April 6, 2023 • 11 mins
In this episode, Brick and Will, Solutions Architect at Blue Margin, discuss the four areas of questioning he uses when interviewing SMEs or stakeholders for building a report. It's important to understand the context first before choosing the right metrics and slicers to ensure that the report is actionable and meets stakeholders' needs.
Click here to watch this episode on our YouTube channel.
Blue Margin helps private equity owned and mid-market companies organize their data into dashboards to execute on strategy and create a culture of accountability. We call it The Dashboard Effect, the title of our book and podcast.
Visit Blue Margin's library of additional BI resources here.
For a free, downloadable copy of our book, The Dashboard Effect, click here, or buy a hardcopy or Kindle version on Amazon.
#dashboarddesign #datastrategy #BI
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Brick Thompson (00:04):
Welcome to The Dashboard Effect podcast. I'm
Brick Thompson and today I havewith me Will Trickett who is one
of Blue Margin's solutionsarchitects. How's it going,
Will?
Will Trickett (00:13):
Doing good, how are you, Brick?
Brick Thompson (00:14):
I'm good. So today we're going to be talking
about a topic that has come upin a lot of our previous
podcasts, sort of interspersedthroughout and I wanted to just
focus in on this topic, sincethis is something you do and are
very well aware of. That is, howdo you interview an SME or
(00:34):
stakeholder in order to build areport that really meets their
needs? I'll preface it by sayingit seems like this would be
obvious, like how hard could itbe? Ask them what they need,
they'll tell you, you build thereport. As it turns out, you can
actually get tripped up if youdon't follow a methodical
approach and make sure you sortof check all the boxes. So with
(00:58):
that, why don't you starttelling me, how do you go at
this when you start a project?
Will Trickett (01:03):
So I guess to start it, we probably learned
the hard way, as an organizationby probably doing it wrong a few
times, and then just learningsome best practices. We like to
focus on kind of four particularquestions that really drive
those conversations. I feel likeby the end of those four, you
can generally have a really goodsense of what is going to be
(01:24):
valuable. Or are we justbuilding something to build it?
Brick Thompson (01:27):
Okay, so It's like four areas of questioning,
not just four questions, right?
So what's the first area?
Will Trickett (01:34):
Yeah, so the first one we asked about is just
the business goal of the report.
The reason we start with that isI think a lot of times we get
stakeholders that come and theysay, "here's what we want built.
Here's all of the metrics andslices", and they just hand it
to you to build, What we find isreally helpful is to just zoom
out first and say, "okay, whyare we tackling this effort in
(01:54):
the first place? Help meunderstand what the the return
on investment is that's intendedfor This report? Are we trying
to save someone time for manualreporting? We're trying to
generate more revenue, becausewe have increased visibility.
There's a lot of differentexamples there, but that's the
first thing we try to zoom inon.
Brick Thompson (02:14):
Okay. So it's really about the business goal.
What are we going to beaffecting, in the business by
having this visibility that thereport provides?
Will Trickett (02:24):
Exactly. Yeah.
One example just to help kind offrame that. We have a client
that we've worked with for anumber of years. They do medical
imaging, and they have kind ofthis referral process that comes
from doctors who send thempatients who need to get scans.
Then from that point, when theyreceive this order, there's 15
(02:45):
to 20 different stages that theperson goes through before they
actually get to recognizerevenue on that order. So
working with them, the firstthing we identified was that
conversion between thosedifferent stages was going to be
crucial. When we asked thisquestion, they said, "Well, for
(03:06):
every 1% in conversion from thestart to the finish of that
process, we generate anadditional million dollars in
revenue, just from bumping thatnumber up." So That helped us
frame the importance and goal.
We know that we're trying to getpeople from A to B, and that's
the key here.
Brick Thompson (03:26):
We know the goal is conversions. So the goal of
the report is to help withfiguring out how to increase the
conversion percent, because a 1%increase is a million dollars in
revenue. So after you'veidentified the business goal,
what's the next area you goafter?
Will Trickett (03:44):
So after the business goal, it's specifically
about the personas who are goingto access the report. A persona
is just a group of users thatkind of have the same maybe
level in the organization or asimilar job site and function.
So it just represents who'sgoing to be viewing this. That
(04:04):
helps us kind of identify withthe audience and what the
particular view is that personmight need.
Brick Thompson (04:11):
Okay. Have you ever had a case where you built
a report where you didn'tunderstand the personas and it
not go well.
Will Trickett (04:21):
Unfortunately, yes. I think what tends to
happen is, you might think youknow who the personas are. Then
sometimes there'll be new usersthat are introduced later. So
we've definitely had thatbefore. I can think of a couple
of cases where there's theinitial group we gathered the
requirements from and then afterwe're into the report build, we
(04:45):
find out that there's kind ofthis unknown shadow person who's
actually really a keystakeholder and is going to be
accessing this report. Theymight have very different ideas
of what needs to be viewed thanthe others.
Brick Thompson (04:57):
So you've built a report, present it and this
new persona that you didn't knowabout shows up and says "What?
This isn't gonna work."
Will Trickett (05:06):
Yeah, and "I need a whole other page for this."
Brick Thompson (05:09):
Hopefully not back at the drawing board, but
you need to do a bunch of workto get it right. So really
important to drill down. I'mguessing that you somehow need
to dig in and try to figure outif there is a hidden persona,
that maybe the stakeholder whoyou're interviewing is not aware
of?
Will Trickett (05:27):
Yeah, I think it's super helpful if you can
highlight the fact thatdifferent people need different
things. You need to really buildthings to a certain group of
people. Make sure that's clear.
Brick Thompson (05:40):
Do you always want to talk to that persona?
Will Trickett (05:43):
Absolutely. Yeah.
Brick Thompson (05:45):
Will you ever proceed without being able to
talk to that persona?
Will Trickett (05:48):
90% of the time, we would require it. Maybe in
certain cases, if it was clearwhat they needed. I think
generally speaking, you want toget on the phone with them, or
at least get their input viaemail or something.
Brick Thompson (05:56):
Because you don't want to have the telephone
game. The persona told theperson who's telling you. So
then what's the third area ofquestions?
Will Trickett (06:14):
Yeah. So once we know the goal and the user, then
we're asking, what questions areyou trying to answer with This
report? This is kind of the nextlayer down, we're really trying
to understand like, What are theknowledge gaps that they're
looking to fill with the report?
Maybe there's just not goodvisibility into a certain area.
(06:36):
I think back to that conversionsexample, one thing there is, you
know, a lot of differentpatients had demographic
characteristics that mightsuggest that they would be
easier or harder to convert.
Then also age, gender, location,things like that. So we wanted
(06:58):
to really identify were theircharacteristics that we could
find that would help them targetthose populations to keep
converting through the funnel.
Secondly, within the differentstages of the funnel, there
might be one particular areawhere it took a long time for
them to convert to the nextstage. So another piece of that
was, is there an associationbetween the length of time
(07:21):
something is stuck in a bucket,and whether or not they convert?
When you kind of start tounderstand those questions, and
you can design a report thatanswers those effectively,
Brick Thompson (07:33):
So you want to make sure that it answers
questions, it's going to allowthe user the report, then to go
and do something about it. SoI'm assuming in those examples,
that would allow them to finetune their funnel management
processes. Figure out ways tohelp certain demographics, get
(07:56):
this diagnostic test that thedoctor ordered, or figure out if
there's a way that you can makethe process of getting a persona
go smoother. That way you don'tget stuck in one of these
buckets of the whole process?
Will Trickett (08:09):
I think that's the other side of the coin is,
like you said, what action doyou take with answering that
question, right? If you get someof that, as you understand what
they're trying to try to answer.
Brick Thompson (08:21):
Then I'm guessing you also want to make
sure this is contributing to theoriginal goal that we talked
about. I mean, the exampleyou're giving, obviously it
does, but there might be caseswhere it's not quite so obvious,
where you can sort of get lostin the okay, we're answering all
these questions. Was it reallygetting us to that goal?
Will Trickett (08:39):
This would help weed that out to make sure that
you're, you're delivering valuewith what you end up sending to
end users?
Brick Thompson (08:47):
So you said there were four areas of
questions. What was the fourthone?
Will Trickett (08:51):
Yeah. So the fourth one is like the lowest
grain, where you're asking them"what are the particular metrics
and slices of data that you needto support answering those
questions." That's sort of wheremost people start, is what are
the things That you need on thisreport? We like to keep that
last. That way, we understandall that context, and then we
(09:11):
can say, based on these fewquestions we're trying to
answer, here's the things thatwe need to solve that. That's
what we use to inform ourdesign. It's just an inverse way
of thinking, yeah.
Brick Thompson (09:22):
You might think you want to start with what
metrics do you want, what arethe measures that you need to
see you get to those- butunderstanding all that context
before you get to that is reallyimportant to making sure that
you're coming up with the rightmetrics.
Will Trickett (09:36):
Exactly. Kind of as a side note, when you're
taking the time to design areport, I think most most
developers have reached a pointwhere they're, they're just
stuck there and think they knowwhat they want to build, but
there's a blockage, you justdon't understand like exactly
what the user needs. If you'veanswered those first three
questions, I think it helps youin those moments, say I actually
(09:57):
need to focus on this more orthis more versus just kind of
being unclear on what the goalis.
Brick Thompson (10:03):
So metrics are obvious- the numbers, the
measures, that type of thing.
Slices is really just how you'relooking at the data in subsets,
right? So, in your earlierexample, you want to look at
data by certain demographiccharacteristics, or maybe you
want to look at the data bygeographical regions or that
(10:24):
type of thing.
Will Trickett (10:27):
Exactly. That's a perfect example.
Brick Thompson (10:31):
Those all make sense. So I understand the four
areas. It sounds like it mightbe a lot of work to get through
those. What's your typicalamount of time? Let's say you're
the report engineer working onthis- how long does it take you
to go through this process?
Typically?
Will Trickett (10:46):
Yeah, I think 45 minutes is usually what it would
take.
Brick Thompson (10:52):
Oh so this isn't gonna take hours and hours. It's
also not, "hey, you can send mean email telling me what you
want and I'll get it done."
Will Trickett (11:01):
Right. I think to do it well, you need you need
the time. It shows if thestakeholders are invested in
making sure that the report isgoing to be really actionable,
too. Yeah, 45 minutes, you canrun through those questions and
get a pretty good sense of whatyou need to build.
Brick Thompson (11:17):
Okay, great. I know there's a lot that comes
after this- doing wireframingand building a first draft and
having to get the data tovalidate and all that stuff, but
we'll leave that for anotherday. Appreciate your time.
Thanks!
Welcome to The Dashboard Effect podcast. I'm
Brick Thompson and today I havewith me Will Trickett who is one
of Blue Margin's solutionsarchitects. How's it going,
Will?
Will Trickett (00:13):
Doing good, how are you, Brick?
Brick Thompson (00:14):
I'm good. So today we're going to be talking
about a topic that has come upin a lot of our previous
podcasts, sort of interspersedthroughout and I wanted to just
focus in on this topic, sincethis is something you do and are
very well aware of. That is, howdo you interview an SME or
(00:34):
stakeholder in order to build areport that really meets their
needs? I'll preface it by sayingit seems like this would be
obvious, like how hard could itbe? Ask them what they need,
they'll tell you, you build thereport. As it turns out, you can
actually get tripped up if youdon't follow a methodical
approach and make sure you sortof check all the boxes. So with
(00:58):
that, why don't you starttelling me, how do you go at
this when you start a project?
Will Trickett (01:03):
So I guess to start it, we probably learned
the hard way, as an organizationby probably doing it wrong a few
times, and then just learningsome best practices. We like to
focus on kind of four particularquestions that really drive
those conversations. I feel likeby the end of those four, you
can generally have a really goodsense of what is going to be
(01:24):
valuable. Or are we justbuilding something to build it?
Brick Thompson (01:27):
Okay, so It's like four areas of questioning,
not just four questions, right?
So what's the first area?
Will Trickett (01:34):
Yeah, so the first one we asked about is just
the business goal of the report.
The reason we start with that isI think a lot of times we get
stakeholders that come and theysay, "here's what we want built.
Here's all of the metrics andslices", and they just hand it
to you to build, What we find isreally helpful is to just zoom
out first and say, "okay, whyare we tackling this effort in
(01:54):
the first place? Help meunderstand what the the return
on investment is that's intendedfor This report? Are we trying
to save someone time for manualreporting? We're trying to
generate more revenue, becausewe have increased visibility.
There's a lot of differentexamples there, but that's the
first thing we try to zoom inon.
Brick Thompson (02:14):
Okay. So it's really about the business goal.
What are we going to beaffecting, in the business by
having this visibility that thereport provides?
Will Trickett (02:24):
Exactly. Yeah.
One example just to help kind offrame that. We have a client
that we've worked with for anumber of years. They do medical
imaging, and they have kind ofthis referral process that comes
from doctors who send thempatients who need to get scans.
Then from that point, when theyreceive this order, there's 15
(02:45):
to 20 different stages that theperson goes through before they
actually get to recognizerevenue on that order. So
working with them, the firstthing we identified was that
conversion between thosedifferent stages was going to be
crucial. When we asked thisquestion, they said, "Well, for
(03:06):
every 1% in conversion from thestart to the finish of that
process, we generate anadditional million dollars in
revenue, just from bumping thatnumber up." So That helped us
frame the importance and goal.
We know that we're trying to getpeople from A to B, and that's
the key here.
Brick Thompson (03:26):
We know the goal is conversions. So the goal of
the report is to help withfiguring out how to increase the
conversion percent, because a 1%increase is a million dollars in
revenue. So after you'veidentified the business goal,
what's the next area you goafter?
Will Trickett (03:44):
So after the business goal, it's specifically
about the personas who are goingto access the report. A persona
is just a group of users thatkind of have the same maybe
level in the organization or asimilar job site and function.
So it just represents who'sgoing to be viewing this. That
(04:04):
helps us kind of identify withthe audience and what the
particular view is that personmight need.
Brick Thompson (04:11):
Okay. Have you ever had a case where you built
a report where you didn'tunderstand the personas and it
not go well.
Will Trickett (04:21):
Unfortunately, yes. I think what tends to
happen is, you might think youknow who the personas are. Then
sometimes there'll be new usersthat are introduced later. So
we've definitely had thatbefore. I can think of a couple
of cases where there's theinitial group we gathered the
requirements from and then afterwe're into the report build, we
(04:45):
find out that there's kind ofthis unknown shadow person who's
actually really a keystakeholder and is going to be
accessing this report. Theymight have very different ideas
of what needs to be viewed thanthe others.
Brick Thompson (04:57):
So you've built a report, present it and this
new persona that you didn't knowabout shows up and says "What?
This isn't gonna work."
Will Trickett (05:06):
Yeah, and "I need a whole other page for this."
Brick Thompson (05:09):
Hopefully not back at the drawing board, but
you need to do a bunch of workto get it right. So really
important to drill down. I'mguessing that you somehow need
to dig in and try to figure outif there is a hidden persona,
that maybe the stakeholder whoyou're interviewing is not aware
of?
Will Trickett (05:27):
Yeah, I think it's super helpful if you can
highlight the fact thatdifferent people need different
things. You need to really buildthings to a certain group of
people. Make sure that's clear.
Brick Thompson (05:40):
Do you always want to talk to that persona?
Will Trickett (05:43):
Absolutely. Yeah.
Brick Thompson (05:45):
Will you ever proceed without being able to
talk to that persona?
Will Trickett (05:48):
90% of the time, we would require it. Maybe in
certain cases, if it was clearwhat they needed. I think
generally speaking, you want toget on the phone with them, or
at least get their input viaemail or something.
Brick Thompson (05:56):
Because you don't want to have the telephone
game. The persona told theperson who's telling you. So
then what's the third area ofquestions?
Will Trickett (06:14):
Yeah. So once we know the goal and the user, then
we're asking, what questions areyou trying to answer with This
report? This is kind of the nextlayer down, we're really trying
to understand like, What are theknowledge gaps that they're
looking to fill with the report?
Maybe there's just not goodvisibility into a certain area.
(06:36):
I think back to that conversionsexample, one thing there is, you
know, a lot of differentpatients had demographic
characteristics that mightsuggest that they would be
easier or harder to convert.
Then also age, gender, location,things like that. So we wanted
(06:58):
to really identify were theircharacteristics that we could
find that would help them targetthose populations to keep
converting through the funnel.
Secondly, within the differentstages of the funnel, there
might be one particular areawhere it took a long time for
them to convert to the nextstage. So another piece of that
was, is there an associationbetween the length of time
(07:21):
something is stuck in a bucket,and whether or not they convert?
When you kind of start tounderstand those questions, and
you can design a report thatanswers those effectively,
Brick Thompson (07:33):
So you want to make sure that it answers
questions, it's going to allowthe user the report, then to go
and do something about it. SoI'm assuming in those examples,
that would allow them to finetune their funnel management
processes. Figure out ways tohelp certain demographics, get
(07:56):
this diagnostic test that thedoctor ordered, or figure out if
there's a way that you can makethe process of getting a persona
go smoother. That way you don'tget stuck in one of these
buckets of the whole process?
Will Trickett (08:09):
I think that's the other side of the coin is,
like you said, what action doyou take with answering that
question, right? If you get someof that, as you understand what
they're trying to try to answer.
Brick Thompson (08:21):
Then I'm guessing you also want to make
sure this is contributing to theoriginal goal that we talked
about. I mean, the exampleyou're giving, obviously it
does, but there might be caseswhere it's not quite so obvious,
where you can sort of get lostin the okay, we're answering all
these questions. Was it reallygetting us to that goal?
Will Trickett (08:39):
This would help weed that out to make sure that
you're, you're delivering valuewith what you end up sending to
end users?
Brick Thompson (08:47):
So you said there were four areas of
questions. What was the fourthone?
Will Trickett (08:51):
Yeah. So the fourth one is like the lowest
grain, where you're asking them"what are the particular metrics
and slices of data that you needto support answering those
questions." That's sort of wheremost people start, is what are
the things That you need on thisreport? We like to keep that
last. That way, we understandall that context, and then we
(09:11):
can say, based on these fewquestions we're trying to
answer, here's the things thatwe need to solve that. That's
what we use to inform ourdesign. It's just an inverse way
of thinking, yeah.
Brick Thompson (09:22):
You might think you want to start with what
metrics do you want, what arethe measures that you need to
see you get to those- butunderstanding all that context
before you get to that is reallyimportant to making sure that
you're coming up with the rightmetrics.
Will Trickett (09:36):
Exactly. Kind of as a side note, when you're
taking the time to design areport, I think most most
developers have reached a pointwhere they're, they're just
stuck there and think they knowwhat they want to build, but
there's a blockage, you justdon't understand like exactly
what the user needs. If you'veanswered those first three
questions, I think it helps youin those moments, say I actually
(09:57):
need to focus on this more orthis more versus just kind of
being unclear on what the goalis.
Brick Thompson (10:03):
So metrics are obvious- the numbers, the
measures, that type of thing.
Slices is really just how you'relooking at the data in subsets,
right? So, in your earlierexample, you want to look at
data by certain demographiccharacteristics, or maybe you
want to look at the data bygeographical regions or that
(10:24):
type of thing.
Will Trickett (10:27):
Exactly. That's a perfect example.
Brick Thompson (10:31):
Those all make sense. So I understand the four
areas. It sounds like it mightbe a lot of work to get through
those. What's your typicalamount of time? Let's say you're
the report engineer working onthis- how long does it take you
to go through this process?
Typically?
Will Trickett (10:46):
Yeah, I think 45 minutes is usually what it would
take.
Brick Thompson (10:52):
Oh so this isn't gonna take hours and hours. It's
also not, "hey, you can send mean email telling me what you
want and I'll get it done."
Will Trickett (11:01):
Right. I think to do it well, you need you need
the time. It shows if thestakeholders are invested in
making sure that the report isgoing to be really actionable,
too. Yeah, 45 minutes, you canrun through those questions and
get a pretty good sense of whatyou need to build.
Brick Thompson (11:17):
Okay, great. I know there's a lot that comes
after this- doing wireframingand building a first draft and
having to get the data tovalidate and all that stuff, but
we'll leave that for anotherday. Appreciate your time.
Thanks!
Popular Podcasts
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com
24/7 News: The Latest
The latest news in 4 minutes updated every hour, every day.
Therapy Gecko
An unlicensed lizard psychologist travels the universe talking to strangers about absolutely nothing. TO CALL THE GECKO: follow me on https://www.twitch.tv/lyleforever to get a notification for when I am taking calls. I am usually live Mondays, Wednesdays, and Fridays but lately a lot of other times too. I am a gecko.