Video summary
The video introduces Peter, a seasoned professional who has transitioned from managing designers to becoming a design operations consultant. He explains that his role involves elevating the standards of design within client organizations by focusing on the structural circumstances that enable good design to happen. A key argument presented is that as design organizations grow, many administrative responsibilities traditionally held by design managers—such as defining processes, managing tools like Figma, and organizing team meetings—are increasingly delegated to specialized roles known as design operators. This shift allows design managers to return to their core strengths in creative direction, storytelling, and developing talent, while specialists handle the operational infrastructure that supports the entire team.
Peter defines design operations as the continuous and structural improvement of conditions for designers to increase the likelihood of successful outcomes. He emphasizes a mindset of constant evolution, noting that if an organization stops improving, it effectively stops growing. To illustrate this, he shares examples from his time at ServiceNow and Miro, where dedicated teams acted as "design police" to review projects at various stages of development, ensuring adherence to guidelines, design systems, and accessibility standards. These initiatives were not merely about policing but about supporting designers by providing clear feedback loops and preventing bottlenecks before they occurred, thereby creating a smoother workflow for everyone involved in the product lifecycle.
To effectively execute these improvements, Peter outlines a five-step program that transforms a chaotic list of ideas into a structured roadmap. The process begins with inventorying potential initiatives, followed by prioritizing them based on urgency and alignment with organizational goals like OKRs. He introduces the concept of horizontal versus vertical work: horizontal initiatives benefit the entire organization, such as creating career ladders for all designers, while vertical initiatives are tailored to specific teams or clusters, like helping a growth hacking team develop a unique process variant. The roadmap also involves drafting briefs for each initiative to assign clear goals and stakeholders, and finally, establishing metrics to measure impact, such as designer happiness, time-to-productivity, and the number of active designers counted within the organization.
Ultimately, the video concludes that design operations is not just an administrative function but a strategic driver for business success. By systematically implementing these initiatives, companies can significantly increase their revenue growth compared to competitors lacking such support structures. Peter highlights that the true goal of all this work is to make designers happier and more effective, which in turn benefits the business. He encourages leaders to adopt a design ops mindset or hire specialists to ensure that the right initiatives are executed in the right order, proving that investing in the infrastructure of design yields tangible returns for the entire organization.
Read the full video transcript
I am Peter. Uh I am Dutch. I am old. Um
I am married to a woman that I met at
this conference 14 years ago. So
um I've got two sons. I sort of say I
can sort of say I'm a designer. I've
been a manager of designer. I've been a
manager of managers of designers. Um but
uh nowadays I'm a a design operations
consultant which means I try to raise
the majority of design in my clients
organizations. I've worked at Miro uh
most recently as a full-time design
operations manager. Before that, at
Service Now, a large workflow provider.
Before that, at the craft, one of these
uh regional UX companies. Um uh and like
I said, I'm a design ops consultant. Uh
I tell my people uh they should I tell
my my client that they need to work on
design ops things. Um and I'm I'm not
the only one. Um here's a prediction uh
about design ops by uh Adrien Alnot uh
who was a director of design or is a
director of design uh uh design ops at
service now uh and she says in a couple
of years and she made this prediction in
2020 so sort of now we will see that the
responsibilities of design managers move
back to their core what of what design
managers should be focusing on which is
uh uh giving creative direction uh you
can read this creating innovative
designs
storytelling, developing designers and
stuff like that and other design
responsibilities that have sort of
accumulated on design managers will be
dedicated to specialists that are called
design operators.
Um so to to confirm that to see if if
you agree, do you think that these
responsibilities are responsibilities of
a design manager
uh defining the design process, keeping
track of all of the work that's
happening? uh making sure that your
designers are trained well. Uh making
sure that there's a design culture in
your in definitely in your design team,
but maybe in the in the in the entire
organization. Um picking the next uh
Figma uh or moving from Sketch to Figma.
Um and then uh making sure that that
happens uh smoothly. Um running all
hands for all designers. So your either
your weekly all hands meetings or your
offsite or stuff like that. Is that the
responsibility of your manager? That's a
lot of work. Um and a manager would
probably rather uh uh work on design
strategy and uh checking the work. So um
nowadays
these responsibilities are often seen as
the responsibility of design operators,
design operations specialists. Um this
is my definition of design operations.
Yeah, I'll be focusing on different
words of this definition throughout the
rest of the talk. Um, let's read it out
loud. We design ops continuously and
structurally improve the circumstances
for designers in the design organization
to increase the chances of good design
happening. That's what we all want,
right? Good design happening. Okay,
let's start in the middle.
Designers and the design at work.
That's you. Yeah, we're doing this for
you.
Um
we want designers in general and their
organizational their business unit to
thrive.
Um who are the wei then? Um we can be
anyone with the mindset that a good
design ops person has. I often I
mentioned this in the workshop. I often
call it like a gene that says I want to
do this. I want to every now and then
take a step back look at the
circumstances of this of design. How are
we doing things? Why are we doing this
this way? Is there a way one way that we
can make it a little bit better? If you
have that mindset, you're a good
candidate to start executing design ops
initiatives, to start thinking of new
ones, to start putting them in the right
order, to start influencing the the
circumstances for designers. Um, it
could be that your organization is ready
to have someone with the role of a
design operator that you can make it a
part-time responsibility, a full-time
responsibility, even for larger design
organizations, a a team of design ops
people, specialists, even a manager,
tool person, culture person, uh, process
person, specialists within the design
ops team. If the team is large, if your
design organization is large enough, it
starts making sense to set up a team
with specialists for design ops. Um, so
there doesn't have to be someone who has
the title. It's okay if there are
multiple people with the role. It's
actually lovely if there are more
multiple people with the role because
then you can work together uh and uh
distribute the responsibilities and
distribute the load. Um, so that's the
Wii part.
Um,
let's talk about the continuous
improvement. Why do you want to
continuously improve? Um, and Brad and
like I was tweaking my slides earlier. I
put in a photo of Brad that says if you
stand still, you die. Like if you're if
you if you're how did you phrase it,
Brad?
>> When you're finished changing, you're
finished.
Don't stop. Don't stop. Continue.
continue improving. Um there's this sort
of statistical rule. If you do 1%
improvement every single day in a year,
365 times, uh then in total you improve
38 times, forget about 10x engineers. We
can 38x design. If we make one
improvement every day, including
Saturdays and Sundays,
maybe we need to settle for 10x. Um so
that's why we want to continuously
improve. It makes all of us better in
total in aggregate. Um and what do we
do? We improve the circumstances for
designers. What does that look like?
It's it means we execute design ops
initiatives that make one of those 1%
improvements or maybe 5% improvements or
2% improvements. We execute design ops
initiatives. And in this talk I want to
show you some design ops initiatives.
That's what it said in the in the
abstract. So, I better do that. And for
that, I need my my mouse cursor. So, I'm
going to move over there. But, um, I'm
going to let you choose.
You have a hand over there. Hi. Which
one would you like to see?
>> Thank you very much. That's the one that
has sound. Um, and I want to challenge
the people in the back to to uh to do
the sound. Let's pick that one. Design
review service. It's one um from uh
service now where um
we had when I joined 275 designers,
researchers and uh writers 275 people we
were supported by a design operations
team of 25 people uh that had
specialists on uh uh balancing designers
over business lines that had uh uh
design system uh team that had uh this
review team, the design experience
review team, the dirt team um and they
would perform reviews um uh at set
stages in the product development life
cycle. So in in in during projects,
design projects um they would do one uh
sort of early conception where they
would talk about are you following our
general guidelines for how applications
general design work uh general service
now work. um do they would have do one
where they would review whether you're
using the right design system components
or whether you were using them at all uh
um and whether you were using the right
ones. So the short version of this is
this and whoop the sound of the police.
Um
they are the design police basically uh
which sounds bad but they can send you
back. They can say stop you cannot go
any further. you need to turn around, do
something over and do it better and come
to come come back to us. You would not
be allowed to proceed to the next stage
of design without their approval
on u multiple stages of your design
process. So they we had a defined design
process and since we were a workflow
tool, we were using our own uh our own
tool to manage our workflow of of design
projects. So we knew how it worked and
we all knew we respected this uh this
process. Uh so we respected the police.
Um the goal of this initiative was um to
support designers in the concept phase
in detailed design phase in usage of the
design system. They even had an
accessibility specialist on the team to
do reviews regarding accessibility of
your proposed solutions. Um the team
that uh was working on this was um this
design experience research u a review
team that was part of our designs
organization and it took a little while
to set up this service like define when
do we want to uh do these reviews how do
we make it happen uh how do we make sure
that designers come to us. So setting up
office hours for all teams in the
agenda. So this part of the business
unit could could go come on Tuesday
morning this part of the of the company
could come on Wednesday afternoon.
Setting up those kinds of systems took a
while, but then it was running and it
worked. Um, let's go back to the
overview. Let's do another one.
The first one, design process
definition. Okay. Um, that was a Miro
thing. Um, Miro's design process
definition, the large overview looked
like this. You can focus on the the bit
at the top first. Um the discover,
define, design/ress research, build and
ship was our product development life
cycle that was agreed upon with product
management, engineering, design, uh
product marketing and data analytics
team. We all followed these same steps
in uh in the development of new
products. So the design process was a
natural part of this. Um so for every
phase in our product development life
cycle we could define and this is where
I click we could define what are the
goals for the product design team in
this stage what are typical deliverables
in this stage what are typical
activities by designers in this stage
and not unimportant who do they
collaborate with and the word amp means
that's an acronym for our product
organization and in this overview of
proposed deliverables activities there
were uh research related activities,
accessibility related, design system
related. They were all marked in in a
color. Um, this was a like a large
diagram and then there were chunks of
this that would uh describe in more
detail on the wiki and people would
hopefully follow this. We didn't have a
design review police team. Um, but this
is what uh what Mir's design process
looked like. And the goal of of
documenting your design process is to
support designers uh uh in doing things
in the right way and mostly the right
way. Um uh and actually promoting this
way of working also to member other
members of project teams. So to the
engineers so they would know what what
they could expect from designers.
Researchers knew how they could
contribute uh product managers knew what
what they could expect. So it's not just
for designers. on the team was me as a
design manager. I was responsible for
for maintaining this design process and
all all of his documentation. Um but I
had the help of several fellow process
freaks uh fellow designers who liked to
think about the way they were doing work
and about and enjoyed standardizing that
this for themselves and their teams. Um
it took several months and then to
document this for the first time in a uh
uh in enough detail and all of its
details around this big diagram. Um and
every now and then there was follow-up
necessary for example because we started
developing a growth team that was
responsible for product hacking or a
growth hacking getting more people into
into our product through sort of
marketing and sometimes a little dark
pattern here and there. um uh and while
they were in the product to get them to
explore more of it and use the new
features. That was our growth team. They
were they were doing a lot of um uh sort
of hypothesis-based quick AB testing uh
experiments instead of the longer
month-long projects. Uh so their design
process was different, their timing was
different, their activities were
different. They were often based on uh
data and analytics uh uh input and
output. Um so they had a variation of
this large process uh just for them. So
that was the tweak or followup that we
developed after we developed the the
overall standard. We developed a
specific variation for the uh uh for the
growth team. Let's try one more. Try
let's do
>> IKEA.
>> The IKEA one. Okay. Uh the IKEA one. Um
that's good. I can introduce a little
bit of terminology. So let's we'll get
back to this. Um there is this thing of
horizontal and vertical design ops work.
Horizontal design ops work is work for
that's meant to benefit everyone.
Everyone everyone. Uh an example is
career letters for designers.
Career letters for designers are uh uh
are created so that everyone in a design
organization knows this is what it takes
for me to move to the next level to get
to get a promotion or at least
contribute to getting a promotion. Um
these are the things that are expected
from me in terms of skills and behaviors
and and typical deliverables in in this
role. Uh scope of work stuff like that.
Career ladders are a horizontal activity
that should developing those career
letters is a horizontal activity should
benefit everyone. uh an example of the
other one the vertical uh work is when
uh the example I just gave where we
helped the growth team develop a variant
of the design process. It was only
beneficial for the growth team. Uh just
them they benefit from it. Uh it was
initiated by their design leader the the
the director of design for the growth
team. They benefited from it vertical
activity.
um at IKEA
uh when I joined them almost all of the
work was horizontal.
So IKEA has uh like 300 digital
designers to not just not create new
flatp pack furniture but create all of
the digital work that IKEA needs
including the kiosks in the shops,
employee applications and the websites
and apps and stuff like that. Uh so they
have 375 300 designers, a design ops
team of 10 people. Uh I joined them for
a while and all of their activities, all
of their initiatives were horizontal
initiatives. Um but there was beginning
to be a need for more vertical more
vertical work. So I helped them specify
this role that one that every member of
the design ops team could take on for a
while sort of wearing a hat uh of being
the design operations partner for one
specific team of designers or cluster of
designers. Um and their manager, the XD
manager could request such a partner and
on this wiki page, it was a wiki page.
Um they could uh uh they could find out
about the existence of this service uh
realize that they could design uh or
request such a design ops partner, read
the whole role description,
um and then uh say if you want one,
click here. So they could request a
design ops person wearing this hat of
vertical uh support uh request that for
four months, eight months, 12 months.
This is how IKEA counts. Um, so the goal
of this uh uh of this new service, this
this hat that design ops people could
wear to benefit uh design uh leaders was
to optimize the implementation of the
most of the existing design op services
that were meant to be horizontal but
then tweak tweak those services to make
them fit for this vertical team and
maybe even develop new services that
were fit for this individual team. um
the the team to that worked on this
initiative was me. Um I did research on
what other companies call these roles.
Uh there there's a a typical design
program manager role or job description
in in the world of design ops that sort
of is like this. You temporarily help a
group of people get a program in place.
Um we had someone who who played this
role for a couple of months for one
team. So, I asked him all kinds of
questions. When I suggested this job
description, he said, "Yeah, let's tweak
it a little bit." Um, and a couple of
the design managers who were supposed to
push this button when they needed
support uh were also involved in the
review of this uh wiki page basically.
Um, it took a couple of weeks of
research. It took a couple of week or a
week of writing and reviewing um and and
tweaking. And then I had to wait three
weeks for approval from somebody high up
in the design organization.
Here are more examples. These are not
just the ones I I can show you more
details of, but these are known examples
of more design potential design
obsitives that you and your team could
develop. Um, and there's more more.
Uh, then the last one is tool research.
Like are we really going to go for the
enterprise version of Figma or not? If
so, who's going to do the quarterly true
ups of supposedly uh active users versus
that you need to pay for for real active
users. So these kinds of things are all
potentially in the scope of a design ops
person or team or people with the
mindset. Um so this is how you improve
the circumstances for design. You do
that by executing these types of design
ops initiatives. Um, and even though it
sounds it looks like we haven't touched
most of it, we're we're getting there.
Um, the the part of the doing it in a
structural way means you need to
well add some organization to the
potential uh of this long list of uh of
design ops initiatives. Uh, you want to
do it in a slightly more structural way.
And I have a five-step program for for
it that we went through in the in the
workshop on Wednesday. Um, and I I won't
go through everything. I'm just going to
show you quickly what the uh sort of the
output of each of these steps looks like
in the in the workshop. Step one, an
inventory of initiatives is just a
really long list of post-it notes uh uh
that have have been given a little bit
of structure. Ideas, you know, post-it
notes on a piece of paper. Um, step two
is where you try to prioritize this
list. Which ones are urgent? which ones
are important, which one are urgent and
important and which ones support the
goals that we have as a design
organization or as a larger organization
in general. If you have like an OKR
system or KPIs and you can show how your
design ops initiatives contribute to
reaching the goals uh or getting you
closer to the to the key performance
indicator, closer to the number that
that you're supposed to hit, then um
your design option initiative rise to
the top. if they check all the boxes,
they're they're at the top. Um, if the
ones that don't really contribute to the
goals that you've agreed on or as an
organization go down on the list until
they become more important and urgent or
until you've tweaked the OKR system that
came up in the uh in the workshop. Um,
this is one type of pri prioritization.
There's also the do we have the
resources and capability available right
now or do we need need to wait a little
bit? Uh one example was when at Nero we
wanted to change some some attributes in
Jira uh so that we could track design
work better in Jira. Um we wanted to do
that and then the IT people said ah wait
we're going to migrate Jira from on
premise to the cloud. Jira is frozen for
a couple of months. You cannot change
anything. You can still add tickets and
and stuff like that but you cannot
change the uh the uh the instructions or
the the implementation. Okay, that mean
meant that initiative had to wait. So
now next later there is another format
for for road maps that I that I prefer.
It has this now next later uh circles
with uh room for bands of work uh that
you can do like in the now in the next
and later based on certain themes that
you may have identified in your
inventory like themes like uh people
work u on skills and on boarding and
happiness surveys and stuff like that.
Uh another theme could be process stuff,
process documentation, tracking work,
tracking whether projects follow the
process, stuff like that. And it could
be a theme on tools and licensing and
stuff like that. Um, the themes are your
themes, the initiatives are your
initiatives, the goals are your
initiatives. You need to tweak this. I'm
not going to I won't tell you what
should be on here. That's up to you. Um,
this is what my road map at Miro looked
like. My design operations road map
looked like this. I'm not kidding you.
It looked like this. Well, actually, I
am kidding you. I'll show you in a in a
in a second why. Um it had like
references to OKRs for each of the
design ops initiatives. Uh uh so we
could show how design- ops work would
contribute to goals of the organization.
Um uh there's another step in the five
or two steps in the five-step program
that is about distributing the work
because you can never do it alone and
then coordinating it again making sure
that progress is measured and stuff like
that. A good way to kick off that is to
write briefs briefings for uh for your
design ops initiatives. And this is an
example of a briefing for one of the
examples that we didn't show. But it a
briefing has a title, a goal for the
team, uh a goal for the initiative, uh
the expected outcome, uh who would be
the stakeholders for that are relevant
for this initiative, who should be uh
who could be good team members, uh and
how we how do they report on progress,
which could be every Wednesday morning
they have a meeting and you just listen
in. Um step five is about metrics,
measuring the impact of such an
initiative. What is the current state?
What is the state that we want to reach
when the initiative is running when it's
or when it's done? What what kind of
behavior change do we see? What kind of
change in percentages, numbers, whatever
do do we see? Um there are many
potential metrics and I can say it
depends only so many times a day but it
depends on your organization which
metrics are good for you. But here are
some examples and the it starts with the
typical product project management triad
of uh quality, speed and budget and you
can pick two and uh and the other the
other one will suffer. Um but it could
be like the number of designers if
you're growing how many designers do we
have? How many people do we count as as
designers? Because there may be some
people in the front end uh uh
engineering world that you want to count
as designers. Do researchers count as
big D designers? Do they count? Are they
part of of our team? I was once in an
innovation unit and my designers were
not counted as as designers. They were
not sent the newsletter. They were not
uh supported by by the design ops team.
They were sort of the kids in a garage
that didn't count. Uh and I I made them
count. I made that change and suddenly
the number of designers went up. That's
good. Um then there are some some design
ops metrics that as a design person you
want to care about and designer
happiness. are designers happy with the
circumstances that uh that are currently
in place u is a big one but for example
and I'll go fast the last one the time
to first review was one uh at Mero when
Mero was growing fast and we were adding
more designers we wanted to track
whether they were uh productive fast
enough like you want you you start
hiring people you start paying them you
want them to be productive members of
the design community and uh uh one uh um
sign that they were productive members
was that they contributed to uh what we
called product review meetings where
they were presenting their work in a
product review meeting. If that was the
case, we c we we use that as a as a
place or as a as a sign that they were
productive members. So the time from
when they got hired to the moment that
they were productive members of the of
the of the community was uh uh something
that we wanted to measure and we wanted
to make short. We wanted a time to make
sure so they became effective fast which
meant uh if this number wasn't going
down then we we weren't do doing
onboarding well or they it took too long
before they got known and assigned to
teams. So onboarding and assignment of
designers were uh were investigated if
this number if this number would be
would stay low we would start fixing the
onboarding and the assignment process.
So this is where I said I lied. This is
what the actual Miro design ops work uh
road map looked like. It had metrics as
well for each of the for for the
initiatives. So we had uh an initiative
linked to a goal and uh with metrics uh
associated to it.
That's my five-step program.
Um I think you should have a design ops
road map uh to make sure that the right
design ops initiatives so the ones you
inventoried uh are organized in the
executed in the right order with the
right people with the biggest possible
impact and that impact is um increasing
the chances of good design happening
right and good design means it's good
for business and Gartner the the big
business analyst they've already
predicted that teams that have a design
support from design ops are uh uh more
effective. They increase the revenue of
the company at twice the rate of
competitors who don't have a design ops
team. Design ops is good for business.
If you need to convince your boss that
they need to give you time to spend on
design ops, use this quote.
Um so again we do all this we execute
these initiatives with the intent of uh
um making designers happier and those
designers are you. So um thank you