COMMA Shares a Service Design Framework for Corporate Websites in the AI Era
Posted September 24, 2026 by comma_service
COMMA applies service design to corporate websites, helping businesses improve clarity for users, search engines and AI through better content, structure, SEO and GEO.
Shanghai-based brand experience and service innovation firm says corporate websites increasingly need to communicate clearly not only with people and search engines, but also with AI systems.
SHANGHAI, China — September 24, 2026 — COMMA (COMMA咖墨), a Shanghai-based brand experience and service innovation firm driven by art and technology, is applying service design principles to a new challenge facing corporate websites: how organizations can make their businesses easier to understand for both human visitors and AI systems.
For many years, corporate website projects have focused primarily on two questions: whether a website looks good and whether it is easy to use.
COMMA believes a third question is becoming increasingly important:
Can search engines and AI systems accurately understand what the company does, who it serves, and why it should be considered relevant to a particular customer need?
As AI-assisted search and generative interfaces become part of how people research companies, products and service providers, the corporate website is taking on a broader role. It is no longer only a destination for visitors. It is also becoming a structured public source from which machines may interpret an organization.
From Website Navigation to Business UnderstandingTraditional website planning often begins with a sitemap.
A company decides which pages it needs — Home, About, Services, Products, Cases, News and Contact — and organizes these pages into a navigation structure.
COMMA argues that this is no longer sufficient.
A website may have a perfectly logical navigation system while still making the company difficult to understand.
For example, a business may list many services but fail to explain which are core capabilities. Case studies may show attractive visuals without identifying the client's problem, the company's actual role or the work delivered. News pages may contain years of company updates but have little connection to stable service pages.
For a human visitor, this creates friction.
For an AI system, it can create ambiguity.
COMMA therefore approaches website planning not only as information architecture, but as a network of relationships between business facts, services, cases, expertise and evidence.
A service page should connect to relevant cases.
A case should explain which capabilities were involved.
An industry solution should connect to appropriate services and supporting knowledge.
Professional articles should clarify how the company thinks, while stable corporate pages establish what the company actually does.
Together, these relationships form a more coherent digital representation of the organization.
A Website Is Becoming a Public Knowledge LayerCOMMA describes this shift as moving from a corporate website being primarily a marketing interface toward becoming part of the company's public knowledge infrastructure.
This does not mean websites should become large internal knowledge bases.
Instead, important corporate facts should be clearly stated, consistently maintained and easy to verify.
These may include:
the company's identity and business scope;
clear descriptions of products and services;
relevant industries and customer scenarios;
project cases and the company's role in those projects;
professional viewpoints and methodologies;
official contact and organizational information;
relationships between different websites, brands or business units.
When these elements are fragmented or inconsistent, both users and automated systems may struggle to build a clear picture of the organization.
COMMA therefore recommends treating website content as an ongoing governance task rather than a one-time copywriting exercise.
GEO Is More Than Adding AI KeywordsThe growth of Generative Engine Optimization, or GEO, has created another common misunderstanding.
Some organizations treat GEO as a new form of keyword optimization.
They add more AI-related terms, create large FAQ sections or implement technical files and structured data, expecting these changes alone to improve visibility in AI-generated answers.
COMMA takes a different view.
Technical foundations are important, including crawlability, semantic HTML, heading structure, internal linking, sitemaps, canonical URLs and structured data.
But these elements cannot compensate for unclear business information.
The more fundamental question is:
If an AI system needs to answer a real customer question, does the company provide enough clear, relevant and verifiable information to support that answer?
A statement such as "we provide digital transformation solutions" may be technically readable but still too vague to be useful.
A stronger website explains what kinds of problems the company addresses, what it actually delivers, which customers or industries it serves, and which cases or evidence support those claims.
In this sense, GEO is closely connected to content architecture, corporate identity and evidence management — not only SEO.
Service Design Connects User Needs, Business Logic and TechnologyCOMMA uses service design as the underlying method for this work.
Instead of beginning with individual pages or visual styles, the process starts with the questions users are trying to answer.
Who is visiting?
What are they trying to understand?
What evidence do they need before making a decision?
Which page or touchpoint should answer each question?
What should happen next?
The same questions can then be extended to search engines and AI systems.
This allows website strategy, content, user experience, technology, SEO and GEO to be treated as one connected system rather than separate optimization tasks.
According to COMMA, this approach is particularly relevant for B2B companies, complex service organizations, technology companies, group enterprises and companies undergoing brand or business transformation, where the value of the business cannot be explained through a single product page.
Designing for People and Machines at the Same TimeCOMMA does not view AI readability as a replacement for visual design or user experience.
A corporate website still needs to create trust, communicate brand character and help users complete tasks efficiently.
The change is that design now has an additional audience.
People often understand hierarchy through typography, imagery, color, motion and spatial relationships.
Machines rely more heavily on explicit language, semantic structure, consistent entities, links and contextual relationships.
The next generation of corporate websites therefore needs to work at both levels.
They need to be visually understandable to people and structurally understandable to machines.
For COMMA, this represents a broader shift in digital experience design:
the objective is no longer simply to build a website that can be seen, but to create a digital source that can be correctly understood.
More information about COMMA's digital experience, corporate website, SEO and GEO services is available at:
Related perspective:
About COMMACOMMA (COMMA咖墨) is a Shanghai-based brand experience and service innovation firm driven by art and technology, with service design as its core methodology.
Founded in 2010, COMMA works across four connected areas: brand design and experience transformation, digital experience and product development, events and spatial experience, and AI applications and service innovation.
Its digital experience practice covers corporate websites, website ecosystems, WeChat mini programs, digital products, ongoing operations, SEO and Generative Engine Optimization (GEO).
Official websites:
| Contact Email | [email protected] |
| Issued By | COMMA 咖墨 |
| Website | COMMA |
| Country | China |
| Categories | Advertising , Arts , Design |
| Tags | comma , comma , service design , corporate website , ai website , digital experience , generative engine optimization , website design |
| Last Updated | September 24, 2026 |