Self-assessment tool

There is no change management process in place for the technical documentation.

The product content developer acts as the information content creator. Thus, the content creator is someone whose main focus is not on technical documentation but on other tasks, and who may lack experience with proper technical documentation principles.

The requirements are based on purchasing and R&D, and lack input from other relevant stakeholders and consideration of user needs.

There are no structural requirements for the content supply. Content creators work with what they are given, without the possibility or experience to request another format or workflow.

The requirements are based on purchasing and R&D, and lack input from other relevant stakeholders.

There are no structural requirements for the content supply. Content creators work with what they are given, without the possibility or experience to request another format or workflow.

Source information is provided in a structured and bidirectional way.

Requirements are based on purchasing, R&D, aftermarket, sales, and legal requirements. The needs of users are known and taken into account for user information creation.

There is a clear process for content creation triggers (new/change/update) including relevant roles and stakeholders.

Requirements are set to enable a seamless flow of content from supplier to user - the process is clear and works.

A user-centric information architecture is in place.

Content is supplied from various data sources and created to be not only user-centric but also machine-centric.

Field feedback is communicated seamlessly via the content structure back to the supplier - the process is clear and works.

User-centric data is used in new development cycles.

Content is automatically supplied from various data sources.

The product content developer and/or content creator manually aligns input data by calling stakeholders (example: technical data).

The product content developer creates data per machine by copying versions from similar machines based on their experience (example: visualisation, e-plan).

Product data is changed on the fly during commissioning / installation due to customer wishes or late technical changes.

The content creator manually gathers and consolidates siloed information across purchasing, R&D, and project-specific sources based on their experience.

The product content developer creates data per machine by copying versions from similar machines based on their experience (example: visualisation, e-plan).

The content creator mostly gathers and consolidates siloed information manually across purchasing, R&D, and project-specific sources based on explainable and documented rules.

For some data, consistent source information is available that can be imported into the technical writing environment.

The product content developer creates data per machine by copying versions from similar machines based on explainable and documented rules (example: visualisation, e-plan).

Processes define the source for information within the lifecycle of a product.

The product content developer creates data per machine based on templates and documented rules (example: visualisation, e-plan).

Feedback loops guarantee that information from the shop floor and the installation site at the customer finds its way to the data set in the database.

Interfaces ensure handover of data from one phase in the lifecycle to the next.

The product content developer creates data per machine based on templates and documented rules (example: visualisation, e-plan).

Verification routines ensure data consistency.

Verification routines ensure data consistency.

Ontologies / Knowledge graphs map metadata between different source systems.

A semantic layer enables leveraging knowledge and data from several source systems to find answers to questions that were previously impossible, or required a huge manual effort, to create.

Interfaces guarantee dynamic updates of information, such as technical data, on the website.

Content creation is performed in silos, for instance in different departments or R&D sites. There is no joint process for content creation in place.

Content creation differs depending on who creates the content.

Technical information is created by non-specialists in technical writing.

Content is manually structured using authoring tools such as Word.

A basic documentation process is established based on the experience of the content creator. There is no consistent written style guide.

Content is structured in information units.

A rule-based process is documented in a style guide and consistently followed.

Content is future-proofed from a format perspective using XML, and metadata is applied.

Content is findable and reused from a content creation perspective.

Content is organized according to the needs of users in their situation and the legal requirements for providing user information (for example, digital product pass, VDI 2770).

Content granularity is defined at topic level based on information type (concept, task, reference, troubleshooting).

Content is integrated from multiple sources (structured).

Controlled language is applied.

Metadata concept set up for digital, user based content delivery using facetted search and AI based search.

Data mapping and compilation across different data sources is automated through import from source systems.

Content is created as information objects (nuggets) for topic-based content delivery, readable by both people and machines.

The content is represented and divided into unique and reusable topics prepared for automated content creation based on feature lists or similar.

Compilation and enrichment are automated across multiple sources and information objects.

AI-based sources support and improve the content creation process.

The values for technical data may vary in the set of documents

The same tasks and functions are written slightly different in different documents so that it is hard to recognize tasks and functions that are the same for different products

Figures are a broad mixture of unrefined fotos and graphics from technical drawings

Technical data are checked for consistancy manually

The same tasks and functions are written slightly different in different documents so that it is hard to recognize tasks and functions that are the same for different products

Systamtically created figures from technical drawings are used when available. Other graphic sources are used in addition

Technical data are checked for consistancy manually

Content should be created using the same style and terminology but the consistancy is checked manually

Systamtically created figures from technical drawings or CAD data used when available. Other graphic sources are used in addition

Reuse ensures consistancy of technical data

Text and graphics are used so that the information is easy to consume with a good text - graphic concept

Technical data are consistant and up to date due to import from source system

Tool-based checks ensure text and graphic quality

Communication with users in different roles is in focus for the information concept

AI-based chats bring user communication to the next level.

Automaitzed QA checks based on well defined and machine readible rules ensure consistant quality

Publishing is product- and document-centric.

Publication practices vary by product or project.

Content is published without structure or metadata.

Publishing remains product- and document-centric (large PDFs).

Publication types are defined and stored consistently.

Basic governance is applied to formats and templates.

Multiple document types are produced from a single source.

Publications are delivered through multiple digital channels.

Content is indexed with metadata and taxonomy.

Multiple information products are created from a single source.

Publication decisions are based on audience, role, and purpose.

Metadata, lifecycle, and reuse rules are governed.

Information is managed as reusable content objects.

Content is published in any format or interface (web, portal, API).

Publication adapts to who is using the content, and to what, when, why, and how.

AI assists with enrichment, validation, and channel adaptation.

No explicit user action is required to access content.

Publication becomes continuous and responsive.

AI-driven decisions optimize timing, format, and channel.

Content adapts automatically to changing contexts.

The information is sent by email to the customer/end-user on demand.

The organization hands over documents as they come via unregulated portals such as SharePoint folders or Dropbox.

The organization sends out printed documents if requested.

The organization provides printed documents with the product; for example, boxes of printed documents in file folders are shipped together with the machine.

The organization provides PDF files via digital portals such as customer portals or websites.

The organization provides individually created online help, such as the old CHM format, for the UI of a machine or software.

The organization processes the PDF file from the multichannel publishing pipeline for customer portals.

The organization processes the PDF file from the multichannel publishing pipeline for printed documentation.

The organization processes the HTML documentation from the multichannel publishing pipeline for web and UI.

The organization processes information in special formats using tools prepared for DCC or Digital Twin.

The organization processes the PDF file from the multichannel publishing pipeline for customer portals, using faceted search and AI-based search at document level (e.g., basic requirements, machine directive, DPP).

The organization processes the PDF file from the multichannel publishing pipeline for printed documentation.

The organization processes the HTML documentation from the multichannel publishing pipeline for web and UI, using faceted search and AI-based search at topic level.

The organization makes information of any format available per role, customer, or machine state, using AI-based search and chat.

The organization combines user information with 3D data for training applications.

The organization combines user information with spare parts information (3D-based) and the e-shop.

The organization integrates user information into AI-based assistance systems of any kind.

The organization provides user information to AI-agent-based applications, where AI uses the information rather than humans.

Content is integrated into an AI-agent-based product that searches for content when required, consumes it for AI-based processes, and informs the user only when a human-based task is needed or for informational purposes.

The system evaluates the interaction of users (AI agent and/or human) and sends information to the creation tool for more context, missing information, or error-prone information.

Consumption is document- or regulation-driven. In general, users get product information from sources other than documentation, such as support or training, while installation/commissioning personnel are on site. They ask for specific information only if needed.

Users must know where information exists.

There is little or no insight into user needs or context.

Document-based usage is still dominant.

Uniform documentation types and templates are used.

There is limited understanding of how content is actually used.

Users locate content through navigation or search.

Findability is improved but personalization remains limited.

Use is still largely document-driven.

Accurate and relevant information is available across the value chain.

Users pull information based on task and role.

Analytics and behavioral insights are used at an early stage.

Digital-first information usage is in place.

Delivery is contextual, based on role, task, and situation.

The system is scripted to deliver information when triggers occur.

Information is used on a task basis and is embedded in the workflow.

AI-based search and chatbots help users find information in text, graphics, drawings, PDF, or HTML.

Information is proactively delivered to users.

Agentic AI solutions with integrated search and chatbot: AI recommends actions, autonomous agents execute feasible tasks, and humans receive information on results and remaining work items.

The system continuously learns from usage and feedback.

There is no regular feedback loop; feedback occurs only in escalation mode, if an accident takes place or the customer explicitly insists on having a feature or function documented.

If feedback is given, it is difficult to connect it to a specific document and version.

There is no regular feedback loop; feedback occurs only in escalation mode, if an accident takes place or the documentation does not match the contract.

If feedback is given, it is possible to connect it to a specific document version.

There is no regular feedback loop; feedback occurs only in escalation mode, if an accident takes place or the documentation does not match the contract.

If feedback is given, it is possible to connect it to a specific content version, and a systematic update of all instances where the information occurs is possible.

A process-based content feedback loop, based on R&D change management or manual feedback, is in place.

It is possible to connect feedback to a specific product, content, and role.

A tool-based content feedback loop is integrated into the IT landscape and forms part of the content creation process.

It is possible to connect feedback to a specific product, content, role, and context.

In addition to active feedback from users, analytics track content usage and use user behavior for recommendations.

If a user searches for content that is not available, a request for creating that content is automatically generated.

When an AI chatbot or search tool cannot fully answer a user's query, it automatically generates a content creation request with associated job data—including time, methodology, resources, and context extracted from the user's natural language.

The organization provides the absolute minimum of required documentation, based on the misleading assumption that avoiding a systematic documentation process saves costs.

There is a lack of knowledge of impact.

Too much time is spent searching for the right information, and the information found is often not sufficient to understand a function or perform a task.

The organization complies with regulatory requirements.

There is a lack of knowledge of impact.

Too much time is spent searching for the right information.

Productivity in content creation is optimized.

Content is findable from a content creation perspective.

Translation costs are decreased.

Consistency is increased.

Dependency on individuals is reduced.

Availability is increased.

Content reliability is increased.

Time spent finding content is decreased.

The link between content and task execution is improved, reducing frustration.

Information availability is increased.

Relevance for the user is increased.

Time spent on information search is minimized.

Task execution efficiency is improved.

Right content is available directly.

The process is optimized from a user perspective.

The content is ready to be part of AI-driven products.

Customer engagement is increased.

Information strategy is missing entirely.

A basic, product-centric information strategy driven by regulatory requirements is in place.

A basic set of organized and defined document types exists.

AI-based information processing focusses on PDF file handling.

A systematic, product-centric information strategy driven by regulatory requirements is in place.

AI-based information processing focusses on PDF file and HTML format handling.

A basic, user-centric information strategy is in place.

There is a good understanding of the traditional elements in the content value chain, and well-established solutions for system architecture, automatic functionalities, and information formats are in place.

AI-based information processing focusses on PDF file and HTML format handling.

There is optimal reuse across the value chain, such as metadata, terminology, and single sourcing of information.

Information is connected to humans and machines as user, product, and context.

The information strategy is seen as an opportunity to increase productivity and customer experience.

Strategic use of AI in the creation and use of information is planned or has started.

The information strategy is part of the business and operations strategy.

Structured and unstructured information data are treated as a critical business resource.

Technical information is an integral part of the strategy for AI-based products and the operational AI strategy.

An information management concept is missing.

Information is structured manually.

Metadata use is limited.

Document categories defined (installation, training, user, services documentation)

Clear guidelines and standards exist for the creation of content.

Structured content, including metadata and XML, is applied in content creation.

Information is structured based on the whole content value chain and product lifecycle.

Metadata based on both product and user is applied.

There are no exceptions; all information and data are covered and managed.

All decisions are taken with information as an important stakeholder.

Information acts as a business enabler.

Information management is integrated into the company DNA.

Information acts as a business enabler.

A paradigm change from humans to machines as consumers of information is ongoing.

Structured and unstructured information are being used seamlessly in an integral way.

Multiple different processes are used, based on individuals or projects.

Information and documentation are part of many unrelated processes with high variability.

There is no end-to-end visibility.

Documentation processes are defined and followed by the technical writer.

Structured formats, templates, and basic guidelines are used.

Processes are stable and repeatable across the basic elements of the content value chain.

Clear inputs, outputs, and handovers are defined.

Structured, compliant workflows are supported by tools.

The end-to-end information flow is managed and monitored.

The right information is delivered at the right time through governed processes.

Ownership, metrics, and continuous improvement are defined.

AI begins assisting with process analysis and optimization.

Workflow structure and operational processes are driven by information needs.

Automatic rules and triggers manage most flows.

Human-in-the-loop exception handling is used.

AI assists decision support, routing, and prioritization.

Workflows automatically detect information needs and initiate creation, adaptation, and delivery.

Processes are adaptive and continuously optimized.

AI continuously learns from usage, outcomes, and feedback.

The operating model is adapted to AI-based, agentic processes.

Individuals define and create information independently.

Technical information roles or ownership are undefined.

Work is dependent on personal expertise.

Roles and responsibilities are defined within the documentation function.

Roles are primarily generalist.

Documentation activity is centralized, with limited integration into other functions.

Technical information activity is connected to R&D, training, and service.

Specialization is increasing (information specialists, architects, tool experts).

The organization is multi-skilled and aligned across the value chain.

Skill development is driven by growing complexity.

The organization is multi-skilled and aligned across the value chain.

Cooperation replaces functional silos.

Product, information, and lifecycle ownership are clearly defined.

Make/buy decisions and interfaces to external specialists are established.

Information specialists and architects manage information and data models.

There is strong collaboration between business, IT, and information functions.

The organization is federated, with shared standards and governance.

The operating model is open, with external partnerships.

The organization is fully multi-skilled and adaptive.

Roles focus on governance, optimization, and innovation rather than execution.

There is a fluid ecosystem of internal and external capabilities.

AI stewardship and accountability are clearly defined.

There is a fix mentality, and work is executed when needed upon product release, customer complaint, or update.

Technically skilled people create technical information as part of development.

People are skilled in communication, technical writing, and technical subjects.

Technical information is approached as a specialism, but with a manual "Word / InDesign" way of working, such as manually laying out each document.

The focus is on document structure, setup, and completeness rather than on usability and applicability.

Topic-based authoring is used.

The level of technical information skills and seniority is high.

New specialist roles emerge (e.g., information architects).

Controlled language is applied.

New specialist roles emerge (e.g., user experience, terms and terminology).

Information architecture is established.

Data mapping and compilation are performed.

User-centric specialists work alongside technical information specialists.

Digital information skills are needed.

Information architects are employed.

Information control and deployment quality are ensured.

Prompting and other skills are used to operate and optimise AI-based information processes.

Tools that do not create extra costs for technical documentation are used, e.g., freeware graphic tools.

Office applications are mostly used without conditions, variables, or reusable text blocks.

Unspecific AI tools are used unsystematically.

The technical writing environment is largely manual.

Office applications, FrameMaker, InDesign, and Oxygen are used without a CCMS.

A limited set of templates is used, but documents are often created by copying a similar document.

The only help for reducing translation cost is translation memory (TM) at the translation agency, or out-of-the-box AI-based translation without terminology support or similar.

A CCMS is implemented.

Interoperability is in place from a content creation perspective.

Content creation is the main objective.

AI functions are integrated into tools such as the CCMS.

Content delivery platforms, including feedback, are used.

End-to-end information flows are supported by middleware.

A Simplified Technical English checker is used.

IoT and connectivity are used, with triggers based on events.

Publications are available on demand in any application or channel.

User and customer experience is the main objective (UX/UI).

AI tools are integrated into the content value chain.

User and customer experience and business goals are the main objectives.

Agentic AI tools are integrated into the content value chain.

Emerging technologies can continuously be implemented.