Those affected should participate in design decisions

What it is

Participatory design holds that people actually affected by a system should take part in decisions about it — not merely be researched or tested. The object of participation is decisions — scope, priorities, trade-offs — not feedback alone; participants act as co-decision-makers, not data sources. The classic formulation is "nothing about us without us," rooted in treating design as a distribution of power: whoever is in the room gets their reality treated as the design's basis.

Why it happens

Affected people hold knowledge designers cannot substitute: fine-grained knowledge of their context (how the flow actually runs at home at midnight), early perception of harms (which phrasing makes people leave), and lived imagination of alternatives (local workarounds that already exist). Conventional user research samples this knowledge but keeps control with the design side — research questions, interpretation frames, and adoption decisions all belong to designers, and once lived experience is translated into a requirements document, its original context is gone. Participation transfers some of that control to the affected: problems are defined jointly, solutions are evaluated with context attached, and adoption is decided in shared settings. A second mechanism is legitimacy: trade-offs imposed on others (convenience sacrificed for safety) gain ethical force only through the affected's participation; unilateral benevolent decisions constitute no authorization.

Studying it

Participatory design is a research tradition of its own, born from co-designing workplace technology with workers, with a method family of workshops, future workshops, joint prototyping, and design probes. Research asks two kinds of questions: how to measure participation depth (actual influence on decisions, not attendance), and what participation does to outcomes and relationships (decision quality, trust, subsequent self-determination). Methodological cautions: satisfaction metrics easily dilute participation studies — participants can be satisfied while their input vanishes — so measurement must trace decision correspondence (which input entered which decision); and since participants are always a sample of a few, representativeness (who may speak for the silent) must be handled explicitly, not avoided.

Where it stops holding

Not every decision suits participation: internal technical choices (which storage engine) are neither accessible nor relevant to the affected, and forcing participation there is theater. Participation also does not replace professional accountability: final judgment on safety engineering and legal compliance stays professional; participation governs values and trade-offs. Time and scale are hard constraints — participatory processes are too costly to cover every feature, and typically apply to high-impact, high-controversy decisions while the rest run the normal process; the boundary must be stated in advance, or unmet expectations will backlash.

Applying it

• At project initiation, list affected parties (beyond direct users, include the indirectly affected) and attach a participation mechanism (workshop, advisory group, co-design session) to high-impact decisions.
• Bring genuinely open questions to the table: "what is not yet decided," not already-made proposals seeking endorsement.
• Route participation outputs directly into decision documents alongside professional assessments; where they conflict, record the trade-off instead of merging it away.
• Verify: track the input-to-decision correspondence rate (how many proposals, how many adopted, how the rest were accounted for); a persistently low rate means the participation is ritual and the mechanism needs rebuilding.

Related

• Same group: P4.06.2 Token participation is not real participation · P4.06.3 Participation requires actual allocation of resources and power
• Adjacent: P4.13 Operational dimensions of participatory design · P4.01 Value-sensitive design
• Search terms: participatory design · co-design · affected stakeholders