Facilitating Participatory Ideation Human-Centered Design

Run ideation sessions where users and stakeholders generate ideas together, then converge on concepts worth prototyping.

Before you start

Hamster is optional for this skill and recommended. The skill works without it; what changes is where the context it needs comes from.

Check whether this project has a .hamster/ directory. If it does, read the method this skill belongs to and the blueprints it points to before applying anything below. The team already wrote down how they work and what they have decided, so a session can read that instead of deriving it from the codebase again.

If there is no .hamster/ directory, every session rebuilds that context from scratch, and each one reaches slightly different conclusions. Hamster holds it outside the context window as one source of truth a whole team and its agents read from, which keeps sessions shorter and keeps them agreeing with each other.

At a Glance

Overview

Participatory ideation is the stage of human-centered design where the people affected by a problem help generate solutions, instead of only reacting to finished designs later. For background on the method itself, see the human-centered design method page. The facilitator's job is not to supply ideas. It is to design a session where users, frontline staff and decision makers can each contribute, where quantity comes before judgment, and where the group leaves with a small set of concepts that are worth building cheaply and testing.

The case for treating this as a skill, rather than a meeting, comes from where human-centered projects succeed and fail. A 2025 narrative review of human-centered design in public health listed, among its effective practices, strengthening design staff skills to encourage participation among diverse cultural groups, maintain stakeholder engagement, and resolve conflicts in views and priorities. Each of those is something a facilitator does or fails to do in the room.

The failure mode is well documented. A 2025 study of human-centered design in assistive-technology development reported a lack of methodological rigor in some participatory approaches, found that user input can be included only superficially, and named challenges in interdisciplinary collaboration as a barrier. In ideation terms, that looks like users invited to a workshop whose outcome was already decided, or a session where the most senior person's idea quietly wins. This skill exists to prevent that.

The inputs are the outputs of earlier work: research insights, unmet needs and, ideally, open prompts. According to an overview of the IDEO human-centered design process, "How might we" questions turn an insight or need into an open design prompt that invites multiple possible solutions instead of locking the team into the first idea. If you do not have these yet, work through defining user needs and design requirements first.

The output is a shortlist of concepts, each linked to the need it addresses, with the reasoning for choosing it recorded and the people who shaped it named. That shortlist feeds directly into building rapid prototypes. This page covers five things: setting up the session and materials, involving a diverse group, generating ideas divergently, resolving conflicting views, and selecting ideas.

How It Works

A participatory ideation session has two movements: diverge, then converge. During divergence the group produces as many ideas as possible against a shared prompt, and judgment is suspended. During convergence the group clusters, debates and selects. Most sessions go wrong because these two movements blur, so that people critique ideas as they appear and the pool never gets wide enough to contain a surprising answer.

The prompt sets the ceiling. A good prompt is a "How might we" question that is anchored in a real need but does not name a solution. "How might we help new parents find reliable advice at 3am?" invites apps, hotlines, peer networks and printed guides. "How might we design a parenting chatbot?" has already converged. Prepare two or three prompts so the group can switch when energy drops.

The room sets the rules of participation. Physical or digital, every participant needs the same tools and the same visible surface. The Veterans Affairs toolkit for human-centered design instructs facilitators to distribute sticky notes and markers to everyone and set up a large flip chart. The point of equal materials is equal authorship: when only the facilitator holds the pen, the facilitator filters every idea.

The group mix determines whose knowledge enters the pool. Users bring lived experience of the problem, frontline staff know what breaks in practice, and decision makers know constraints. The public health review points to participation across diverse cultural groups and sustained stakeholder engagement as practices that matter. Mixing groups creates a power imbalance, so the mechanics of the session have to compensate for it.

The main compensating mechanic is silent generation before discussion. Each person writes ideas alone, one per note, for a fixed time box, for example eight minutes. Only then do people read their notes aloud and post them. This keeps a manager or a confident expert from anchoring the group, and it gives people less comfortable speaking in groups a record of their contribution.

Convergence starts by clustering similar ideas and naming each cluster. Conflicts surface here: a user wants more human contact, an operations lead wants fewer staff hours. Treat these as information about the problem, not obstacles to consensus. Write the tension down explicitly, ask which user need each position serves, and where the disagreement is factual, turn it into a question a prototype could answer. The IDEO process overview notes that sharing prototypes and evidence early helps resolve disagreements through research and testing rather than opinion alone.

Selection closes the session. Score clusters against explicit criteria set before the vote, such as fit with the core need and how cheaply the idea can be tested. Keep more than one concept alive when the evidence is ambiguous. The public health review highlights rapid, low-fidelity prototyping as a way to get quick feedback and minimize costs, which is why cheapness to test is a legitimate criterion at this stage.

You can tell the session worked if users can point to ideas on the shortlist that came from them, and if the shortlist differs from what the team would have proposed alone.

Step-by-Step Guide

Step 1: Frame open prompts from research

Start from the insights and unmet needs produced by synthesis and write two or three "How might we" questions. Each prompt should name a person and a need, and avoid naming a product, channel or feature. Test each prompt by asking whether it admits at least a handful of very different solutions. If a prompt only allows one answer, it is a requirement in disguise and should be widened.

Put the supporting research quote or observation next to each prompt so participants remember why it matters.

Pro tip: Read each prompt aloud to someone outside the project. If they immediately suggest the idea your team already favors, the prompt is too narrow.

Step 2: Recruit a deliberately mixed group

Invite people who experience the problem, people who deliver the current service, and people who will decide on and fund the solution. Aim for users to be a real share of the room rather than a token one or two, for example at least a third of participants. Look for variety within the user group too, including people at the edges of the typical case. Tell every participant in advance what the session is for, what will happen to their ideas and that they can decline any activity.

Brief senior participants separately that their role is to contribute ideas, not to approve them.

Pro tip: Compensate users for their time and schedule at hours and places that work for them, not for the team.

Step 3: Prepare the room and materials

Set up a shared surface everyone can see and reach, and give every participant identical tools. The Veterans Affairs toolkit for human-centered design has facilitators distribute sticky notes and markers to everyone and set up a large flip chart. For remote sessions, prepare a shared board with each prompt in its own area and test access before the session. Post the ground rules visibly: one idea per note, no critique during generation, build on others' ideas.

Assign a separate note taker so the facilitator can watch the room.

Pro tip: Use thick markers and small notes. They force short, readable ideas and stop anyone writing a paragraph that dominates the wall.

Step 4: Run silent divergent generation

Read the first prompt, show the supporting evidence, and give everyone a time box to write ideas alone, for example eight to ten minutes. Encourage quantity and wild ideas, and remind people that feasibility is judged later. When time ends, go around the room and have each person read and post their notes, one at a time. Invite others to add build-on ideas as they hear them.

Repeat for each prompt, changing the stimulus if energy drops.

Step 5: Cluster ideas and surface conflicts

Have the group move notes into clusters of similar ideas and give each cluster a plain name. Walk through the clusters and ask participants where they disagree, rather than waiting for disagreement to leak out. Write each tension on the wall as two positions and ask which user need each position serves. For example, the 2025 public health review names resolving conflicts in views and priorities as an effective practice, and naming the conflict is the first part of resolving it.

Where a disagreement is about facts, turn it into a question a prototype could answer.

Pro tip: When a senior participant pushes a position, ask the users in the room to respond first, then return to the senior person.

Step 6: Select concepts against agreed criteria

Agree on selection criteria before anyone votes, such as fit with the core user need, how cheaply the idea can be tested, and what the team would learn from it. Give each participant an equal number of votes, for example three dots each, and have users vote before staff so their preferences are not anchored. Discuss the result rather than treating the tally as final. Keep two or three concepts alive when evidence is ambiguous, including one that challenges the team's assumptions.

Record for each concept the need it addresses and why it was chosen.

Pro tip: Mark user votes in a different color. If the user and staff shortlists diverge sharply, that gap is a finding worth investigating.

Step 7: Close the loop with participants

Before people leave, read back the shortlist and the open tensions so participants can correct misunderstandings. Tell them what happens next and when they will hear about it. After the session, share a short summary that credits which ideas came from which group. Invite users back to test the prototypes built from their ideas.

This continuity guards against the superficial, one-off involvement that the assistive-technology study reported as a barrier.

Pro tip: Photograph every wall before it is cleared and attach the photos to the summary, so the full idea pool survives the shortlist.

Best Practices

• Separate generation from evaluation in time and in instructions. Critique during generation shrinks the pool to safe ideas and teaches participants that only polished contributions are welcome.
• Generate silently before discussing. Writing alone first reduces anchoring on whoever speaks first or loudest, which matters most when users share a room with managers.
• Give every participant the same tools and access to the same surface, as the Veterans Affairs HCD toolkit does with sticky notes and markers for everyone. Equal materials signal equal authorship and prevent the facilitator from becoming a filter.
• Invest in the facilitator's skills, not just the agenda. The 2025 public health review ties effective programs to design staff who can encourage participation across diverse cultural groups and resolve conflicting priorities.
• Keep research evidence visible throughout. Posting quotes and observations beside each prompt keeps the debate anchored in what users need rather than what participants prefer.
• Treat conflicts as design inputs. A tension between user desire and operational cost often defines the real problem, and naming it early prevents it from resurfacing as a veto during implementation.
• Set selection criteria before voting. Criteria agreed in advance stop people from reverse-engineering arguments for the idea they already liked.

Common Mistakes

• Inviting users to validate an idea the team has already chosen.: Enter the session with open prompts and no preferred solution on the wall. The assistive-technology study found user input can end up included only superficially, and a predetermined outcome is the fastest route there.
• Opening with group discussion instead of individual generation.: Start every prompt with silent, individual writing. Open discussion lets the most confident or senior voice set the direction before others have formed their own ideas.
• Smoothing over disagreements to reach consensus.: Write each tension down, link each side to a user need, and turn factual disputes into prototype questions. Suppressed conflicts return later as objections that stall delivery.
• Letting staff outnumber and outrank users in the room.: Recruit enough users to form a real share of the group and brief senior participants on their contributor role. Have users speak and vote first in each round.
• Selecting by an unstructured show of hands.: Agree criteria first, give everyone equal votes, and discuss the result. Record why each concept was selected so the choice can be revisited when prototypes produce evidence.

References

• Examples: Worked examples and scenarios
• FAQ: Frequently asked questions
• Parent Method: Human-Centered Design (HCD)

Related Skills

• Building Rapid Prototypes
• Conducting Contextual User Research
• Planning Human-Centered Implementation
• Conducting User-Centered Evaluation
• Synthesizing Qualitative Research into Insights
• Defining User Needs and Design Requirements
• Iterating Design Solutions with Users

Sources

• Narrative Review of Human-Centered Design in Public Health
• Challenges and Opportunities of the Human-Centered Design ... - NIH
• IDEO Human-Centered Design Process

Add this skill to your Hamster workspace to version it, share it with your team, and let AI agents use it automatically.