
sources?
A
Unlike
a
traditional
analytic
system,
we
are
using
the
Cassandra
capabilities
to
also
open
it
up
as
an
application.
This
complete
set
of
data
that
can
be
further
curated
and
the
field
itself
are
the
sales
reps
in
the
field,
so
every
company
has
about
you
know.
Let's
say
when
you
look
at
Eli,
Lilly
or
mark
or
some
of
the
large
organizations
they
have
about
five
to
ten
thousand
sales
reps
in
the
field
and
in
those
cases,
any
information
that
they
can
bring
in
and
help
curate.
B
A
Cassandra,
so
maybe
like
everybody
else,
went
down
the
traditional
path
tried
using
article
defining
star,
schemas
and
bringing
together
some
of
this
information.
In
the
past
we
come.
The
team
at
relative
comes
from
an
area
of
work
known
as
master
data
management,
where
it
focuses
on
essentially
combining
some
of
the
dimensional
or
the
what
you
would
call
as
the
differential
data
sets,
which
are
typically
demographic
or
formal,
graphic
or
product
type
of
information
entity
type
of
information,
but
in
case
of
the
type
of
problems
that
we
were
trying
to
solve.
A
Now
we
had
to
combine
that
dimensional
data
with
all
of
the
fact
data
and
look
at
the
analysis
across
the
board
and
in
that
case,
trying
to
support
both
the
master
data
management
capabilities
which
require
matching
disparate
data.
Data
sets
merging
them
together,
maintaining
crosswalks
for
the
different
sources
that
provide
different
types
of
information,
as
well
as
maintaining
all
of
the
history
and
lineage,
because
from
a
compliance
standpoint,
you
need
complete
visibility
into.
When
was
a
change
made.
Who
made
the
change?
Where
did
the
information
come
from?
A
What
were
the
contributing
sources
for
a
specific
attribute?
All
of
that
lineage
is
something
that
we
needed
to
maintain,
which
increased
the
total
volume
of
information
that
would
build
up
over
a
period
of
time,
but
on
top
of
that,
combining
it
with
claims,
data,
sales,
prescription,
sales,
type
of
information
or
any
of
the
transactions.
That's
where
we
ran
into
challenges,
because,
first
of
all,
it
was
desperate
type
of
Records
with
completely
different
fields
on
them
and
we
needed
a
flexible
solution
for
it
in
order
to
move
forward.
A
Maintaining
crosswalks
are
doing
dimensional,
roll-ups
and
aggregations,
and
in
terms
of
how
we've
you
know
deployed
it
so
a
typical
model.
We
have
the
cassandra
faster,
but
we
have
a
separate
cluster
for
elastic
search
which
takes
the
real-time
information,
anything
that
is
flowing
in
and
then
essentially
any
of
the
API
or
the
UI
capabilities
that
are
distributed
and
again
deployed
in
a
cast
cluster
and
that's
our
deployment
model
for
what
we've
built
out
using
some
of
the
cassandra
capabilities.