A design can accomplish its immediate goal and still ask too much of the person experiencing it.
It might secure a purchase someone did not fully understand, move work out of one department by quietly placing it on the customer, or make a business process more efficient while making the people inside it less visible.
Ethical Design means taking responsibility for those consequences. It asks how the choices we make affect people's understanding, agency, access, privacy, and well-being—not only whether the product or campaign performs as intended.
At Abstraction, this is part of the work from the beginning. It influences the questions we ask, the ideas we recommend, the tradeoffs we discuss, and the decisions we are willing to put into the world.
Human value is not separate from commercial value. Understanding what people need and creating something worthy of their trust is part of how we want a business to succeed.
Care has to affect the decision
It is easy to say that people matter. The more demanding question is what changes because they do.
Does the person understand what they are agreeing to? Can they correct a mistake or decline an invitation? Are we asking for information that genuinely supports the experience? Have we considered someone who cannot participate in the way the design assumes?
We also look beyond the visible interaction. A smooth customer experience may depend on an employee repeatedly repairing a broken process. An automated response may sound reassuring while leaving the underlying issue unresolved. A beautiful space may communicate welcome without providing a practical way for someone to use it.
Human-Centered Design helps us investigate the people and circumstances around the work. Ethical Design asks us to take the implications seriously when deciding what to do.
The point is not to claim that we can anticipate everything. It is to remain responsible for what we can reasonably examine—and what we learn afterward.
A real professional boundary
Abstraction will challenge—and, when necessary, refuse—decisions that knowingly exploit people, conceal material information, misuse their data, or create avoidable harm.
That applies to how we design for end users and how we work with our own clients.
We should not use our knowledge to make someone less able to understand a choice. We should not present a proposal as certain when important questions remain open. We should not manufacture urgency around an engagement or make our reasoning inaccessible to keep a client dependent.
Challenging a decision does not have to mean abandoning the business objective. We can often examine a different way to reach it: make the value clearer, improve the service, change the sequence, reduce unnecessary work, or reconsider what is being measured.
But when the proposed mechanism depends on knowingly misleading or exploiting someone, it is not a creative challenge we should quietly help solve.
Responsibility extends beyond the intended user
The people affected by a design are not always the people named in the brief.
Customers, employees, clients, partners, visitors, and others may bear different consequences of the same decision. We want to understand those relationships while keeping the clearest emphasis on the end user's experience.
Who gains convenience? Who takes on the work? Who can make a meaningful choice? Who is being observed, represented, excluded, or asked to trust something?
These questions can change the scope of a design problem. A request for a more effective interface may reveal an operational issue. A campaign may expose a gap between the promise being made and what the organization can deliver. A proposed use of data may require a more careful conversation about purpose and access.
Service Blueprinting can help make those dependencies visible. Co-Creation helps bring relevant knowledge into the decision instead of asking one team to imagine every perspective alone.
Fewer support contacts is not the whole answer
When a business wants to automate part of its support, we examine what success would mean for customers as well as staff. The questions belong in the brief, before fewer contacts becomes the only measure.
The initial goal is to reduce the number of questions reaching staff. That may be a useful operational objective. It does not, by itself, tell us whether customers are getting the help they need.
A design could reduce contacts by providing clearer information and helping people resolve routine issues. It could also reduce contacts by making the support route difficult to find.
Those outcomes might look similar in one report and feel very different to the person seeking help.
We would want to examine the goal and the proposed mechanism together:
| Question | What it could change |
|---|---|
| Which questions can the experience genuinely resolve? | Define a useful scope instead of pretending every issue belongs in automation |
| How does someone recognize the limit of that help? | Explain what the system can do and provide an honest next step |
| What happens when a person still needs support? | Connect the interface to an actual service process, with realistic expectations |
| What does the business count as success? | Look beyond reduced contact volume to appropriate evidence of resolution, effort, and unresolved needs |
| Who reviews recurring difficulties? | Give the learning an owner and a route back into improvement |
We might recommend better self-service content, a more focused assistant, clearer handoffs, or a change in the underlying product that prevents the question arising in the first place. Where dialogue is useful, Conversational UX helps shape how someone asks, corrects a misunderstanding and reaches the right next step.
The ethical concern does not require us to oppose automation. It requires us to distinguish useful assistance from an obstacle designed to make someone give up.
We work with the team to connect those judgments to the interface, the underlying service and the evidence reviewed after release. A useful support experience should help someone resolve the need—not simply disappear from the report.
Transparency is something people experience
Important information can technically exist and still be difficult to recognize when it matters.
We consider the timing, language, hierarchy, and context of what people need to know. The question is not only whether the business has stated something somewhere. It is whether the interaction helps someone understand the decision in front of them.
Ethical Persuasion develops this responsibility around influence and choice. Inclusive Design examines the assumptions that can make participation harder for some people than others.
The same principle applies to AI. We should distinguish an automated exchange from a human conversation, a generated suggestion from verified information, and a proposed action from one actually completed. Human-Centered AI connects those distinctions to permissions, oversight, and the limits of the system.
Honesty should be carried by the experience itself, not left for the person to discover after something goes wrong.
Data is not detached from people
Research, analytics, customer records, and operational information can help us design more thoughtfully. Their usefulness does not make every possible use appropriate.
We ask what information the work needs, who can access it, what people understand about its use, and how long it should remain available. We distinguish a useful observation from an unsupported conclusion about someone's motives, identity, or circumstances.
We also consider whether the same learning could be retained without unnecessary personal detail. A report can preserve an insight without exposing the person whose experience helped reveal it.
Privacy, security, accessibility, and other specialized requirements need appropriate expertise and evaluation. This ethical position is not a substitute for legal advice or technical assessment, and a statement of intent does not certify an implementation.
Our responsibility is to recognize when those questions matter and make sure they are not silently excluded from the design conversation.
Work through the tradeoff; do not hide it
Projects have constraints. Time, budget, technology, and operational capacity affect what can be done now.
Ethical Design does not make those constraints disappear. It changes how we respond to them.
We can narrow a release, simplify an approach, retain an appropriate existing route, investigate an uncertainty, or explain what remains unresolved. We should not describe a known exclusion or unsupported promise as acceptable simply because correcting it would be inconvenient.
The response needs judgment. We examine the consequence, the people affected, the alternatives, and the work needed to make a responsible next move. Where necessary, that may mean stopping or refusing the proposed direction.
A record of the concern, decision, rationale, and follow-up helps prevent a temporary compromise from becoming an invisible permanent condition.
Keep responsibility in the loop
Ethical Design is not a final review applied to a finished idea. It belongs throughout Design Thinking:
Discover → Define → Explore → Design / Develop / Market → Learn → Advance → Discover.
What people experience may differ from what we intended. Feedback, observation, appropriate evaluation, and operational learning can reveal consequences we did not anticipate. We need to be willing to change the design, the measurement, or the original idea in response.
Evidence-Informed Design helps us examine what we know. Continuous Improvement gives that learning a way to affect the next version.
The ambition is not to declare ourselves incapable of making a mistake. It is to do work we can explain honestly, take concerns seriously, and act when the evidence tells us something needs to change.
If a business goal and a human concern seem to be pulling in different directions, bring us the situation. We can help examine what a more responsible response could look like.