Welcome to Clean Data
is Now Essential.
Thank you for joining
me today, this afternoon
I trust all of
you are having
an amazing first
day, so thank you
for spending 40
minutes with me.
I also want to give
a shout-out to--
we have so much staff here
that's been helping us.
Everybody in
a yellow shirt
has been extremely
friendly and helpful.
So an extra smile or
thank you goes a long way
They've been
helpful for me
in setting up my
presentation today as well
as our sound and
recording crew over there.
Forward-looking statement.
Quick reminder
that Salesforce
is a publicly traded
company and customers
should base their
purchasing decisions
on products and
services that
I know we've heard some
exciting announcements.
So I'm super geeked
about some of the things
that were
announced earlier
But, again, base
your decisions
on things that are
currently available.
My name is Lizz Hellinga.
I am also the Chicago
Salesforce WIT Leader.
Even though I live
in Holland, Michigan,
Chicago is my closest
nexus of community groups.
I'm also a Salesforce
Hall of Fame MVP,
and I really like
to talk about data.
And that's what we're
here to talk about today.
So finally, data is
like front and center
because of all the
cool things that
Our agenda today
is we're going
to share some strategies
and tactics that
will help you lead your
organization's CRM data
We'll build out a little
bit around the why.
We'll talk through
some strategies.
And, of course, it
wouldn't be a Salesforce
presentation
without some tactics
that you can use within
the platform to help
At the end of
this presentation,
I will share a
QR code for some
of the resources
that I'll reference,
as well as some articles
that I'll reference too
So don't stress
about taking pictures
unless you truly want to.
It helps your
note-taking process.
So last year, McKinsey,
around the same time
as Dreamforce last
year, came out
with an article around
generative AI that I read.
And this quote has
been this driving force
behind a lot that I've
done over the last year
It says, "If your data
isn't ready for generative
AI, your business
isn't ready."
So if your
organization's data
is incomplete,
outdated, inconsistent,
that means that AI
could potentially
be unreliable, misleading,
or even harmful
to some of your
organizations.
So having your
data ready for AI
will help you be
better positioned
to innovate
faster, to adapt
to some of these changing
market conditions,
as well as, most
importantly,
we can use AI to
really personalize
So clean data is
crucial for all of us
in building trust
in our AI outputs.
Let's take a look at some
of the different types
I'm curious to see how
many of these types
How often do we talk
about duplicate data?
I think I hear about it
almost minimum of at least
But what is
duplicate data?
Duplicate data
means there could
be multiple records
of Lizz Hellinga
It could be a contact
record, account record,
a lead record, but there
could be multiple of me
And you can determine
if that's good or not.
But duplicate data is
the number one thing
The second is
inaccurate data.
So let's say you
have a contact record
with an address, but
perhaps the zip code
We all have some of
those typing challenges.
And I know for
sure I've entered
in the four-digit
extension
incorrectly on a couple
of records in my lifetime.
So you can have
inaccurate data.
So are you missing
key data points
that allow your
organization
to be productive from
a specific record?
So for example,
let's say you
need to know if a
person in a household
is the designated
emergency contact.
Or perhaps you
need to know
if an individual
holds the insurance
There's a lot of reasons
why you could have
incomplete data, but
having a checkbox
may be that you need
to have it filled
in but it's not could
lead to some challenges
So incomplete
data is something
that usually is determined
by your specific org
The other kind of
bad data that we have
is stale data, so
data that's older,
maybe that hasn't
been touched
It really depends on
your organization.
So some
organizations may say
a lead record
that's a year old
is fine, whereas
another organization may
say that a lead record
that is 90 days old
But again, it depends
on your specific
And then this is
my favorite one,
I've recently
learned about this
over the last year
or two as I've
been going on this
deeper dive into data
and what people
do with it.
But it's data that people
just won't get rid of.
So I once had
an organization,
they had their Salesforce
instance for 15 years.
I mean, how many of
you had instances that
They had 500,000 accounts.
Do you know how many
had actually been
touched in the last year?
So they had 500,000
account records.
And it honestly took six
months of discussions
with the CEO to be able to
remove a portion of those.
So we weren't able
to remove all 5,000
But it was data
that just wasn't
touched within their org.
They just had
it, some really
they had had since day one
in their Salesforce org.
So how many of you
have at least one
of these in your
Salesforce org today?
Two, three, four,
five, all five.
So what my main point is,
though, is that bad data,
it's really what
you determine
within your organization.
So it's what you and
your key stakeholders
So, yes, we know like
duplicate data is bad.
We don't want to have
all these duplicates
But when you think
about incomplete data,
like incomplete
data in my org
could mean something
completely different
So, again, it's gathering
your stakeholders together
to start to have
those discussions.
And I'll share a few
strategies to help you
because sometimes it
takes a little bit
to get those stakeholders
on that journey with you.
We all know stakeholder
management is probably
the one thing with our CRM
that we didn't realized
that we needed to
do, but we actually
have to do almost
all the time to have
It all starts
with process.
And I am not like
this superstar
at creating process flow
maps or Visios or Miro
boards, but you have
to really document
It can be in a Google
Doc or some document
where you just write
step 1, step 2.
But you have to understand
what processes Salesforce
is supporting within
your org in order for you
to be able to answer these
questions around what
type of data
you have, what
is the quality of the
data that you have.
If you don't understand
your processes,
you won't understand which
data points are essential.
So from a high
level, when you
think about, let's say,
like an application
process for a student
within a university, there
are certain data
points that you
need at the beginning
of the process.
And there are
certain data points
that you gather
throughout that process
until maybe they apply
and they get accepted
But you have to know
what that journey looks
like before you can
determine what that is.
Because we've all been
there where somebody says,
well, whenever they create
that account record,
I want them to add all
45 required fields.
And we can't have that
because data is a journey.
And so understanding
your processes
So the second point, which
I kind of highlighted,
is identify those
essential data points.
So if you're mapping
something out,
and I have an example
on the next slide,
start to list out what
they need at each stage.
So thinking about
maybe the lead process
or the sales
process, start
to map out what points
are important for you
to be able to progress
to that next stage,
or even with your
case, [? lifecycle ?]
And then lastly, you
have to understand
the who, what, where,
and why and when as well.
So who is the data
about, and who
within that process
needs to see it?
So sometimes in
a sales process,
when a deal is
[? closed-won, ?] finance
Maybe delivery needs
to be notified.
Different people need to
get different data points
And what are
the data points?
What needs to
be sent when?
What needs to be reviewed?
And then obviously,
where, where in the cycle
is those data
points important?
So let's look at
a sample process
to talk through this
example a little bit more.
So we're going to use the
simplified lead process.
It's going to be a
business development, BDR,
So when that lead record
comes into Salesforce,
we don't need that
much about them.
So we're really
looking at, oh,
give me the first name,
last name, company,
maybe an email or a
phone, and then they can--
BDR probably can take
that data and run with it
and start to put them
through the lead cycle.
And again, this is a very
simplified-- like I said,
I'm not great with flow
charts or any of these.
So this is just using the
slide where available to
me to map out the process.
You can even start
it on paper too.
So once the BDR contacts
that lead and starts
to qualify them, there's
additional data points
Are they the right person?
Is there an actual
need for your product
What is the timeline
for purchasing?
And what are the
products that they're
They don't need that to
be a record initially,
but once they
start getting
to the
qualification stage,
they need all
that data in order
to be able to take it
to that next step, which
is then to pass it
to the salesperson.
So what does the
salesperson want?
The salesperson will
want some additional
Maybe they're collecting
who the decision makers
are or who are the
key influencers
What is the next
meeting date?
This is just a pretty
simplified example
of how you can chart
out the different data
points that you need
within your process.
And sometimes
this is all you
need to start driving
those discussions
Because if you
do this and you
start to have
conversations,
let's say, with the VP of
sales or with marketing,
oh, they're going to
have some opinions on it.
And so that'll
be where you
start to negotiate amongst
your stakeholder groups
on what data is
actually required.
And that would be
another session.
So the second
strategy to consider
So all data has
a lifecycle.
It starts with
the creation
And as Salesforce
professionals,
we have so many
tools available
within Salesforce to
help us gather data
from our end users in
a clean and simple,
well, sometimes a
little bit more steps
We'll share some
of those tactics
Data needs to
be maintained,
so you have to
think through what
are your maintenance
processes?
How are you ensuring that
the data stays accurate?
How do you ensure
that it stays relevant
and potentially compliant
if you are a regulated
So these are all
things that you
have to consider from a
maintenance perspective.
And then you also have
processing of data.
So clean, structured
data allows
for efficient processing,
so it reduces errors.
We've all been
in a situation
where maybe we're on
the phone with a vendor,
and they don't have
all of our information
or they have
incomplete information.
You're like, OK, now I've
got to update my address.
And so having clean,
structured data
helps smooth those
customer service request
processes, or
sometimes it's
even when you're changing
hotels or things like that
with your hotel provider.
And then you have to
have the management
So this is sort of
that undercurrent.
So how do you ensure that
the data is organized?
Since we're all
Salesforce professionals,
there's a lot of
things that we
can do from an
architecture perspective.
We also want to make sure
that it's accessible.
That's another
key thing that--
Who can't access it, as
well as we want to ensure
And we have so many
tools within Salesforce
from like masking
fields to using profiles
and permission [INAUDIBLE]
to ensure that data
Analysis and
visualization.
So a lot of times when
we're talking about data,
people are talking
about that output
because we want to be
like-- we want clean data
so that we can have
strong analysis
And we also want
to make sure
that we can use it for AI.
And so having accurate,
clear visualization
is truly dependent
on having clean data.
But it can also help
you manage data.
And we'll talk about
that in the next section
And then lastly, to my
hoarding point earlier,
you need to have a
strategy for archiving
Like maybe you don't
need to destroy all
of your data, but you
do need to think about,
how do you remove data
from the system that is
stale or old, incomplete
because it can get costly
maintaining 500,000
records account records
Or maybe sometimes if
you have connected tools,
it can be based on the
number of maybe contact
records or lead records
you have in the system.
So having a method and a
structure for archiving
data and removing it from
Salesforce is important.
And there's lots
of backup tools
that you can use for
some of that too.
So have to think through
the data lifecycle.
So you have a
couple strategies
to use, but
then from there,
you have to build
your data plan.
And the data plan really
is a standard operating
It's not something,
unfortunately,
that can be one and done.
Like this has
to become a part
of the lifecycle
of your CRM.
So we talked quite a
bit around stakeholders.
Stakeholders have
to be involved
And so those process maps
that you start to work on,
that will help
you identify
the key stakeholders
that need
to join your
discussions around data.
So it could be anything
from, let's say,
We think about
the sales process.
Like right off the bat,
we know from a sales
perspective,
you're probably
going to have to talk
with the VP of sales,
probably going to have
to have marketing,
may have to have
someone from finance.
You may have to have
customer success
because they're getting
handed off that.
But the process
map will show you
who needs to be involved.
And then from there,
you can always
ask if additional people
need to be looped in.
From there once you
start gathering them--
and this is kind of
like an informal data
governance, I
will say that.
I just shy away from using
governance as a term.
But once you get
your group together,
you have to determine what
your goals and objectives
In large orgs,
even in small orgs,
it can feel like
a giant elephant
that you have to
tackle in order
to be able to get through
this data journey.
I think you should
pick one process,
and it could be
something simple
like the lead
process to identify
which key data points are
essential for that process
to be successful
for your company.
So when you're
thinking about some
of those key
initiatives, maybe
tackle one of
those processes.
For example, I was working
within a governance group,
We have one form
on the website
that the clients
have to complete.
And because they're a
regulated organization
and the data has to go to
the federal government,
there are probably like
50 data points that
And so we went
through that form.
And it was a
tedious process,
but it was our
intake process.
And we went through
step by step
as a group to figure out
what needed to be there.
That honestly took
us like three months
to do because we only
had biweekly meetings.
But figuring out what
your key initiatives are
will help you start
to wrangle the data.
And then you
have to consider
on top of that,
what are some
of the regular activities
that you as a Salesforce
admin or Salesforce
product owner
need to think through
around your data
How do you maintain data?
Are you following
up with things?
There's a lot of
different activities
that you can do to ensure
that you're holding
the data to the
standards that you
and your stakeholders
have established.
And then lastly, like
measure and respond.
So just like any other
business function,
you have to be able
to measure it in order
to see its effectiveness.
So maybe there's
a baseline
right now where you're
looking at records
and you're saying,
OK, we only
have 20% complete records
for our actual customers
And maybe your goal is
to get that to 40% or 50%
But working with that
stakeholder group,
you determine what
those objectives are
And they need to
be measurable.
So they need to be smart.
Once you have
that data plan,
your next action
to think through
is training and adoption.
So I can put all the
things in Salesforce
We can have all
the conversations,
but it doesn't
mean anything
if your organization
isn't trained
and isn't
adopting what you
need to do for clean
data within your org.
I think that training and
adoption or enablement
is the undercurrent
for building culture
And this is probably what
I spend most of my time
with in relation
to data governance
You have to weave
your data goals
What are you
hoping to achieve?
Why is the data important?
I would share and
educate high level
on the processes
during training.
So obviously,
you don't want
We don't want
people to be bored.
Or you want them
to fall asleep
during the explanation
of a process.
But you want
them to know what
key data points
about maybe
your client or
your customer
So thinking back to
that lead process,
if I was sharing it with
a new salesperson joining
the organization,
I would say, look,
to create a lead, you only
need five data points,
first name, last name,
company, email, or phone.
But as you take them
through the cycle
and they start
to qualify them,
you need to know
more about them.
And so educating
them on the journey
of the value of that
data will help them
have more context
around why
they need to ensure
that they're updating
So sharing client
specific needs, so
understanding the impact.
If you don't have certain
data points, what happens?
I mean, there
is a true impact
if you have incorrect data
about a current client
and they call in and
they're constantly
like having to re-explain
themselves or tell
the customer service
person something,
it may frustrate them,
and they may leave.
So there are real effects
of having poor data
in the system on
your client base.
And then also I would
create feedback channels
on data for
your end users.
So if there's
something that--
maybe you have a
ticketing system
or you have some sort of
office hours or check-in.
Give them a place to talk
through with you the data.
Because the more
involved they are,
the more they're willing
to bring data questions
to you, the more
you know that you're
starting to have an
impact on their adoption.
One of the things that I'm
constantly training people
Sometimes I feel
like that's--
I'm like, just
send me the case,
and I'll just
merge it for you
because we've
all been there
where it's like somebody
merged a record.
And thank you for
recycle bin for 30 days
so that people can go back
and restore some of that.
But we want to
empower people.
We want to have
them manage data.
But then we also
want to protect it.
So it's kind of weighing
all those options.
But training and
adoption and enablement
is probably your best path
to building that culture
and respect around the
data within your org.
So we've talked a lot
about some strategies
and the why,
but let's share
some tactical Salesforce
ways to help support
The first one is
reports and dashboards.
So without clean data, we
can't have great reports
But we have to
inspect the data.
And so I usually will
build reports just
So I'm talking
with someone
and mapping out
that process.
And they say, oh yeah,
all of our qualified leads
And I build a
report, and I show,
oh, look, you
have probably 70%
It helps to have
that discussion
around what can we do to
improve that data input.
How can we get
the people that
are entering the data to
fill in that missing data?
I typically will use
the last activity
or last modified to
help me figure out
But again, based
on the strategies
we discussed
before, you have
to work with your
stakeholders to figure out
Now, it's not always easy
to get your stakeholders
And I also think reports
and dashboards can
So if you're struggling
with somebody,
I recommend pulling
some stale data reports
to be like, well, look,
your team isn't really
managing their
opportunities on time
or they're not updating
their key target accounts
or what you deem as
team target accounts.
Reports and dashboards
will help drive
some of those
conversations
if you have
individuals that
are reluctant to get on
that clean data journey.
When possible,
when you get
further along
in your journey
So people do
enjoy seeing--
in some orgs, I've
built out a scorecard
I don't use it for every
like record type or record
But typically,
accounts seem
to be really
important to people.
So if you can build
out a rating system,
it can help the end user
see, oh, well, my data
quality score is like
a 5, and it needs
They may be competitive
and want to fill it in.
I will sparingly add
some of these reports
or dashboards to
the home page.
I don't want to
clutter the home page.
I want it to
be actionable.
So usually, if I'm
putting something
on the home page,
it's to alert someone
like, hey, you need to
take action on this now.
So typically,
we've all used it
a lot with opportunities.
That is a main place
where maybe you'll
see no activity
on an opportunity
within the last 30 days
because we want our sales
You also see it
with cases too,
and that's a
shortened timeline.
How many of you have
used Salesforce Labs?
So they have some
great resources.
And one of my favorites
is to Clean Your Room
So you can install it in
one of your sandboxes.
I link to it
[? at ?] the QR code
at the end of
this session.
But it's a fun one
to play around with.
Obviously, you
have to modify it,
but it's a
springboard that you
can use within your org
to shorten your time you
It's a great
time-saving dashboard
that gives you
some good insights
into the quality of
your data in the system.
So reports and
dashboards always
the first place that
I start whenever
The second thing
that we can do
is the ease of data entry.
Uh, we have to always have
a mind towards UX, User
Experience, when we're
building and designing
And there's so many tools
that we can take advantage
The number one tool that
I love, that I see that--
I'm like, I can't
believe people are not
How many of you are
familiar with Path?
OK, so good, like
a big chunk of you.
You can use it for
other processes
within your
Salesforce org.
And I use it quite
a bit to highlight
the key fields by stages
or the picklist values.
While you can't edit
the Highlights Panel,
you can't click
in it to edit it,
the Highlights Panel is
where key data points
like on account,
contact, case record
you can place up there for
people to see right away
and instantly know if
things are out of date
We can use dynamic forms
and screen flows also
to help with that
ease of data entry
And then the last one is
the email integration.
I think a lot of
people think about, oh,
what can I do from a
setup perspective on page
layouts or with
tools within it
but also email
integration?
So whether it's a Gmail
connector or your Outlook
connector, I find
that that is probably
one of the best
ways to get
executives
using Salesforce
So if you're
integrating it
with your Outlook
or G Suite,
they're more easily able
to add some of that data.
And you can use
quick actions
within that side panel
to help build that out
So I don't know
how many of you are
It's usually a
pretty quick win.
Salespeople love
it, but also, I
find it to be great
for executives
because we want them
to lead by example.
But sometimes they
can be some of our--
I'm not sure of the
right word to use,
but they don't
always adapt as well
So outside of ease
of data entry,
we also want to be able to
automate where possible.
It helps us improve
data quality.
So there's a lot of
opportunities to automate.
There's so many ways
that we can automate
But where do you need
to begin when you
You have to go back
to your process maps.
So looking at
your processes,
determine where can I
automate, where can I
use Salesforce
to improve this?
I see a lot with
people sending
data between different
Salesforce objects.
I have been in orgs where
they'll have people add
And they'll have the
same field on the account
object, but it won't
map, it won't update,
and it creates a
frustrating user
And then the individuals,
the end users,
don't know what to trust.
So using automation to
help you update some
of those fields
across objects
will help improve
trust in the system
and the quality
of the data.
We all have
email, and we all
have a lot of email
notifications.
And I've had a lot of
sales leaders come to me
and say, I want
you to email them
when the opportunity is 30
days stale, 60 days stale,
and old or hasn't
been touched
And it's so
easy to ignore.
It's so easy to ignore
those email notifications.
So I recommend using them
sparingly, using them
So for example,
usually, when
I'm talking with a
sales leader that
wants me to email the heck
out of their sales team,
I say, let's
talk through what
is most important to
your organization.
It's usually like those
late-stage opportunities
that maybe are over a
certain dollar value.
And I ask them
if, hey, can we
I don't want to
send them an email
every time there's a stale
opportunity over 30 days
because it's just not
productive for us.
And it creates a
frustrating experience.
So use the email
notification sparingly.
A couple of other
things to use sparingly
We know that we
need to have them.
I would just go with the
minimum required based
on your analysis
of the process
and your discussions
with your stakeholders.
Where possible, please
group them together.
It is so hard when you're
trying to fill in a record
or update a record
and you're like, oh, I
got a required field here,
a required field here,
It just creates a
frustrating user process.
So I recommend trying to
group them where you can.
And then if possible,
optimize the Create Record
And, again,
include the minimum
that you need
in order just
So when you think
back to my BDR example
with the lead, they
only needed five things
for that, just to
create that record.
Maybe to move down
through the process,
they needed additional,
but they just needed five
There will be
people that will
want you to add
like 20 or 30 fields
when they create
an object.
And I would say
use the process map
to show them where maybe
those data points need
to be and where
they can be gathered
The next one to
use sparingly
We try to not have
too many on an object,
but they can become a
little bit frustrating,
especially when the end
user hasn't been trained
on the process
and doesn't know
to anticipate or
expect it when they're
So when you
think about the--
a lot of times we'll
have to add a key contact
or somebody on
an opportunity
at a certain stage,
like a finance person,
for example, or we have
to update the shipping
And we use validation
rules to enforce that when
they get to maybe proposal
stage or [? closed-won. ?]
Educate them through that
so that when they are
building it in, it becomes
just second nature.
They're already filling
out that information.
And they don't have to be
inundated with validation
rules when they try to
move to that next stage.
But when you do
have to use them,
please use clear
error messages.
Nothing is more
frustrating than getting
an error message
and trying
to figure out where is
this field on this page.
You want to end
all that scrolling.
So try to locate
those error messages
on the actual field when
you can because it creates
a frustrating process
when they don't.
But a lot of this you
can use these sparingly
if you take the
approach where
you embed a lot of
this into your training
I'll use even in-app
guidance prompts.
How many of you are using
prompts or walkthroughs
Those are a great way to--
You can prompt to
your heart's content,
but you can
pop up a prompt
on an actual
process and just
tell them like, oh, you're
going to make sure you
need to add these fields.
It can be like an
education journey
without stopping them
in the flow of work.
So use these as
sparingly as possible.
I mean, sometimes
I just want
to have a ton of required
fields or validation
rules, but it creates
a frustrating user
experience if
there's too many.
So we've covered
a lot today.
So we've talked through
some of the strategies
that you can use within
your organization
to get people on board, so
starting to build out that
why around your data,
thinking about how
you can better use clean
data within your group.
But your next
action, like what
you need to do if there's
one thing that you do when
you get back
from Dreamforce,
is map out at
least one process.
If you can get, I
think, up to three
like Miro boards
for free--
There's Visio,
whatever you have.
You can even use
Google Slides
and use the pictures
to map it out,
but start to map
out your processes.
And that will naturally
show you a lot of the data
that you need
throughout that journey
for your client or
for that end user.
And you can even
just write it out
I mean, sometimes
just start somewhere.
And then once you
start socializing it
with your
stakeholders, they'll
all get on board with the
discussion around data
because they'll
realize like, OK,
if we want to have
things at key points,
we have to make sure that
the system supports it.
And it'll help
get them on board.
It's just a URL
link to my website
I have the article
for McKinsey
around clean data is just
paramount for gen AI.
It's helpful especially
for your executives
I also have an
article around a lot
of alternatives to
validation rules
that's super helpful
for you for Salesforce
And also, please feel
free to reach out to me.
I'm happy to have
conversations
with you around
things that I've
done with clean
data or mainly
how to wrangle your
stakeholders to get
on board with clean data.
And I'm happy to take
any of your questions.
We have a few
minutes for Q&A.
Have about seven minutes.