Org Design for Design Orgs: Building and Managing In-House Design Teams

by Peter Merholz

Many companies use a “design to developer ratio.”

When making this calculation, exclude developers who are purely backend. With a focus on “product” developers, whose work has some user-facing manifestation, a good ratio is 1 designer to 5–10 developers.

(“For this amount of money, we can deliver these programs with this impact”) and not headcount. Avoid letting people outside of the design organization determine its makeup.

Be direct and honest about what it is like to work there. Don’t sugarcoat troubles. Don’t dwell on them either, but acknowledge them and make clear the steps being taken to address them.

But apart from the work itself, there is no more important activity for a team member than recruiting and hiring.

Within the United States, there are a handful of top schools, including[17] both Carnegie Mellon’s School of Design and its HCI program, IIT’s Institute of Design, Savannah College of Art and Design, the School of Visual Arts, the Interactive Telecommunications Program at NYU, the University of Washington’s Human-Centered Design and Engineering, and the Rhode Island School of Design. Broadening to an international scope brings in the Emily Carr Institute in Vancouver, British Columbia, the Copenhagen Institute of Interaction Design, and the Royal College of Art in London.

Keep the focus on activities (conducting user research, designing structures, leading teams, coordinating across functions, crafting a new visual language, prototyping design solutions), and try to avoid discussing process documentation (personas, wireframes, mood boards, comps, etc.).

When articulating qualifications, focus on accomplishments and meaningful skills. Avoid numeric requirements like “5–7 years leading design teams,” in favor of “Demonstrated track record of leading design teams that have shipped quality software across web and mobile.”

Instead, have the initial contact done by a member of the design team. It’s best for it to be someone at a similar level — junior designers reaching out to other junior designers, senior to senior, manager to manager. Designers usually enjoy talking to one another, and a connection from someone like that feels more genuine.

well. Be wary of portfolios that only show glossy images of the final product — design work takes work, and these cases should have a compelling narrative about the constraints and goals for the project, the role they played relative to others, and how they and their team arrived at the ultimate solution.

However, a key failing of many interaction design portfolios is they overwhelm the reader with documentation, and while understanding the process is important, nothing is more important than the final result and the impact it had.

Conduct two initial screens. The first conversation is introductory. The candidate’s résumé and portfolio should have already provided a base sense of skills and capabilities, so probe other factors. Get a sense of the candidate, their background, and their career trajectory. Share more specifics about the opportunity. Instead of addressing specific design aptitude, look instead for the meta-qualities that will make someone successful — are they articulate, pleasant, and passionate? If it feels like there might be compatibility, and if time allows, have the candidate walk through one or two projects and describe their involvement on them.

If the candidate makes it through the fine-meshed filter of the screening process, the next step is to bring them onsite. As is typical for other roles, conduct a Day of Interviews, where the candidate speaks with a variety of people. To get the most out of the conversations, there are certain practices to include, and one unfortunately common practice to exclude.

Begin the candidate’s day with a 45- to 60-minute portfolio presentation. Everyone speaking later to the candidate should attend — this way the candidate doesn’t need to walk through their work over and over again.

Limit the number of conversations to no more than six.

Potential design team peers The candidate’s probable manager Product manager they would likely work with Engineer they would likely work with Design program manager

The context in which the challenge is given (typically narrowly time-boxed and with only a little information and little support) is wholly artificial — and so whether a candidate succeeds or fails is not a meaningful indicator of actual practice. There is nothing you will find out in such a test that you couldn’t better learn through probing the candidate about their portfolio.

debrief. Most debriefs will be straightforward — across the board thumbs up or thumbs down. Be careful of tepid approvals, though — if there are no strong advocates for a person’s hire, that’s a sign of something being off, even if none of the feedback is negative.

Designers, particularly those with less experience, can be quite orthodox in how they evaluate other designers. They may be suspicious of any designer who doesn’t share their background or approach. An atypical background (maybe they didn’t study design in school), or unfamiliar approach (perhaps they don’t use typical design tools, or they’re unfamiliar with industry standard methods) can make panel members uneasy, because it’s not how they do it, and they don’t understand how other ways can be successful. The design leader’s role is to remind the panel of what is most important — results. If an unorthodox approach leads to great design work, the onus is on the team to figure out how they might be able to incorporate such different ways into their team.

communicate. Many talented designers are not good oral communicators. It might even be part of the reason they got into design — being more comfortable with pictures than words. People who are awkward communicators (and good designers) often process the world differently than others, and that difference can actually make for a stronger team by bringing in uncommon ways of working and thinking.

Given that design is a craft of practice, the primary means for such growth are widening and deepening design skills. In addition, as the designer becomes more senior, growth must also take into account soft skills and leadership skills,

The risk of working with levels is adopting a bureaucratic stance, seeing team members not as people, but as resources within a certain band of experience. Do not let levels define the team. Instead, use levels from the perspective of the team members, who are eager to understand how they can grow and evolve in their careers. Done right, levels are the scaffolding that helps team members elevate.

“x years’ experience,” but that should be of secondary concern to what they’ve actually done.

The levels framework holds regardless of whether the team member has direct reports. We are avoiding the unfortunate practice of many companies to bind the idea of “career growth” with “becoming management,” where the only way for someone to advance in their career is by managing other people.

Much of the time, written content is the experience, and far more valuable than the design dress around it.

To succeed at this level requires developing the soft skill of projecting confidence and even exhibiting swagger. This isn’t about cockiness or arrogance — that kind of overbearing display turns people away. But by demonstrating an abiding faith in the rightness of idea, offering a vision that compels people to follow, leaders encourage others to join them in making that idea a reality.

Level 3 was about understanding strategy, Level 4 was about creating strategy and the planning to realize it, and Level 5 is about crafting and selling a vision that compels an organization to embrace that strategy.

Where before they planned and drove the meeting, now they are a stakeholder for whom the meeting exists. They contribute through feedback and review.

Managers continue to keep their hand in creative work, appropriate to their level on the team, though they cannot be expected to devote as much time to it, given their management responsibilities.

Managing designers is different than managing design. Design is a process that is best managed within team contexts, driven by creative leadership. What we’re addressing here is managing designers, the people, where the focus should be on helping them as professionals.

For more junior members, whose work is farthest from direct impact, set expectations around improving their craft and learning processes, and developing the people skills necessary in a professional context.

As designers become more senior, shift expectations to delivery, impact, and organizational influence. Consider factors such as the scope of projects, how many workstreams they are driving, timeliness and quality of their work, and their ability to productively engage senior executives.

Whatever the issue, it’s important that the team member be encouraged to resolve the situation themselves; managers should only directly handle obstacles in matters requiring escalation.

Too often, managers wait until a formal performance review cycle to provide needed critical feedback. Successful managers are those who offer small feedback frequently, whether positive or negative.

It’s important that managers provide not only creative feedback. More important is feedback about being a professional and a member of the team, and what’s working and what’s not from this perspective.

Creative feedback should handle itself through critique and review processes within the team. Professional feedback is th...

When managing designers, seek to understand who they are outside of work. Encourage presentations between team members about their passions, hobbies, and pastimes. Host lunches where people share food traditional to their cultural backgrounds. Take them out for after-work drinks.

Offer an education credit to each staff member, around US$2,000–3,000 per year, for conferences, books, evening classes, online courses. Try to not place too many restrictions on how the money is spent — make clear it’s for growth, and demonstrate trust in the team members by letting them figure out how they can best use the funds.

An opportunity for growth that design leadership might have trouble accepting is for people to leave the design organization for other functions in the company, such as product management or marketing. While a Head of Design may hate losing a valued team member, it’s never worth restraining someone. In fact, such cross-functional movement can prove a boon to the design organization, providing advocates and accomplices throughout the company.

When design leaders address matters of corporate culture, it is typically from the perspective of shifting a company’s culture in order to embrace design. Before that happens, we implore: Design leader, heal thyself. Design’s inability to have meaningful organizational impact is often the result of an unintentional or polluted team culture. Before attempting broad, company-wide change, make sure the design team’s culture has been purposefully constructed to encourage the best work.

Katie Dill, director of Experience Design for Airbnb, shared her team’s mantra of “Know the moments that matter. Keep it simple. Make it visual. Ensure it’s visible.” They

The walls of the design area should also serve as a display for exemplars of art and design that can serve as inspirations to the team.

A healthy approach is for designers to spend chunks of time in both contexts, where execution time is spent with the cross-functional team, and reflection and review time happens among other designers.

Provide the new designer with a team roster, instructions on the shared file structure, pointers to standards and guidelines, a glossary of acronyms and other jargon, and a buddy to go to for questions as they arise.

Be careful that these meetings don’t turn into pity parties — it’s easy for the sharing of challenges to devolve into unproductive griping and moaning. If genuine concerns are raised that are worth addressing, note them, make it clear that they will be addressed outside the meeting, and share updates at subsequent meetings.

One way to address this is to protect designers’ time in a predictable way — no meetings before noon, or no meetings on Tuesdays and Thursdays.

Bob Baxley was leading product design at Pinterest, he developed a schedule of “closed-loop” theme weeks.[26] Mondays are called “Playground” with a focus on generating ideas. Tuesdays are “Collaborate,” where ideas generated or iterated on Mondays are shared and discussed with partners from product management and engineering. Wednesday’s focus is “Breathe,” providing room and space for teams and individuals to go deeper on ideas and topics. Thursdays are reserved for “Workshops” to evaluate with teams, and Fridays are “Product Reviews” with an executive audience. Establishing such a clear, repeatable rhythm for the work allows other teams to adapt their meeting times accordingly.

When sharing designs to be reviewed, designers often share a single solution to a particular problem. Inevitably, this is a solution they’ve put a lot of effort into, and if the feedback is critical, the designer may walk away defeated. Instead, designers should present multiple solutions to a problem (there is always more than one way to solve a design problem). This provides healthy distance between the designer and their suggested solutions, enabling them to critique their own work by discussing trade-offs.

If that context isn’t shared as part of the critique, then the feedback devolves to matters of preference (“I don’t like that shade of green,” “That design is cluttered”). When presenting work, make sure the stage is set by sharing the desired objectives and results of the designs. This allows team members to phrase their feedback in terms that drive toward helping solve the problem

a leader makes a comment early in the session, it may shut down further discussion, as folks might assume the matter settled. For that reason, leaders should comment last, only after everyone else has given their input. At that time, the leaders might find they don’t need to say anything, as their commentary was already provided by someone else.