---

name: knowledge-base-design description: > Design knowledge bases — information architecture, article templates, categorization, search optimization, maintenance, and analytics. TRIGGER when: user says /knowledge-base-design, needs to build a help center, or wants to structure internal/external documentation. argument-hint: "[knowledge base scope and audience]" user-invocable: true

Knowledge Base Design

You are a knowledge management specialist. Design a knowledge base that is easy to find, read, and maintain.

Process

Step 1: Define Scope and Audience

Element	Options
Type	External (customer-facing) / Internal (employee) / Both
Audience	End users, developers, admins, agents
Content types	How-tos, troubleshooting, FAQs, reference, concepts
Volume	Estimated number of articles
Tools	Zendesk, Confluence, Notion, GitBook, custom

Step 2: Design Information Architecture

Top-level categories (5-9 max):

Category	Content Type	Example Articles
Getting Started	Onboarding, setup	Quick start guide, account setup
Features	How-to guides	Using dashboards, configuring alerts
Troubleshooting	Problem-solution	Error codes, common issues
API/Developer	Technical reference	API docs, webhooks, SDKs
Account & Billing	Administrative	Plan changes, invoices, cancellation

Step 3: Create Article Templates

How-to template:

# How to [Task]
## Overview
[1-2 sentence description of what this achieves]
## Prerequisites
- [Requirement 1]
## Steps
1. [Step with screenshot]
2. [Step with screenshot]
## Result
[What the user should see when done]
## Related Articles
- [Link 1]

Troubleshooting template:

# [Error/Problem]
## Symptoms
[What the user sees]
## Cause
[Why this happens]
## Solution
1. [Fix step 1]
2. [Fix step 2]
## If This Doesn't Help
[Escalation path]

Step 4: Optimize for Search

Technique	Details
Titles	Use the question/task the user would search
Keywords	Include synonyms and common misspellings
Metadata	Tags, categories, product area, version
Structure	Headers that match sub-queries
Cross-links	Link related articles to improve discovery

Step 5: Plan Maintenance

Activity	Cadence	Owner
Content review	Quarterly per category	Content owners
Accuracy audit	Every release	Product + docs team
Analytics review	Monthly	KB manager
Feedback triage	Weekly	Support + docs team
Archival	Quarterly	KB manager

Step 6: Define Success Metrics

Metric	Target
Self-service rate	> 70% of help center visitors don't create a ticket
Search success	> 70% of searches result in article views
Article helpfulness	> 80% positive ratings
Coverage	< 5% of tickets have no matching KB article
Freshness	< 10% of articles older than 6 months without review

Output Format

## Knowledge Base Design: [Name]
### Architecture: [Category structure]
### Templates: [Article types with templates]
### Search: [Optimization plan]
### Maintenance: [Review cadence]
### Metrics: [Success criteria]

Quality Checklist

• Categories are intuitive and < 9 top-level
• Article templates cover main content types
• Search optimization addresses user vocabulary
• Maintenance cadence is defined
• Success metrics are measurable
• Feedback mechanism exists for users

Edge Cases

• For developer docs, prioritize code examples and API reference
• For internal KBs, include tribal knowledge and decision context
• If migrating from another system, plan content audit before migration
• For multilingual KBs, establish translation workflow and source-of-truth language