Knowledge Management Process
Knowledge base structure, article creation standards, review cycles, AI-powered search, documentation standards, and wiki contribution guidelines.
Knowledge Base Structure & Taxonomy
Our knowledge base is organised into a hierarchical taxonomy designed for both human browsing and AI-powered search.
| Category | Sub-Categories | Article Types | Audience |
|---|---|---|---|
| Business Processes | Client onboarding, service delivery, incident mgmt, change mgmt, project delivery, sales, procurement, QA, security | Process documents, policies, procedures | All ASI staff |
| Technical - Infrastructure | Networking, servers, storage, virtualisation, backup/DR, cloud (Azure, AWS, GCP) | How-to guides, troubleshooting, runbooks | Engineers |
| Technical - Security | Endpoint, email security, SIEM, firewalls, identity, vulnerability mgmt | Configuration guides, incident playbooks, hardening guides | Security & engineering teams |
| Technical - Applications | M365, LOB applications, SaaS integrations, custom applications | Configuration, troubleshooting, user guides | Engineers & service desk |
| Client-Specific | One space per client (restricted access) | Environment docs, runbooks, known issues, contact lists | Assigned team members only |
| Service Desk | Common issues, password resets, software installation, VPN, printing | Step-by-step resolution guides, scripts | L1/L2 service desk |
| Training | Onboarding, certifications, product training, soft skills | Training materials, course links, certification guides | All staff |
| Internal Operations | HR policies, finance, travel, office admin, IT self-service | Policies, how-to guides, FAQs | All staff |
Article Creation Standards
Article Template
📄 Knowledge Article Template
Article ID
[Auto-generated: KB-NNNNNN]
Title
[Clear, descriptive title. Start with action verb for how-to articles. Max 80 characters.]
Category / Sub-Category
[Select from taxonomy above]
Article Type
[How-To / Troubleshooting / Reference / Policy / Runbook / FAQ]
Audience
[L1 Service Desk / L2 Engineer / L3 Specialist / All Staff / Client-Facing]
Summary
[1-2 sentence summary. This appears in search results.]
Symptoms / Problem Description
[What does the user experience? Error messages, screenshots.]
Resolution / Steps
[Numbered step-by-step instructions. Include screenshots where helpful. Note any prerequisites.]
Related Articles
[Links to related KB articles]
Tags
[Relevant keywords for search: product names, error codes, technologies]
Author
[Name]
Review Date
[6 months from creation or last review]
Writing Standards
- Language: Australian English. Write in clear, simple language. Aim for reading level Year 10 or below.
- Voice: Active voice, second person ("You can..." not "The user should..."). Direct and concise.
- Steps: Number all procedural steps. One action per step. Start each step with a verb.
- Screenshots: Include screenshots for UI-based procedures. Annotate with red boxes/arrows. Ensure no PII is visible.
- Code/Commands: Use code blocks for commands, scripts, and configuration snippets. Test all commands before publishing.
- Links: Link to related articles. Never duplicate content; reference the canonical article instead.
- Version-specific: Clearly state which software version the article applies to. Update when new versions are released.
- Accessibility: Use alt text for images. Ensure content is navigable via keyboard.
Review & Update Cycle
| Article Type | Review Frequency | Reviewer | Trigger for Early Review |
|---|---|---|---|
| Process documents (this wiki) | Every 6 months | Process Owner | Process change, audit finding, major incident |
| Technical how-to / troubleshooting | Every 6 months | Subject Matter Expert | Product update, resolution failure, user feedback |
| Client-specific runbooks | Every 3 months | Assigned SDM/Engineer | Environment change, incident |
| Security playbooks | Every 3 months | CSIRT Lead | Security incident, new threat vector |
| Service desk scripts | Every 6 months | Service Desk Lead | FCR decline, user feedback |
| HR/internal policies | Annually | People & Culture | Policy change, legislative update |
Review Workflow
- Automated reminder sent to reviewer 2 weeks before review date
- Reviewer checks article accuracy, completeness, and relevance
- If still accurate: update "Last Reviewed" date and extend review date
- If updates needed: make changes, submit for peer review, then publish
- If obsolete: archive the article (do not delete; mark as "Archived" with date and reason)
Knowledge Sharing Culture
Knowledge management is everyone's responsibility. We foster a culture of sharing through:
KB Contribution KPI
Every engineer is expected to create or update at least 2 KB articles per month. Tracked in ServiceNow and included in performance reviews.
Knowledge Champions
Each team has a designated Knowledge Champion who ensures articles are created for recurring issues, reviews team content quality, and mentors new contributors.
Post-Incident KB
Every P1/P2 incident and every PIR must result in at least one new or updated KB article. This is a mandatory step in the incident closure checklist.
Weekly KB Digest
A weekly email digest highlights the top 5 most-used articles, new articles published, and articles flagged for review. Distributed every Monday morning.
AI-Powered Knowledge Search
ASI AI Sentinel integrates with our knowledge base to provide intelligent, context-aware search capabilities:
- Natural language queries: Users can search using natural language questions (e.g., "How do I reset MFA for a client user?") and AI returns the most relevant articles
- Automatic KB suggestion on tickets: When a new incident is created, AI analyses the description and attaches the top 3 matching KB articles to the ticket. Engineers see these in the ticket sidebar.
- Semantic search: AI understands synonyms and related concepts. Searching for "email not working" also returns results about "Outlook issues", "Exchange problems", and "M365 outage".
- Gap detection: AI identifies frequently searched terms that return no results, flagging these as knowledge gaps for content creation. Reviewed weekly by Knowledge Champions.
- Auto-summarisation: For long runbooks and process documents, AI generates a brief summary that appears at the top of search results.
- Usage analytics: AI tracks which articles are most useful (measured by "resolved with this article" clicks) and surfaces these higher in search results.
Documentation Standards
Naming Conventions
| Document Type | Naming Format | Example |
|---|---|---|
| KB Article | KB-NNNNNN (auto-generated) | KB-004521 |
| Process Document | [Process Name] - v[X.Y] | Incident Management Process - v4.0 |
| Client Runbook | [Client]-RB-[System]-[Topic] | AcmeCorp-RB-Azure-DR-Failover |
| Architecture Diagram | [Client]-ARCH-[System]-[Date] | AcmeCorp-ARCH-Network-202603 |
| Meeting Minutes | [Meeting Type]-[Client]-[Date] | QBR-AcmeCorp-20260315 |
Storage Locations
| Content Type | Primary Location | Access |
|---|---|---|
| Business process documents | This Wiki (wiki.asiaisolutions.com.au) | All staff |
| Technical KB articles | ServiceNow Knowledge Base | All staff (some articles client-visible) |
| Project documentation | Confluence (project spaces) | Project team + stakeholders |
| Client documentation | Confluence (client spaces) + ServiceNow CMDB | Assigned team only |
| Architecture diagrams | Confluence + draw.io (source files in SharePoint) | Engineering teams |
| Meeting recordings | Microsoft Teams / SharePoint | Meeting attendees |
| Financial/HR documents | SharePoint (restricted) | Finance/HR teams |
Wiki Contribution Guidelines
Who Can Contribute
- All staff can submit articles for review via ServiceNow "KB Article Request"
- Engineers and above can create and edit articles directly (subject to peer review)
- Process Owners are responsible for their process document pages
- Knowledge Champions can approve and publish articles in their domain
- QA Lead has global editorial authority for standards compliance
Publication Workflow
StartDraft Article Created
Step 1Author Self-Review (checklist)
Step 2Peer Review (technical accuracy)
GateKnowledge Champion Approval
Step 3Published to Knowledge Base
LiveArticle Available & Searchable
Author Self-Review Checklist
- Title is clear, descriptive, and under 80 characters
- Correct category and sub-category selected
- Summary is concise and appears in search results
- Steps are numbered, start with a verb, one action per step
- Screenshots included and annotated (no PII visible)
- Commands and scripts tested and verified working
- Related articles linked
- Appropriate tags added for searchability
- Spelling and grammar checked (Australian English)
- Access level set correctly (internal only vs. client-visible)
- Review date set (6 months from creation)
Knowledge Management Metrics
2,847
Total KB Articles
89%
Search Success Rate
62%
KB Attachment Rate (Tickets)
94%
Articles Within Review Date
127
New Articles / Month
18
Knowledge Gaps Identified
Metric Definitions
| Metric | Definition | Target | Current |
|---|---|---|---|
| Search Success Rate | % of searches that result in the user clicking on an article (vs. refining or abandoning search) | ≥ 85% | 89% |
| KB Attachment Rate | % of resolved tickets that have at least one KB article linked | ≥ 60% | 62% |
| KB-Assisted FCR | % of first-contact resolutions where a KB article was used | ≥ 50% | 54% |
| Article Currency | % of articles reviewed within their scheduled review date | ≥ 90% | 94% |
| Knowledge Gap Backlog | Number of identified knowledge gaps without a corresponding article | ≤ 20 | 18 |
| Contribution Rate | Average KB articles created/updated per engineer per month | ≥ 2 | 2.3 |