How any team can create, populate, and manage a searchable Knowledge Base — using Just247Pipes’ visual pipeline designer, no coding required.
The Knowledge Problem Every Business Faces
Your company has knowledge. Lots of it. Product documentation, FAQ pages, internal policies, support articles, onboarding guides — the list goes on. But here’s the problem: that knowledge is scattered, hard to find, and harder to use.
When a customer asks a question, your support team searches through folders, wikis, and PDFs. When a new employee needs a policy answer, they ping a colleague who pings another colleague. The information exists — but it’s not accessible.
A Knowledge Base solves this by centralizing your organization’s information into a structured, searchable system. And with Just247Pipes, building and managing a Knowledge Base isn’t an IT project — it’s a visual workflow anyone can design.
In this guide, you’ll learn how to:
– ✅ Create a Knowledge Base component in Just247Pipes
– ✅ Populate it with your existing documents and FAQs
– ✅ Configure intelligent search with auto-categorization
– ✅ Deliver instant, accurate answers to your users
– ✅ Keep your Knowledge Base up to date automatically
What Is a Knowledge Base on Just247Pipes?
The Knowledge Base component in Just247Pipes is more than a searchable FAQ — it’s an intelligent, categorized, and automatically enriched repository that integrates directly into your data pipelines.
Think of it as the brain of your information system. You feed it knowledge (documents, FAQs, articles), and it gives you back instant, relevant answers — either directly or as part of a larger AI pipeline.
Three Core Operations
| Operation | What It Does | When to Use |
| Add FAQ | Store new question-answer pairs into the Knowledge Base | When ingesting documents, importing FAQs, or enriching content |
| Get FAQ | Retrieve specific FAQ entries by ID or category | When browsing, listing, or displaying knowledge base content |
| Search FAQ | Find the best-matching answer for a user question | When a user asks a question and needs an instant answer |
These three operations cover the full lifecycle: add knowledge, retrieve it on demand, and search it intelligently.
Step 1: Create Your Knowledge Base Pipeline
On the Just247Pipes canvas, create a new pipeline and drag the Knowledge Base component onto it.
Pipeline for Populating the Knowledge Base
File Input → Processor → Knowledge Base (Operation: add_faq)This pipeline ingests your documents, processes them into clean Q&A pairs, and stores them in the Knowledge Base.
Pipeline for Searching the Knowledge Base
Text Input → Knowledge Base (Operation: search_faq) → Response OutputThis pipeline accepts a user question, searches the Knowledge Base, and returns the best-matching answer.
Let’s configure each component in detail.
Step 2: Populate Your Knowledge Base
Before your Knowledge Base can answer questions, it needs content. Here’s how to feed it.
File Input — Connect Your Documents
Drag a File Input component to start your ingestion pipeline:
– Supported formats: PDF, TXT, DOCX, CSV, Markdown, JSON
– Configuration: Point to a local directory or connect cloud storage (S3, Azure Blob, GCS)
– Scheduling: Set up automated ingestion on a cron schedule (e.g., nightly) to keep content current
Tip: Start with your most valuable content — product FAQs, support articles, and policy documents. You can always add more later.
Processor — Clean and Structure Your Content
Raw documents need processing before they’re useful in a Knowledge Base. The Processor component handles this:
| Setting | Recommended Value | Why |
| Chunk Size | 500–1000 tokens | Keeps entries focused and searchable |
| Overlap | 50–100 tokens | Preserves context between chunks |
| Clean HTML | true | Strips formatting artifacts from web content |
| Extract Q&A Pairs | true | Automatically identifies question-answer structures |
The Processor transforms raw text into clean, structured entries that the Knowledge Base can store and search effectively.
Knowledge Base (Add FAQ) — Store Your Knowledge
Configure the Knowledge Base component for ingestion:
| Property | Value | Why |
| Operation | add_faq | Store new entries |
| Auto Categorize | true | Automatically classify entries into topic categories |
| Enable Duplicates Detection | true | Avoid storing redundant content |
💡 Business Value Tip
> Auto-categorization eliminates manual tagging forever. When you enable Auto Categorize, every FAQ entry is automatically sorted into the right topic bucket — billing, technical, account, shipping, etc. For a company with 10,000+ support articles, this saves a full-time employee weeks of work per year. Your Knowledge Base organizes itself.
Example: Ingesting a Customer Support FAQ
Imagine you have a CSV of 500 support FAQs:
question,answer,category
"How do I reset my password?","Go to Settings → Account → Reset Password...","account"
"What are your shipping times?","Standard shipping takes 3–5 business days...","shipping"
"How do I cancel my subscription?","You can cancel anytime from your account page...","billing"Your pipeline processes this automatically:
1. File Input reads the CSV
2. Processor cleans and structures each row into a Q&A pair
3. Knowledge Base (add_faq) stores each entry with auto-categorization
Within minutes, 500 searchable FAQ entries are live in your Knowledge Base.
Step 3: Configure Intelligent Search
Now that your Knowledge Base is populated, let’s make it searchable.
Knowledge Base Search Configuration
Create a new pipeline for the search operation:
Text Input → Knowledge Base (Operation: search_faq) → Response OutputConfigure the Knowledge Base component:
| Property | Recommended Value | Why |
| Operation | search_faq | Find the best-matching answer |
| Max Results | 5 | Return the top 5 most relevant matches |
| Category Filter | (empty = all, or e.g. “billing”) | Narrow search to specific domains |
| Auto Categorize | true | Automatically classify the incoming question |
| Similarity Threshold | 0.6 | Only return results above 60% relevance |
How Search Works
When a user asks “How do I change my plan?”, the Knowledge Base:
1. Analyzes the question’s meaning (not just keywords)
2. Searches all stored FAQ entries for semantic matches
3. Ranks results by relevance score
4. Returns the top matches with confidence scores
This means “How do I change my plan?” matches “How do I upgrade my subscription?” — even though they share only the word “how.” That’s the power of semantic search over keyword search.
💡 Business Value Tip
> Semantic search finds answers keyword search misses. A traditional search for “change my plan” would miss an article titled “Upgrading Your Subscription.” The Knowledge Base’s semantic search understands that these mean the same thing — so users find answers faster, and your support team handles fewer “I couldn’t find the answer” tickets.
Step 4: Category Filtering for Precision
The Category Filter is a powerful feature that narrows search results to specific domains.
How It Works
| Without Category Filter | With Category Filter |
| Searches all FAQ entries | Searches only entries in the specified category |
| Broader results, may include tangentially related content | Precise results focused on the right domain |
| Best for general-purpose queries | Best for domain-specific queries |
When to Use Category Filtering
– Billing questions → Filter to `billing` category
– Technical support → Filter to `technical` category
– HR policy questions → Filter to `hr` category
– Product feature questions → Filter to `product` category
Combining with Intent Detection
For even smarter routing, add an Intent Detection component before the Knowledge Base:
Text Input → Intent Detection → Knowledge Base (Category Filter: {detected_intent})The Intent Detection component classifies the user’s question into a category, and the Knowledge Base uses that category to narrow its search. This two-step approach dramatically improves answer accuracy.
Step 5: Keep Your Knowledge Base Current
A Knowledge Base is only as valuable as its content is current. Just247Pipes makes maintenance effortless.
Automated Re-Ingestion
Schedule your ingestion pipeline to run on a cron schedule:
– Nightly — Re-index documents from your content management system
– On webhook — Trigger ingestion whenever content changes in your CMS
– Manual — Click “Run” whenever you add significant new content
The Knowledge Base’s duplicates detection ensures that re-ingesting the same content doesn’t create duplicates — it updates existing entries in place.
Adding New Content
Need to add a new FAQ entry? Two options:
1. Pipeline ingestion — Drop a new document into your source directory and re-run the ingestion pipeline
2. Direct API — Use the Knowledge Base’s `add_faq` operation via REST API:
{
"question": "What is your refund policy?",
"answer": "We offer full refunds within 30 days of purchase...",
"category": "billing"
}Content Lifecycle Management
| Task | How Just247Pipes Handles It |
| New content | Auto-ingest via scheduled pipeline or API |
| Updated content | Duplicates detection updates existing entries |
| Outdated content | Manual removal or automatic expiration policies |
| Categorization | Auto-categorize handles new entries automatically |
Real-World Examples
Example 1: Customer Self-Service Portal
A SaaS company builds a self-service portal for customers:
Ingestion Pipeline:
Help Center Articles → File Input → Processor → Knowledge Base (add_faq)
Search Pipeline:
Customer Question → Knowledge Base (search_faq) → Best Answer → Chat WidgetResult: 60% of customer questions answered instantly without human intervention. Support team focuses on complex, high-value cases.
Example 2: Employee Policy Assistant
An HR department centralizes all company policies:
Ingestion Pipeline:
Policy Documents → File Input → Processor → Knowledge Base (add_faq, auto_categorize: true)
Search Pipeline:
Employee Question → Knowledge Base (search_faq) → Policy Answer → Slack/Teams IntegrationResult: Employees get instant policy answers in Slack. HR saves 15+ hours per week on repetitive questions. New hires onboard 40% faster.
Example 3: Multi-Department Knowledge Base
A large organization manages knowledge across departments:
Ingestion Pipeline:
Department Docs → File Input → Processor → Knowledge Base (add_faq, auto_categorize: true)
Search Pipeline (with Intent Detection):
User Question → Intent Detection → Knowledge Base (search_faq, category_filter: {intent})
→ Response OutputResult: 80% of internal questions answered automatically. Department-specific routing ensures answers come from the right domain every time.
Why Just247Pipes for Your Knowledge Base?
🎯 Business Value
| Metric | Before Knowledge Base | With Just247Pipes Knowledge Base |
| Time to find an answer | 10–30 minutes of searching | Seconds |
| Support ticket volume | Growing linearly with customers | Deflected by 50–70% |
| Knowledge consistency | Varies by agent experience | Standardized, always current |
| Onboarding time | Weeks of shadowing | Days with self-service access |
| Content maintenance | Manual updates across systems | Automated ingestion and updates |
🧩 Simplicity
– Visual pipeline designer — Drag, connect, configure. No code required.
– Auto-categorization — Knowledge Base organizes itself
– Duplicates detection — Re-ingest without creating duplicates
– Built-in scheduling — Keep content current automatically
🔄 Flexibility
– Multiple input formats — PDF, DOCX, CSV, Markdown, JSON
– Category filtering — Narrow search by domain, department, or topic
– Intent-aware routing — Combine with Intent Detection for smart categorization
– API access — Integrate into any application via REST API
– Scalable — From 100 FAQs to 100,000+ entries
Technology Glossary
| Term | Plain-English Definition |
| Knowledge Base | A structured, searchable collection of your organization’s information — FAQs, documents, policies, product guides — organized so it can be instantly queried and referenced. |
| FAQ (Frequently Asked Questions) | Question-answer pairs that address common inquiries. In Just247Pipes, FAQs are the primary unit of knowledge stored in the Knowledge Base. |
| Auto-Categorization | Automatic classification of knowledge entries into topic categories (e.g., billing, technical, account). Eliminates the need for manual tagging. |
| Semantic Search | Search that understands meaning rather than just matching keywords. “How do I change my plan?” matches “Upgrading Your Subscription” — even with different words. |
| Category Filter | A setting that narrows Knowledge Base search results to a specific domain, improving precision for domain-specific queries. |
| Chunking | Breaking large documents into smaller pieces so they can be individually stored and retrieved. Without chunking, a long document would produce a single entry that dilutes specific topics. |
| Duplicates Detection | A feature that identifies and merges identical or near-identical entries when re-ingesting content, preventing your Knowledge Base from growing redundant. |
| Intent Detection | An AI technique that classifies what the user wants from their message. For example: “I need a refund” → Intent: `billing_refund`. Useful for routing queries to the right Knowledge Base category. |
| Similarity Threshold | A minimum relevance score (0–1) below which search results are discarded. Setting it to 0.6 means “only show results that are at least 60% relevant.” |
| Cron Schedule | A time-based job scheduler that runs tasks at fixed intervals (e.g., every night at 2 AM). Used to automatically re-ingest content and keep the Knowledge Base current. |
Quick-Start Checklist
– [ ] Set up Just247Pipes — Create a FREE Account or Sign In
– [ ] Prepare your content — Collect FAQs, documentation, policies in supported formats
– [ ] Build the ingestion pipeline — File Input → Processor → Knowledge Base (add_faq)
– [ ] Run ingestion — Populate your Knowledge Base with initial content
– [ ] Build the search pipeline — Text Input → Knowledge Base (search_faq) → Response Output
– [ ] Test with real questions — Try queries your users would actually ask
– [ ] Enable auto-categorization — Let the system organize content automatically
– [ ] Set up scheduled re-ingestion — Keep your Knowledge Base current automatically
– [ ] Integrate into your application — Connect via REST API or embed in your chat widget
– [ ] Add Intent Detection — For smarter category routing (optional enhancement)
What’s Next? Supercharge Your Knowledge Base with RAG
A Knowledge Base gives your users instant, accurate answers from your curated FAQ content. But what if you want to go further — answering questions that go beyond your FAQs, drawing from your full document library, and generating natural-language responses with citations?
That’s where RAG (Retrieval-Augmented Generation) comes in. RAG combines your Knowledge Base with an LLM to generate intelligent, source-grounded answers to any question — even complex ones that don’t have a pre-written FAQ entry.
—
Ready to give your team instant access to organizational knowledge?
—
Just247Pipes — Transform your knowledge into instant answers with visual pipeline design.
Leave a Reply