Frequently asked questions.

Word and Google Docs are page-based tools — every document is a standalone file. DITA is topic-based: you write a topic once and reuse it across 50+ deliverables. Formatting is automated through stylesheets, publishing to PDF, HTML5, mobile, and API happens simultaneously, and every version is tracked with full branching and baselining. Word can't do any of this at scale.

For teams producing 100+ pages of regulated or multi-channel content, DITA typically pays for itself within the first release cycle through reduced duplication and translation costs.

More relevant than ever — and for a new reason. While docs-as-code (Markdown + Git) works well for developer documentation, DITA remains the standard where complex hardware, medical devices, defense (S1000D interoperability), insurance policy assembly, and compliance traceability are non-negotiable.

But the real shift in 2026 is AI-readiness. Organizations deploying chatbots and RAG pipelines are discovering that unstructured content (Markdown, Word, PDF) produces retrieval precision around 25–35%. DITA's typed topics, semantic markup, and metadata-rich architecture push that to 85%+. Structured content is no longer just a publishing advantage — it's an AI advantage. See our AI Readiness Guide for the full scoring framework.

Git-based (Docs-as-Code): Content is stored in a Git repository. Writers use VS Code or a text editor. Publishing is triggered by CI/CD pipelines (GitHub Actions, Jenkins). Best for developer-heavy teams comfortable with Git, pull requests, and Markdown/DITA source files.

CCMS: A Component Content Management System (Heretto, IXIASOFT, Paligo, Tridion Docs) provides a web-based authoring UI, granular component reuse tracking, translation management, built-in review workflows, and role-based access control. Best for large teams with mixed technical skill levels, heavy localization, and regulatory compliance requirements.

The trade-off: Git is cheaper and more flexible; a CCMS provides visibility, governance, and localization integration that Git alone can't match. Many of our clients use both — a CCMS for authoring and review, with Git-triggered pipelines for publishing. We help you choose based on team size, content volume, localization needs, and compliance requirements.

Yes. We help teams adopt XDITA (XML-based LwDITA), HDITA (HTML-based), and MDITA (Markdown-based) for simpler content needs. We also build hybrid workflows where developers write in Markdown and tech writers author in full DITA, with automated merge at build time via the DITA-OT. This lets each team use the tool they prefer while maintaining a single structured output.

The DITA Open Toolkit (DITA-OT) is the open-source engine that transforms DITA XML into output formats — HTML5, PDF, EPUB, etc. A plugin extends or customizes that transformation. For example:

• A branded PDF plugin applies your company's fonts, colors, cover page, and headers/footers to the PDF output
• A custom HTML5 plugin generates responsive web pages with your navigation, search, and design system
• A JSONL exporter plugin produces chatbot-ready chunks with metadata for vector store ingestion

Every plugin is a versioned, testable artifact in your Git repository — not a one-off configuration buried on someone's laptop. We build, test, document, and maintain custom plugins for our clients. See our Automation Guide for details on how plugins fit into the CI/CD pipeline.

We score your content on 7 dimensions: content typing, metadata coverage, section addressability, vocabulary control, semantic markup richness, reuse architecture, and output pipeline maturity. Most organizations start at 2–4 out of 14. Our clients end up at 9–12.

The practical work involves: converting content to typed topics (concept, task, reference), enriching every topic with machine-readable metadata from a controlled vocabulary, adding section IDs for precise chunking, and building a publishing pipeline that produces chatbot-ready JSONL alongside HTML5 and PDF.

The result: your AI chatbot retrieves the right answer at the right level for the right audience — with 85%+ precision instead of 25–35%. See the full framework in our AI Readiness Guide.

Instead of running DITA-OT manually on a laptop, every Git commit triggers an automated pipeline: schema validation → link checking → terminology enforcement → multi-format build (HTML5 + PDF + JSONL) → deploy to staging → promote to production. The entire build runs in a Docker container so every environment produces identical output.

We configure pipelines on GitHub Actions (our recommended starting point), Azure DevOps, Jenkins (for air-gapped defense environments), and GitLab CI. A typical pipeline builds 200 topics into 3 output formats in under 3 minutes — compared to 15–40 minutes manually.

See our Automation Guide for the full 8-stage pipeline, Dockerfile examples, and GitHub Actions YAML.

Yes — and we strongly recommend it. A Docker image locks your entire toolchain (DITA-OT version, Java version, custom plugins, corporate fonts, validation tools) into a single versioned artifact. Every CI build pulls the same image. No more "it works on my machine" or configuration drift between environments.

This matters most in regulated industries where you need to prove that the build environment is reproducible and auditable. We build and maintain Docker images for our clients, including custom plugins baked in and corporate fonts pre-installed for PDF generation.

Absolutely. Customizing the DITA-OT PDF plugin (XSL-FO) and HTML5 output is one of our most requested services. We match your corporate branding pixel-perfectly — cover pages, headers/footers, TOC styling, cross-reference formatting, index generation, and accessibility compliance (Section 508 / WCAG 2.1). We also build custom plugins for EPUB, Markdown developer portals, and chatbot-ready JSONL export.

We work across regulated and documentation-heavy industries:

• Government & Defense: S1000D programs, MIL-STD compliance, IETM/IETP publishing, civilian agency content modernization
• Financial Services: Compliance documentation, regulatory filings, multi-channel policy publishing
• Insurance: Policy document assembly, state-variant management, regulatory compliance
• Life Sciences: IFU documentation, regulatory submissions, multi-language labeling
• Transportation: Service manuals, parts catalogs, S1000D interoperability across aviation, rail, and automotive
• Technology: Product documentation, API references, datasheet automation, developer portals

If your organization produces complex, regulated, or multi-channel documentation, we can help — regardless of industry.

We are platform-agnostic. We have deployed and integrated Heretto, IXIASOFT CCMS, Paligo, RWS Tridion Docs, Vasont, and Astoria. Our approach: define your requirements (team size, content volume, localization needs, deployment preference, budget), shortlist 2–3 platforms, then run a structured proof-of-concept with your actual content — not canned demos. We load your topics, configure your workflow, run a full publish cycle, and score each platform on a weighted matrix.

A typical 2,000-page migration takes 3–6 months end-to-end: 2–3 weeks for the content audit, 2–4 weeks for information architecture design and pilot conversion, 6–12 weeks for batch conversion, and 2–4 weeks for validation and stylesheet tuning. Volume, source format complexity (Word vs. FrameMaker vs. InDesign), and the number of languages all affect the timeline.

We always start with a free 20-page sample conversion so you can evaluate quality before committing. See the full process in our Migration Playbook.

Yes — we recommend it. A typical pilot involves converting 20–50 pages of your most representative content, building the publishing stylesheets (PDF + HTML5), and validating the output against your brand standards. The pilot proves the approach, surfaces edge cases early, and gives your team hands-on experience before scaling. Pilot engagements typically run 4–6 weeks and cost a fraction of the full project.

Yes. We build custom DITA training curricula tailored to your team's roles and experience levels. A typical program includes topic-based authoring, metadata and taxonomy, DITA map architecture, content reuse (conref/keyref), review workflows, and hands-on labs with your actual content. We've built curricula with 70+ topics, 10+ learning modules, and structured labs — delivered as instructor-led sessions, self-paced DITA-based courseware, or both. Training is included in most migration and CCMS implementation engagements; it's also available as a standalone service.

See the Enterprise DITA Training Curriculum case study for an example of what we've shipped.

We offer three models: Project-based (fixed scope, fixed price — ideal for migrations, CCMS implementations, and plugin development), Staff Augmentation (embed our DITA/XML specialists in your team for 3–12 months), and Managed Services (ongoing SLA-based support for CCMS administration, publishing pipeline operations, and content engineering). We always provide a detailed quote after a free discovery call — no surprises.