Skip to main content
XenonStack

Service Design

XenonStack Insights Interaction Design Principles and Best Practices

and | Published | Updated

Interaction Design Principles and Best Practices
14:27
Interaction design principles, process and best practices, shown as a four-layer stack from user goals to clear feedback and recovery

Key takeaway

  • Interaction design principles, the five dimensions, process and best practices, with WCAG 2.2 targets and how the heuristics apply to AI agents.

Interaction design is the design of how people and products talk to each other: what a person can do on screen, how the product responds, and how clearly it shows what happened. It covers behaviour, feedback, navigation and error handling, and it sits next to visual design and research rather than replacing them. The Interaction Design Foundation describes it as the design of interactive products and services, looking beyond the product itself to how users will interact with it (read on ixdf.org).

The principles that hold up best are the ten usability heuristics from Jakob Nielsen and Rolf Molich, which Nielsen Norman Group says have stayed unchanged since 1994. For enterprise teams the same ideas now apply to software that acts on a user's behalf: an agent that shows its status, can be stopped and explains its errors is easier to trust. For the wider discipline see our guides to user interface design principles and user experience design tools and principles.

What is interaction design, and how is it different from UI and UX?

Interaction design (IxD) is about behaviour over time. UI design is mainly the look and layout of the screens, and UX is the whole experience of using a product, including research, content and support. IxD is the part that decides how a control reacts, what feedback follows an action and how a person recovers from a mistake. The three overlap in practice, and a product team usually needs all of them.

DisciplineMain questionTypical output
Interaction designHow does the product respond to what I do?Flows, states, feedback rules, prototypes
UI designWhat does each screen look like?Layouts, components, type and colour
UX designDoes the whole experience meet the user's goal?Research, journeys, content and service design

For the UX side of software delivery, see user experience in software product development.

What are the principles of interaction design?

Five principles are widely taught, and each maps to the usability heuristics. They are a checklist for review, not a fixed rulebook.

PrincipleWhat it asks of the designRelated heuristic
Goal-driven designStart from the specific need the user is trying to meetMatch between the system and the real world
UsabilityMake the state of the system easy to see and understandVisibility of system status
AffordancesLet controls show what they do, so form follows functionRecognition rather than recall
LearnabilityReuse familiar patterns instead of inventing new onesConsistency and standards
Low frictionAvoid delay, surprise and needless thoughtAesthetic and minimalist design

The ten heuristics, as listed by Nielsen Norman Group, are visibility of system status; match between the system and the real world; user control and freedom; consistency and standards; error prevention; recognition rather than recall; flexibility and efficiency of use; aesthetic and minimalist design; help users recognise, diagnose and recover from errors; and help and documentation.

What are the five dimensions of interaction design?

Gillian Crampton Smith introduced four dimensions of an interaction design language, and Kevin Silver added a fifth, according to the Interaction Design Foundation. They are a useful way to check that a design has considered every channel a person uses.

Two-column diagram: left lists words, visuals, physical objects or space, and time with behaviour; right lists the review question for each, ending in a reviewed design.
DimensionWhat it coversExample question for a review
1D, wordsText such as button labels and messagesDoes the label say what will happen?
2D, visual representationsImages, typography and icons that support the interactionDoes the graphic help or distract?
3D, physical objects or spaceThe device and setting the person usesDoes it work with touch, mouse, keyboard and one hand?
4D, timeAnimation, sound, progress and resuming laterCan the person see progress and pick up where they left off?
5D, behaviourHow the product responds to input and gives feedbackDoes the response confirm what the person did?

What does an interaction designer do?

An interaction designer works with product, engineering and research teams to define how a product behaves. The work usually follows three activities.

Design strategy

Set out the user's goals and the interactions needed to meet them, so the team agrees on what the product must let people do before screens are drawn.

Wireframes of key interactions

Sketch the main flows and the screens that support them, including empty, loading and error states.

Prototypes

Build clickable prototypes so real users can try the flow and reviewers can apply the heuristics before engineering effort is spent.

What are the steps in the interaction design process?

Teams differ in detail, but the sequence below covers most projects. Each step produces something the next one can use.

StepWhat happensOutput
1. Find user needsInterview and observe users, review current optionsNeeds and goals
2. Analyse and structureTurn findings into scenarios and break tasks into stepsTask flows and scenarios
3. Design the solutionApply the heuristics and patterns, for example for navigation and feedbackWireframes and interaction rules
4. Prototype and testLet users try it, or have experts review it against the heuristicsIssues list and revisions
5. Implement and measureBuild, release and watch how people really use itShipped design and usage evidence

What are the best practices for interaction design?

These practices follow from the heuristics and from published accessibility guidance.

  • Keep it simple. Remove steps and options that do not serve the user's goal, so a person can complete the task without reading instructions.
  • Show progress. Show where the person is in a multi-step task. NN/g's response-time guidance, published by Jakob Nielsen, says delays near 10 seconds need feedback such as a percent-done indicator, because users will probably turn to other work while they wait.
  • Be consistent. Use the same control for the same action throughout. A green button that submits on one screen should not mean something else on the next.
  • Write useful error messages. Say what went wrong and how to fix it, in plain words. This is heuristic nine.
  • Prevent errors first. Confirm destructive actions, give undo, and constrain inputs so mistakes are hard to make.
  • Make targets easy to hit. WCAG 2.2 criterion 2.5.8 (Level AA) asks for pointer targets of at least 24 by 24 CSS pixels, with listed exceptions such as spacing, inline links and browser-controlled elements.
  • Do not hide keyboard focus. WCAG 2.2 criterion 2.4.11 (Level AA) requires that a focused component is not entirely hidden by author-created content such as sticky headers.
  • Give feedback that confirms the action. Each action should produce a visible result, so the person knows the system received it.

The WCAG 2.2 criteria are published by the W3C, which lists the standard as a Recommendation dated 12 December 2024 and adds nine new success criteria to 2.1 (see the W3C specification and its target size explainer).

How should interaction design handle response time?

Response time shapes how a product feels. According to Jakob Nielsen's article on response times, there are three limits to design for.

LimitWhat the user perceivesDesign response
0.1 secondThe system responds immediatelyShow the result, no extra feedback needed
1 secondThe train of thought stays unbroken, though the delay is noticedKeep the interface responsive, and avoid implying direct manipulation if it lags
10 secondsAbout the longest users keep attention on a taskShow progress, such as a percent-done indicator, and let people do other work

How does interaction design change for AI agents?

When software plans and acts for a person, the usual heuristics still apply, but they carry more weight. Visibility of system status becomes a visible record of what the agent did and why. User control and freedom becomes a way to pause, approve, reject or reverse an action. Error recovery becomes an explanation of what the agent could not do and what the person can do next. For one view of this in customer-facing work, see agentic AI in customer experience and interaction.

There is also a transparency point for products sold in the EU. As published in the AI Act Explorer (a Future of Life Institute site, so confirm the wording on EUR-Lex), Article 50 asks providers to design AI systems that interact directly with people so that those people are told they are dealing with an AI, unless that is already obvious to a reasonably well-informed person. Treat that as a design requirement to check with your legal team, not as legal advice.

For enterprise teams, this is where interaction design meets governance. ElixirData describes its Context OS as keeping governed context, versioned policy and decision traces in one decision layer above AI agents, and AgentGRC as keeping autonomous agent actions inside policy, with the audit trail risk teams need. ElixirClaw describes itself as an agentic OS platform to build, orchestrate and govern AI agents and automation workflows while keeping humans in control. Those are the controls an agent interface can then expose to its users.

When is interaction design not the main problem?

Good interaction design cannot fix a feature nobody needs, slow back-end systems or unclear ownership of a process. If users complete tasks but still fail to reach their goals, look at the service, the data or the policy behind the screen. Likewise, a prototype that tests well with five users does not prove the design works for every role, device and assistive technology, so keep measuring after release.

Related reading: conversational user interface best practices.

To map one decision your agents should make, and the context and policy it depends on, book a Context Workshop. See what happens in the workshop.

Frequently asked questions

What is interaction design in simple terms?

It is the design of how a product behaves in response to what people do, including feedback, navigation and error handling. The Interaction Design Foundation describes it as the design of interactive products and services.

What are the main principles of interaction design?

Goal-driven design, usability, affordances, learnability and low friction are commonly taught. They map to the ten usability heuristics from Jakob Nielsen and Rolf Molich, which NN/g says have stayed unchanged since 1994.

What are the five dimensions of interaction design?

Words, visual representations, physical objects or space, time and behaviour. Gillian Crampton Smith introduced the first four and Kevin Silver added the fifth, according to the Interaction Design Foundation.

What is the difference between interaction design and UX design?

Interaction design focuses on how a product responds to input and what feedback it gives. UX design covers the whole experience, including research, content and service, so interaction design is one part of it.

What accessibility targets apply to interaction design?

WCAG 2.2, a W3C Recommendation dated 12 December 2024, includes criterion 2.5.8, which asks for pointer targets of at least 24 by 24 CSS pixels at Level AA, with exceptions, and criterion 2.4.11, which asks that focus is not entirely hidden.

How does interaction design apply to AI agents?

The same heuristics apply, with more weight on visible status, the ability to approve or reverse actions, and clear error explanations. In the EU, Article 50 of the AI Act asks that people are told when they are dealing with an AI, unless that is obvious. Confirm requirements on the official text.

Next Steps with Interaction Design Principles

Talk to our experts about implementing compound AI system, How Industries and different departments use Agentic Workflows and Decision Intelligence to Become Decision Centric. Utilizes AI to automate and optimize IT support and operations, improving efficiency and responsiveness.

Table of Contents

navdeep-singh-gill

Navdeep Singh Gill

Global CEO and Founder of XenonStack

LinkedIn profile

Navdeep Singh Gill is serving as Chief Executive Officer and Product Architect at XenonStack. He holds expertise in building SaaS Platform for Decentralised Big Data management and Governance, AI Marketplace for Operationalising and Scaling. His incredible experience in AI Technologies and Big Data Engineering thrills him to write about different use cases and its approach to solutions.

Get the latest articles in your inbox

Subscribe Now

Interested in solving your challenges with the XenonStack team?
Get Started

Tell us about you and your priorities so we can prepare the right conversation. It takes about 2 minutes.

Please Select your Industry
Banking
Fintech
Payment Providers
Wealth Management
Discrete Manufacturing
Semiconductor
Machinery Manufacturing / Automation
Appliances / Electrical / Electronics
Elevator Manufacturing
Defense & Space Manufacturing
Computers & Electronics / Industrial Machinery
Motor Vehicle Manufacturing
Food and Beverages
Distillery & Wines
Beverages
Shipping
Logistics
Mobility (EV / Public Transport)
Energy & Utilities
Hospitality
Digital Gaming Platforms
SportsTech with AI
Public Safety - Explosives
Public Safety - Firefighting
Public Safety - Surveillance
Public Safety - Others
Media Platforms
City Operations
Airlines & Aviation
Defense Warfare & Drones
Robotics Engineering
Drones Manufacturing
AI Labs for Colleges
AI MSP / Quantum / AGI Institutes
Retail Apparel and Fashion

Continue