Use case · AI Knowledge Base Assistant
An AI knowledge base assistant that answers from your documents, not guesses
Clientager turns your policies, guides, price lists and FAQs into accurate answers for customers, on phone, chat, WhatsApp, email and SMS, and tells you which questions it couldn't answer.
Ready for your review in 24–48 hours
Customer: If I cancel my Lisbon package 20 days before departure, do I get anything back?
Clientager: Hi, I'm the AI assistant for Wayfarer Travel. Under our package terms, cancelling 15–29 days before departure refunds 50% of the package price, minus any non-refundable flight fares shown on your booking.
Customer: And if I had travel insurance?
Clientager: Travel insurance may cover cancellation for covered reasons, such as illness, but that depends on your policy, and claims go through the insurer, not us. I can share the insurer's claim contact if you have a policy with them.
Customer: Can I just change the dates instead?
Clientager: Date changes more than 14 days before departure are allowed once, with a $75 change fee plus any price difference. Would you like me to ask an agent to check new dates for you?
Customer: Yes please, maybe two weeks later.
Clientager: Done. A travel agent will email you options for dates about two weeks later by tomorrow afternoon.
What Clientager did
- Policy answered: 50% refund at 15–29 days, change fee $75
- Task for travel agent: date change quote, +2 weeks
- Source linked: package terms, section 4
Quick answer
What is an AI knowledge base assistant?
An AI knowledge base assistant answers customer questions using your own content: policies, FAQs, product manuals, service guides and price lists. Instead of sending people to search a help center, Clientager finds the relevant information and explains it in plain language on any channel. It stays within your approved content, says when it does not know, and flags gaps so you can improve your knowledge base.
Illustrative example
Customer message
Does my plan cover this?
Happy to help with that. Let me check and get it sorted for you now.
Saved to the customer record
What Clientager handles for ai knowledge base assistant
- Answer from your documents
- Explain policies plainly
- Cover every channel
- Admit unknown answers
Live availability
Booked · reminder scheduled
What gets missed today
Customers don't read the help center
You wrote detailed FAQs and guides, but customers still call or email. Searching a help center feels like more effort than asking.
The right answer is buried in a PDF
Policy documents, manuals and terms are long. Staff spend time hunting for the one clause that answers a simple question.
Old information keeps circulating
An outdated price list or policy lingers in someone's saved replies, and customers receive answers that are no longer true.
You don't know what customers are asking
Without a record of unanswered questions, it is hard to see what is missing from your documentation.
What Clientager can handle for you
- Answer from your documents
- Explain policies plainly
- Cover every channel
- Admit unknown answers
- Flag knowledge gaps
- Keep answers current
- Link to source pages
Enquiries it handles every day
- “What's your cancellation policy?”
- “Does my plan cover this?”
- “How do I set this up?”
- “What's included in the package?”
- “What documents do I need?”
- “Where can I find the terms?”
How it works
- 01
Gather your content
You share policies, FAQs, manuals, price lists, terms and guides. We organize them and mark which sources are authoritative.
- 02
Configure and test
Our team builds the knowledge into the agent and tests it with real customer questions, including tricky and ambiguous ones.
- 03
Answer in plain language
When a customer asks, the agent finds the relevant content and explains it simply, linking to the source page where useful.
- 04
Say when it doesn't know
If the answer is not in your content, it says so honestly and passes the question to your team rather than guessing.
- 05
Improve over time
Unanswered and frequently asked questions are reported, so you can add or update content and keep answers accurate.
Grounded answers instead of guesswork
The main risk with AI answering customer questions is confident, wrong answers. Clientager's knowledge base assistant is configured to answer from the content you provide and approve, not from general internet knowledge. When a customer asks about your cancellation terms, the agent explains your terms. When your content does not cover a question, it says so and routes it to your team instead of improvising a policy.
Answers are written in plain language, not pasted paragraphs of legal text. A customer asking "do I get anything back?" gets a direct answer with the relevant figure, then a link to the full terms where useful. This is especially valuable for insurance agencies, travel agencies and property managers, where policies are detailed, exceptions are common and a wrong answer can be costly for both sides.
Keeping your knowledge current
Knowledge only helps if it is up to date. During setup, we agree which documents are the source of truth and how changes reach the agent. When your prices, policies or procedures change, tell us, and we update and test the agent before the change goes live. Older documents can be retired so they stop influencing answers, and seasonal information, such as holiday hours, can be scheduled in advance.
The agent also shows you where your content falls short. Questions it could not answer, and questions asked repeatedly, are reported to you. These reports are a practical to-do list for improving your FAQs, website and documentation, and they often reveal confusing policies or product instructions that cause problems further down the line. Fixing those at the source reduces questions on every channel.
One knowledge base, every channel
Customers ask the same questions by phone, chat, WhatsApp, email and SMS. The knowledge base assistant uses the same approved content on all of them, so answers are consistent whether someone calls on Saturday or emails on Tuesday. On the phone, answers are spoken naturally and kept short. In email, they can be more detailed with links to the relevant documents.
Where a question needs action rather than information, such as changing a booking, the agent can complete it within your rules or pass it to your team. Our team configures and tests the knowledge with you, usually ready for your review within 24–48 hours. Plans start at $39 per month; see our pricing page, as prices may vary by region.
Frequently asked questions
What kinds of documents can the assistant learn from?
Common formats such as PDFs, web pages, help center articles, documents and spreadsheets. Policies, FAQs, manuals, price lists and service guides work well. During setup we review your content with you and agree which sources should be treated as authoritative.
How do you stop the AI from making up answers?
The agent is configured to answer from your approved content and to say when it cannot find an answer. We test it with real and tricky questions before launch. If a question falls outside your content, it is passed to your team rather than guessed.
Can it link customers to the source document?
Yes, on channels where links make sense, such as chat, email and messaging. It can point to the relevant help article, terms section or guide so customers can read the full detail if they want to.
How quickly are updates to our policies reflected?
When you tell us about a change, we update the agent's knowledge and test it before it goes live. Where content comes from a connected source, such as a help center, the process for keeping it in sync is agreed during setup.
Will it answer questions outside our business?
It is designed to stay focused on your business. If someone asks something unrelated or outside your content, it politely redirects or passes the question to your team, rather than acting as a general-purpose chatbot.
Can it handle technical product questions?
Yes, if the information is in your manuals or guides. It can walk customers through setup steps or troubleshooting in plain language. Problems that need a technician or a warranty decision are passed to your team with the details.
Related industries
Use cases
Guides
Want this for your business?
Tell us how your business works. We'll build the AI around it, test it, and have it ready for your review within 24–48 hours.