AI chatbots that answer from your own content
For teams answering the same customer and staff questions every day across website chat, WhatsApp and email. We build AI chatbots that answer only from your approved content, cite their source and pass the conversation to a person when they are not sure.
- Timeline
- 2 to 4 weeks to first version
- Best for
- Teams answering the same questions in chat and email daily
- Usually starts with
- A website chatbot grounded in your help center and FAQs
Agent flow
What is an AI chatbot?
An AI chatbot is a chat assistant that uses a language model to understand questions in plain language and reply in full sentences. A business chatbot should be grounded in your own help articles, policies and product data, so it answers from facts you control rather than guessing, and it should hand the chat to a person when it cannot help.
- Answers only from your help center, policies and product data
- Works on your website, WhatsApp, Instagram and help desk
- Hands off to a person with the full chat history
- Captures leads and creates tickets while it talks
Typical AI chatbots we ship
Each one scoped and quoted at a fixed price after a free review of your process.
Website support chatbot
A visitor asks about shipping times or sizing. The chatbot searches your help center and policies, replies in plain language with a link to the source, and offers a person if needed.
WhatsApp order and booking questions
Customers message your WhatsApp Business number. The bot checks order status in Shopify or open slots in your calendar, answers or books instantly, and passes complaints to your team inbox.
Lead capture and qualification
When a visitor shows buying intent, the chatbot asks a few qualifying questions, collects contact details with consent, creates the lead in your CRM and offers a meeting slot.
Internal HR and IT helpdesk
Employees ask about leave policy, expenses or VPN setup in Slack or Teams. The bot answers from your handbook and SOPs, and opens a ticket when it needs IT or HR.
Product finder for online stores
A shopper describes what they need in their own words. The chatbot matches it to products in your catalog using live stock and specs, and suggests two or three options with links.
Multilingual customer support
A customer writes in Spanish, Arabic or Hindi. The chatbot replies in the same language from your English content, and translates the summary for your team if it hands the chat over.
A typical build, end to end
Replies on WhatsApp and your website in seconds, in the customer's language.
What changes for your team
Support answers the same ten questions by email all day.
The chatbot answers common questions instantly from your help center, and staff handle the rest.
The old chatbot follows rigid menus and frustrates customers who type real questions.
Customers type naturally and get accurate answers with a link to the source.
Chats after hours wait until morning, and some buyers leave.
Questions get answered at any hour, and leads land in your CRM overnight.
Four steps, one fixed price
- 01
Audit questions and content
We pull recent tickets and chats to find the most common questions, then check whether your help content actually answers them clearly.
- 02
Build the knowledge base
We clean and structure your articles, policies and product data so the bot retrieves the right passage, and we fill gaps with you.
- 03
Set guardrails and handoff
We define topics it must not answer, tone, escalation triggers and how chats pass to staff in your help desk or inbox.
- 04
Test, launch and improve
We test against a large set of real past questions, launch on one channel, then review unanswered questions weekly to improve content.
How we ground AI chatbots in your own help content
The biggest risk with a business chatbot is a confident wrong answer. We reduce that by grounding the bot in your own material: help center articles, policy pages, product data, past ticket replies and internal SOPs. When a question comes in, the bot first retrieves the most relevant passages, then writes its answer only from them. If nothing relevant is found, it says so and offers a person instead of improvising.
Grounding only works if the content is good, so part of every project is fixing the source. We find questions your help center does not answer, policies that contradict each other and pages that are out of date, and we flag them for your team. Once live, the bot logs every question it could not answer, which gives you a running list of articles worth writing.
How chatbot to human handoff should work
A good chatbot knows when to step aside. We set clear handoff triggers: the customer asks for a person, mentions a refund or complaint, gets the same unhelpful answer twice, or the conversation involves payment or legal matters. The chat then moves to your team in Zendesk, Intercom, Freshdesk, HubSpot or a shared inbox, with the full transcript and a short summary attached.
Outside business hours, the bot tells the customer when a person will reply and creates a ticket so nothing is lost. Your staff never have to ask the customer to repeat themselves. Over time, the handoff log shows which topics the bot should learn and which should always go to a person, so the split between bot and team is a decision you make with real data.
Custom AI chatbot vs off-the-shelf chat widgets
Off-the-shelf chat tools are fine for a simple FAQ. They struggle when answers depend on your own systems: a customer's order, an account balance, stock at a specific store, or whether a booking slot is free. A custom chatbot connects to those systems through their APIs, so it can answer with live data and take small actions, like rescheduling a booking or updating an address.
It also means you choose the model, the data store and where conversations are kept, which matters for GDPR and for sectors with strict privacy rules. You own the prompts, the knowledge base and the integration code. We often build on top of the help desk you already use rather than replacing it, and every project starts with a fixed-price quote after a free 30-minute audit.
AI Chatbots questions
How do you stop an AI chatbot from making things up?
We ground it in your own content and instruct it to answer only from retrieved passages. If it cannot find a relevant source, it says it does not know and offers a person. We also block topics it should never handle, show source links in answers and test against a large set of real past questions before launch. No chatbot is perfect, which is why handoff is built in from day one.
Can the chatbot hand conversations over to my team?
Yes. Handoff triggers are agreed with you, such as a request for a person, a complaint, a refund or a repeated failed answer. The conversation moves into your help desk or shared inbox with the full history and a summary, so the customer never repeats themselves. After hours, it sets expectations and creates a ticket instead.
Which channels can an AI chatbot work on?
Most projects start with website chat, then add WhatsApp Business, Instagram and Facebook Messenger, or Slack and Microsoft Teams for internal bots. The same knowledge base and rules serve every channel, so answers stay consistent. We connect to help desks like Zendesk, Intercom and Freshdesk, or route chats to a shared inbox if you do not use one.
How do we keep the chatbot's answers up to date?
The bot reads from your help content, so editing an article changes the answer after the next scheduled sync, which we can run hourly or daily. We connect that sync to your help center, Notion or Google Drive. A weekly report lists questions the bot could not answer, so your team knows exactly which content to add or fix.
Who owns the chatbot and the conversation data?
You do. The chatbot runs in accounts registered to your business, and you own the prompts, knowledge base, integration code and every conversation log. We handle data under GDPR for UK and EU users and can sign an NDA on request. If you ever move to another provider, everything is documented and ready to hand over.
Often built together
AI ReceptionistA 24/7 AI receptionist that answers, books appointments and flags urgent calls to your team.
AI Agent DevelopmentCustom AI agents that do multi-step work across your tools, with a person approving key calls.
CRM AutomationKeep your CRM accurate and moving without anyone typing in updates by hand.
Get a fixed quote for your AI chatbots
Send two lines about the process you want to hand over. We reply within one business day with what we would build, the tools involved and what it would take.
- 30-minute call, no slides, no pressure
- One fixed price before any work starts
- You own every workflow, account and line of code