Developing for "Somebody": How to Truly Know Your Users

hi everybody welcome to designing for

somebody thank you for getting up this

early to come to this Saturday morning

session and thank you to the ultimate

camp planning committee for putting on

this event my name is Kylie orc I was

your house key um and I'm not a typical

mid camp attendee let alone a typical

Smith camp presenter before I tell you

why let me remind you of our keynotes

first principle for inclusion yesterday

don't ever judge anybody speaking out of

your boo conference I have never worked

with Drupal before full disclosure and I

never really knew much about Drupal

before attending some sessions on it

yesterday so why the hell am I here um I

am a graduate student at the University

of Michigan School of Information

currently in an agile project management

agile project management methodologies

cars taught by Michael Hess who is on

the triple security team and towards the

beginning of the semester he encouraged

our class to submit proposals to mid

camp based on whatever sort of value we

thought we could provide to conference

attendees so I figured you know I'll try

and a few weeks maybe months later I'm

here and I'm here to talk to you guys

about something that's a little

different than I think is par for the

course for this conference but something

that I hope will nonetheless be valuable

to you and that is a high-level overview

of the processes of contextual design

and contextual inquiry so some

groundwork for the next however long I

talked for this section will cover why

you shouldn't

system for just anybody or for everybody

it will cover why you shouldn't design

develop or implement a system without

knowing the systems or your users and it

will cover how you can come to truly

know your users through contextual

design and inquiry I'll be drawing from

some of my own experiences in the

classroom and in my internships with

contextual inquiry just to provide some

sort of example that I can point to the

session won't cover anything really

about anything really technical and

anything explicitly related to Drupal um

so if that's not your speed I won't be

offended if you walk up now but just so

you know what you're getting into that's

what we'll be talking about today so

designing for somebody not just anybody

then not everybody what does that mean

so designs that claim to meet the needs

of anybody and designs that claim to

meet the needs of everybody are somewhat

similar a design for anybody would be a

design that any random person off this

tree were able to use this type of

design would necessarily be as

non-technical non-complicated

non-industry specific and not complex as

possible so as to guarantee that

literally anybody would find it

intuitive to use if such a thing exists

this design would necessarily be very

generic appealing to the lowest common

denominator of any potential users

background skills and experiences

similarly a design for everybody is a

design that every single person in this

room at this conference in Chicago

throughout the country would be able to

find a use for this type of design would

necessarily be very bloated with any and

all features that everybody could ever

want to use related to a particular

system or service um

accordingly neither of these user group

examples should ever be you

for a design development or

implementation process using these as

user groups for which to design for

developed for can only lead to like I

said very generic or very bloated

outcomes which no user will really find

and meets their specific needs

so if we're not for trying to not design

or develop or implement but just anybody

but for everybody

we want to develop for somebody we want

to create useful stuff obviously so we

want to develop our design or architect

or build a specific solution that meets

a specific need of a specific group of

users we want to design things for that

somebody and what that somebody does so

how do we go about doing that this is

where contextual design comes in

contextual design is one way to design

for somebody and it's the way that I'll

talk about this morning this is a user

centered design process developed by

Karin holes flat and Hugh buyer and it

incorporates ethnographic methods for

collecting data that's relevant to a

system or product or service being

developed or revised or improve

Martin and Hanson have defined

contextual design as a customer centered

design process in which every step is

anchored in customer data making it feel

less like design magic so this is a very

data-driven design methodology and so

these are the process of contextual

design has you know at least on paper a

lot of these steps

there's many methods of data collection

intermediary steps of data synthesis and

analysis and then some research

deliverables that hopefully result in a

usable useful design product this is

just a brief overview of these steps

laid out by Holtz flat and buyer but we

really will be only focusing on the

first one within the scope of this

session somewhat touching on the two

that follow it because I think these are

the most the most valuable at the

beginning stages of a designer and

fermentation process so as I had

mentioned contextual inquiry will be the

focus of this presentation as it

pertains to the scope of contextual

design so before we get into this first

step of contextual design which is

contextual inquiry I want everybody to

think about what they do every day when

you come into work just in your heads

you don't have to talk or tell me or

anything like that

so accordingly and with that one of

Holt's Bloods and virus principles

underlying the contextual design process

is that which states people are experts

at what they do but unable to articulate

their own work practices so essentially

what this boils down to is this we know

what we do but not exactly how we do it

when you thought about what you did

every day it was likely it was at a

fairly mid-level range you didn't

consider your individual keystrokes that

you do to perform your job nor is it

likely that you considered how your work

fits into the larger organizations

functional workflow that is our work

processes and how we perform them I

really test it especially when we're

asked to consider or recount them when

we're not in our typical work

environment or mindset just like I asked

you to do so this begs the question of

how users developers or researchers are

supposed to actually learn how people

how their users do what they do and

that's through contextual inquiry

according to Martin and hanison again

contextual inquiry is an immersive

method of observing and interviewing

that reveals underlying and somewhat

invisible work structures according to

me contextual inquiry is really

observing somebody doing their job in

their actual workspace and asking them

questions about what they're doing this

is fieldwork at its core for my field at

least if you're able to watch somebody

do their work you'll be able to

understand and observe even if you have

to ask follow-up questions the tacit

parts involved in their task performance

beyond what they be able to articulate

to you in a decontextualized setting as

in a sterile interview or something

similar so this this is something I've

done in a couple different instances in

my coursework at the University of

Michigan it's something I've done with a

local Public Library and consulting

method

Jacquetta class semester and it's

something I'm currently doing on a

larger scale to support a CRM

implementation project with U of M's

office of enrollment management in those

cases when an organization has

identified a problem with their workflow

or when an organization is introducing

or last introduce a new to our system

contextual inquiries can be helpful in

exploring how work is actually being

done in an organization and

recommendations and requirements can

file from the data resulting from the

exploration so these are the key parts

of a contextual inquiry process first we

have context obviously we have to

understand users and their needs in the

actual context in which they do work

this is done by observing them do actual

work in the National workspaces not in a

conference room not at a conference just

in where they work we want to understand

how they do it in that particular

context another key part is partnership

and we have to work with users in this

method as far as partners throughout the

inquiry process a good piece of advice

I've heard for my professors is that you

as a contextual inquiry should situate

yourself as an apprentice to the users

expert in this situation ready and

waiting for the user for the expert to

guide us through their actual work

activities another key part is

interpretation and you similarly related

to the partnership aspect of contextual

inquiry we have to work with users to

come to a mutual interpretation a common

understanding of the work we're doing

because our insights are only so

valuable in that they're accurately

assessing what the user is actually

doing and then the last key part is a

focus contextual inquiry is in most

cases should not require like a

predefined set of questions that you go

through and answer the user it shouldn't

be a standard interview you should come

to a contextual interest session with a

clear project focus that is a clear idea

of what you want to explore the work

processes you

get more involved with and maybe a

protocol that lists the points that you

want to be sure you cover in your

observation session but that's all you

shouldn't use a script but like a

protocol really to guide you through the

process and so this protocol is really

um I guess I'm map you could consider it

for a successful a successful contextual

inquiry um and you can write this

protocol out as a guide for the first

few times you conduct a contextual

inquiry but it really will become second

nature after a while um the first key

part of this protocol is really the

introduction and this is important for

setting the stage for what's to come in

your contextual inquiry because your

users aren't really your your experts in

this case I are really going to

understand what it is that you're doing

why you're observing them do their work

so setting the stage for exactly what

you're doing and the purpose is that

that serves is important in this and

similarly you should introduce yourself

in the scope of your work so that they

know you know what they should be

working on when you're with them you

should set expectations for the length

of the the interview / observation

session which is usually about an hour I

found after anything longer than that

people start to lose focus

and then you should be sure here to set

up the user as the expert in that case

that situating yourself is just there to

learn from them and then during the

observation / interview the contextual

inquiry you should as I've said observe

the user's task because Anax ask

questions according to your research

focus and you can be nosey you know you

can say hey what are you doing here what

are you doing here we're there to learn

accordingly you should take notes and

audio record the conversation with their

consent you should collect any artifacts

you know documents they reference things

like that

at a minimum you should set out to learn

the specific things they do to complete

specific work processes their pain

points of how things are now in their

goals for their own work processes and

here it's super important to ask the

users talk about their actual work and

provide specific examples of their

processes reconstructing situations when

possible if they begin to talk extract

lead or summarize that information is

not really useful in - - contextual

inquiry because it's decontextualize

it's based on their memory and the

accuracy is really reliant on their

memory which is as we discussed not

infallible so we don't want to rely on

what they think they do but we want to

rely on what we're actually seeing them

do and this is hard to do this is hard

to not ask them to predict what they

would do give an X what they would do

with better tools what they would do

with better systems and so this is one

of the hardest parts I found of

contextual inquiries making them making

it a concrete conversation about what

they're actually doing and not straying

into summary or abstraction or

predictions I guess on their end and

then after this main bulk of your

interview observation you should try to

start the wrap-up in which you with the

user creating shared larger

interpretation of what you've observed

in what you've learned so that you're on

the same page about what just happened

next up actor contextual inquiry ideally

you would complete a series of

contextual inquiries because nobody

works in isolation that you would

conduct through the set of observers and

observation interviews and observations

with people performing similar jobs or

people working on a team and after you

have a series of these done he would

conduct interpretation sessions after

these conduct show up and create with

your team to go over your notes and then

observations to essentially interpret

your data capturing any key findings

this interpretation session what you

could call it when you do this you

should start by creating a user profile

which includes like demographic

information about the user's job and the

roles they fill and then in this case

you should assign them a code that

you'll use to refer to them throughout

the inquiry process in this case then

the person who conducted the contextual

inquiry should then walk the team

through their interview without

summarizing just by going through their

data points and a designated a

note-taker should be taking notes on the

walkthrough with other team members

asking questions throughout the walk

through any key issue interpretation

characteristic breakdown or insightful

quote on the end of the user should be

recorded as a stand alone affinity no

attributed to the code of the respected

user profile and if a fitting note is

just that a notable a data point in a

contextual inquiry essentially in a

general rule of thumb that I found is

that you should have between probably 50

contextual inquiry conducted so with

your contextual inquiry is done in

interpreted then the next step usually

is affinity diagramming and this

affinity diagram will bring together

issues and insights across all customers

into like a wall-sized hierarchical

diagram to reveal the scope of the

problem or the scope of the user tax

that you are observing so to do this

after you like I said after you have all

your you know however many hundreds of

affinity notes after a handful of the

enquiries it suggested like I said to

write each on either like a small sticky

note or use an online tool like mural

comb and then with your massive

collection of affinity notes it's time

to create an affinity diagram and if I

can show you a guess what that looks

like real

so this is an example of an affinity

wall that I had created for a different

research process but essentially what I

had done was I conducted a series of

contextual inquiries with users and with

my team we recorded each individual data

point on the yellow sticky notes and

then we group them under blue sticky

notes which summed up any key themes or

interpretations that we found and then

we just kept going higher until the

green sticky notes where we found kind

of the key problems or findings of our

research process and that was just based

on you know shared interpretation of

what the data meant and what the data

highlighted

so that seems like I said to go back to

this recommended steps for the textual

design so we've conducted that we focus

mostly in the contextual inquiry where

we talk to users and we observe what

they do and we talk briefly about the

interpretation sessions I haven't

those are pretty open-ended in terms of

the structure of them but essentially

like I said it's just a walkthrough of

what your data found and then we talked

briefly about the affinity diagrams

where in you you bring your data points

together on this it's been large actual

physical diagram but I based on my

experiences I think it's probably going

to be more common for that to be more

digital using two online tools to do

that because it is a pain to get all

those sticky notes together and go

through that process but so that is the

process that I have mostly been

following in my my own work I haven't

really been in a position where I can do

the visioning storyboarding user

environment design and paper mock-ups

but depending on your role in an

organization and what you're really

looking to get out of your research

process those steps can follow from

there as well

so with your affinity ater diagram

what's what's revealed as I had said was

a critical set of findings related to

how your users currently perform their

day-to-day tasks their pain points with

the current processes and their goals

for their processes in some cases like I

mentioned this may be enough we're

giving you a role this may be enough but

you you could in other cases this may

not be sufficient so you could go ahead

to learn more information about your

users using that process laid out that I

had mentioned by halted Empire these two

readings are things that I would

definitely suggest in there nope 9

coincidentally my work cited for this

presentation as I had mentioned halts

flat and buyer kind of pioneered this

this method of contextual design with

all the steps underneath it and this

Martin and handing ting text is a good

good primer for brief introductions into

various methods of design research

okay little short but any questions

research choices made in wireframe or

design or particular features like how

does this

okay yeah so um to go to my my

internship right now so I'm supporting a

CRA above an implementation project with

the office of enrollment management at

University of machine so what I'm doing

is I'm sitting down with users in like

financial aid in admissions and I'm

looking I'm sitting with them and

watching them do their job and from what

I'm learning I'm making these affinity

notes and I'm looking at their discrete

work processes and from then I'm making

recommendations to the the technical

implementation team about what features

we need in this CRM tool so that we

don't disrupt the users workflows when

we switch to this new tool in a year or

so that makes sense so that's just one

specific example in another instance

where we had really in an in a class I

had taken a public library came to the

class that I was in and said you know

we're having problems collecting

feedback from patrons so we looked at

how they were currently collecting data

from their patrons and then from our

affinity diagram made recommendations

about how they can better do that work

process

so first things and this sounds like

it's focused on kind of like with the

websites that I work on is a feeling to

concentrators

have you seen like maybe is that

accurate or

early

[Music]

I guess

yeah I think I think it's some similar

similarly to like usability test

recruiting where you would want

end-users to look at where you'd want to

see how and users use a tool I think

that this process would be definitely

appropriate for looking um at how

end-users use a system and then making

record or making changes based on how

based on your observations probably use

it but it just might take a little more

recruiting similar to how you would

recruit for standard usability testing

if that makes sense

yeah yeah right right so um the is a

small Public Library in in Flint

Michigan actually and what we did was we

sat in on any like a family area like

kids events or adult events that they

conducted for the community and we took

notes we just observed what the

librarians or the program facilitators

did especially specifically paying

attention to how they distributed

feedback requests at the end of a

program and from then we found you know

patterns between how specific librarians

got feedback and what they did with that

feedback afterwards to share it with the

larger staff and from there we saw a lot

of work process breakdowns wearing some

some librarians were uploading this

information to a shared server space

others were just throwing the feedback

forms on their desks it's forgetting a

bottom for weeks on end you know so with

our observations of what we saw and our

kind

holistic interpretation of the data

points we were able to make

recommendations for like um like a

shared space where they have to upload

their their user feet or their

participant feedback after they host a

program so that they can use to further

drive data driven data driven decisions

about what programs to bring back what's

programs to focus on the things like

that

you know your current project you use

are you gonna use analytics to see how

the system's working now for the site I

compare that writing well it's in what's

interesting about that is that there

this this CRM that's being used in the

office of enrollment management they

don't have one right now so all data is

just thrown wherever in each of these

individual offices and the CRM is hoping

to bring all the data together into one

spot and so we want to just make sure

that we're not breaking anything when we

introduce this new system so I'm kind of

just looking at their current processes

now and how we can sort of support those

in the future so there's not we're just

looking at I guess the current state

rather than the current state of how

they work rather than how they work with

a specific tool or system now what's

like a profound observation you so far

that was well so full disclosure like I

have always been just like amazed by you

know the University of Michigan like

it's this huge school that's like really

prestigious you know beyond my wildest

dreams that I'd be studying there so I

figured their admissions process I

figured everything would be repeated

pretty clean on the backend you know not

really stuck together with duct tape or

anything like that

many of the admissions processes for

specific schools are just copying and

pasting application data into Excel

spreadsheets sharing and I'm like what

like that my my my the curtain has

dropped I guess on anything that I

thought so personally it's just been

weird to see that like how how piecemeal

this system has

but in terms of the how that relates to

the CRM essentially we're just it's just

more fodder for the rationale for

needing a system like that because

there's so much air that can be

introduced in such a manual process so

that's kind of what my data is is

suggesting now that whatever's going on

now is just not foolproof enough in the

case of the librarian I just do all the

reviews hundred I stuff over there she's

not supposed to and they're not supposed

to do that how do you get them to show

you that I mean I feel like if you would

interview them they like one and do it

correctly like how they're supposed to

write not really sure that they kind of

be lazy right yeah that's definitely an

issue people want to clean up their desk

before you come and they want to you

know a lot of honestly with the library

example a big push back we had was that

wanting to meet like in a conference

room where we could you know that'd be

crowded with all their books and all

their papers everywhere but it's really

important for this method to make sure

that it's in context and you know the

first few times I've done this I

probably wasn't successful in allowing

them to show me their actual work

process you know but I'm to be pretty

adamant that I want to see exactly what

they do you know and so with the library

example in particular like sometimes we

you know we would tell them we were

coming but we weren't telling them

exactly like oh we're going to be you

know just observing this session but we

also want to come see how you do this

data process so it was a thing where we

not that we didn't set expectations but

that we didn't tell them exactly what

our process what our what would be

observing I guess you

I'd say just so that we wanted to make

sure they are being as authentic as we

could observe in their process we find

it hard to

we need to find the audience for a

specific person having outside people

say we need to include these people in

these people and these people is part of

the process yeah and it's a just

particularly with the library example

the director of the library was our main

contact and she just wanted us to talk

to other you know executives not

executives but like managers at the

library you know it was really hard

actually to get into to talk to the

actual librarians because they were so

busy with everything else like and

what's a good part what's a good kind of

a benefit of this if you can spin it the

right way is that it shouldn't take up

much time of the participant because

they aren't still technically working

you might just have to take time out for

questions during your contextual inquiry

process but it is it is a bit

challenging to squeeze in on people's

schedules like so it's anytime you can

build into context compensate for that

is definitely a good thing in that case

did you do any reporting of your

findings on if the employee processes

and then the managers see that and say

well no that's not how things are done

here at all right that that definitely

is a big concern I think with this

process especially with the librarians

you know because they didn't want their

bosses to know that they weren't doing

things according to protocol which is

why those user profiles and user codes

are really important because we sort of

anatomize the data with that but there

has been you know there has been cases

where in my CRM project in particular

where two people who are supposed to be

like the admissions directors for a

particular school for example are kind

of in disagreement about what the actual

work process is but it's the goal of

this contextual inquiry session to come

to kind of a shared interpretation of

what is actually going on even if there

is some conflict that just highlights

the need that the CRM tool needs to

implement you know a more streamlined

board process for them to follow

yes he talks about a little bit about

evaluating and for this within the

context of a larger

but I was kind of curious if you like if

you this monster if you repeat this over

and iterate over it and like once you

get the results you kind of refine it

over time and see business change

facilities if you're thinking if some of

these things about doing this is their

recommendations but okay we can't do

this twice you're going to get like

integrated over multiple ones maybe

you're going to get them then success

but you know it's kind of like what is

the range of an applications or having

to be

let me think about that yeah I in terms

of my eye I can I can kind of guess at

an answer but in terms of my my

experience with it I've only been

included on the kind of in the front of

the process you know I haven't been

there for any any revision or any kind

of iteration and in the process but I

imagine that for example with the CRM

system um after it's implemented in a

year so I can definitely see like an

application where we'd kind of do this

process again to see if it's any better

but I would I would kind of venture to

guess that it would be more about

collecting user pain points to

specifically make fixes in the system

rather than kind of map out a

large-scale plan for a revision but more

so how the implementation team can meet

their specific needs in the in the near

future that makes sense

kind of a different I guess a different

purpose with it um after each iteration

this process kind of defined you're

building a new system to educate the

users then do you are you gonna I take

this award to like content people to

kind of after you have the system

defined right yeah so beyond those as it

stands beyond the scope of what I will

be involved with as an intern but from

what I'm understanding there's going to

be a lot of training sessions for users

and a lot of content like help come they

don't documentation developed for use

but in terms of what I will be involved

with I can't really explicitly speak to

anything because it's really useful it's

a develop that kind of chapter right

right yeah though that'll definitely be

on the works I'm just not really privy

to that the end of the conversation yep

in when you're constructing how do you

decide how many levels of hierarchy your

new group things under I think you have

three levels um honestly that was what

our professor told us to do in terms of

that one in particular but I think a

general rule of thumb was to have I want

to say between like three and seven this

great data points under each larger

heading before building on to the next

one just so that it's a manageable range

anything smaller might not can probably

be grouped into something else than

anything

can probably be broken up a little bit

so I would say that three to seven range

is probably the sweet spot for that but

I don't know what that's based on that's

kind of just what I've been taught

can you explain this infinity wall thing

a little bit more detail what what is

this yeah

so I'll show you a video of it actually

I think we have a video somewhere yeah

about five hour process this is just my

my work team from last semester so we

had all the sticky notes on the table

and here we're just kind of constructing

the crude weights at this point kind of

common themes we found so like I said

during our interpretation sessions we

any discrete data point any you know

particularly strong quote anything that

could stand alone as a data point I

guess any particular breakdown in the

work flow we recorded it as an affinity

note on a little sticky note and we had

probably close to 500 at that point this

is a pretty relatively I would say

large-scale this is pretty I realized

this is pretty a pretty big time

commitment on a typical development

process so this is maybe not realistic

in a lot of cases but we had those 500

sticky notes and we kind of just started

grouping and we started merging groups

taking groups apart until we come to me

my team and I came to a common

understanding of what these main

groupings should be and then we just

kept building upwards of the thick

groupings and then that's kind of it

I don't know that answer your question

for us

okay okay yeah it's possible to employ

when you're having these conversations

with your clients because if I find as

often

their clients will hire hire team to to

fix a problem but in order to fix the

problem got to talk to them about you

know pain points and weaknesses and this

could be difficult conversations so

I think one of the things that I have

found is crucial to that is situating

the user as the expert situating

yourself as there just to learn from

them by putting my you know by putting

your confidence in them and situating

them as the expert in the situation and

you're simply just there to learn from

them I think that really from me at

least it has helped and it breaks out

any barrier that might that might arise

but I think also you know it depends on

how you talk to them how you are you

come into the session if you come in

really straight-laced and like you know

and almost intimidating in a way like

like you're hoping to but like we want

to be sure that we're not like it

doesn't come across as like for auditing

them we want to avoid kind of the

evaluation tone of voice evaluation

questions the availa depressions that we

want to just learn in that case so I

think situating it as a learning session

an observation session will sort of work

to break down the barriers but then also

any you know standard good interviewing

techniques come into play with body

language what certain words you choose

and things like that so

so definitely interpersonal skills are

it's really important in this particular

research method

you're recording stuff oh yeah cuz here

baby

oh whoops thank you

that's what that means okay thank you

I'm going to say you know and add to

that because I know what some things

that I do that helps me out this right

so the more you kind of you know get

them up talking about sounds the

processes on video right cuz that's go

very simple but any question and then

they're just going on and on and on

right yeah

I'm just wondering

Senator John

they might be against that you know like

has been talking to

the folks that work and they don't they

don't want to take away for a time

reasons but I think it's more

right so so with my current scope I I'm

just like kind of along for the ride on

the implementation team in particular

and I haven't had to deal with that

necessarily but any I think I think

persuading the it really deals down to

making the use the cake that you know

the argument that this is useful for

your end product this is useful for

productivity this is useful for

streamline work processes things like

that and what I found is that being

really flexible on like availability is

has been important like with this

project I'm currently working on I

started scheduling interviews in January

and I'm scheduling out to April at this

point like I'm trying to be really

flexible with people with when people

are available and taking any time I can

get as I mentioned like a typical

standard session would feel like an hour

if I can get 15 minutes like I'll take

found be patient on my end but I think

there's a lot of negotiation buying has

to happen at a higher level beyond that

which I can really intelligently speak

to you right now

you think an hour is enough because I

cannot explain our working environment

right right so yeah it's I mean an hour

would probably be like I said I've done

like up to two hours and I found after

the first hour focus is lost people are

trying to multitask ideally it's good to

be able to observe like one discrete

task that would take an hour or so but

if it if it comes down to where you're

you need to observe this multi hour-long

process I think that's appropriate but

we just want to be sure that we're being

you know genuine to the project focus I

guess if that makes

[Music]

and the client was a law firm insulting

people's life the hand progress

the water

and you kind of mapped out as people hit

the show we have to show your income

good show you have any assets you've had

show tell the story of what happened

there so it's kind of set up to six

greens you know there's a discrete

screen for your income pre-treating for

your assets and it look like a really

great process so we all went out to see

one of these days we walk in and it's a

room of 75 people that are coming in to

talk about you know how they've learned

losing their house keys and we realized

that what would happen is people say

okay you have your ear a bunch of stuff

and they find water if we find it and

then you five streets later it's a all

the way I felt receipts and so you can't

really apparent that you were going to

be going back and forth between extremes

and it's good

and it was great aha moment but oh we

have to consolidate occipital couple

studies

so these people can jump around quickly

and that's kind of honest things we

wouldn't get unless you actually saw

this happen

right so it's like I said this might be

this the scale might be realistic

unrealistic excuse me for a lot of

organization because it takes so long if

you don't have like a specific research

team in place but even just observing

contextual use even just you know for

for a few minutes or a few hours is

super useful in terms of how usable you

can make the system later on

[Music]

do you find people and get off task as

they're showing you to say well I think

it should do this

sorry oh definitely definitely

um I try to I try to stay clear that as

much as I can but if somebody is off

talking about how a specific a specific

point they have I just let them go like

there's a point where I kind of like

things people think should yeah

definitely it's it's um I found what

specifically with this CRM

implementation process I'm working on

particularly one of the interviews I did

the woman I was speaking with just

before I even introduced myself or

anything she just went our process sucks

I hate doing all this blah blah blah

like I had to reorganize a little bit so

it you know it threw my my internal

protocol off for the system of it but it

was really useful information because I

could see kind of what she was doing and

what she was struggling with and I just

kind of took it I take you know any data

that you get is useful it might just

require some some finagling on your end

to to bring it back around to observing

actual work process

awesome okay thank you guys