Tangensys • Plan, deliver, measure
AI chatbot development
Ground the assistant in approved business information and make the handover obvious. A chatbot should be helpful when it knows the answer and cautious when it does not.
The work to plan together
A scope shaped around this requirement.
- Knowledge base setupScope
- Chatbot build and toneReview
- Hand-over to peopleCheck
- Approved informationWhat the tool is allowed to use
- Routine stepHandled automatically
- Human hand-overYour team takes over when needed
- Regular reviewOutputs checked and corrected
In short
What does AI chatbot development involve?
AI chatbot development builds website chatbots that answer questions from your content, qualify visitors and hand over to your team, including customer support assistants.
When AI chatbot development is a useful fit
Businesses with repeated questions
Start with approved answers to stable questions and a clear route to a person.
Support teams
Test how the assistant handles missing information, uncertain requests and human takeover.
Sites with out-of-hours visitors
Explain response limits and collect a callback request without claiming a human is always available.
Problems worth addressing first
Use these situations to identify the work that deserves attention before expanding the scope.
- Visitors leave with unanswered questions
- Support repeats the same answers
A practical example
A visitor asking about a service can receive an approved explanation and a route to a person. A request outside the knowledge base should not produce an invented price.
This is an illustrative scenario to explain the approach, not a claimed client project.
The work we can include
The final quote identifies the deliverables, quantities, approvals and exclusions. Select the work that serves your goal rather than assuming every item is required.
- Knowledge base setup
- Chatbot build and tone
- Hand-over to people
- Reporting
Where we usually begin
AI can still make mistakes. Knowledge sources, escalation rules, data handling, provider costs and review responsibilities are agreed before launch.
Practical detail
Test the answer and the handover
An illustrative support assistant answers from an approved service guide. Its test set should include ordinary questions and requests it should not answer confidently.
| Part of the work | Example output | How to review it |
|---|---|---|
| Supported question | “How do I request a quote?” → approved steps | The response agrees with the current published process. |
| Unknown information | “Can you guarantee this will work for me?” → limitation and staff route | The assistant does not invent a guarantee. |
| Changed fact | An old service price conflicts with the current guide | The evaluation identifies the stale source and the answer is reviewed. |
| Handover | “Please let me speak to someone” → agreed contact route | The request reaches the nominated route with appropriate context. |
Sign-off records should identify the question, source version, response and pass/fail reason. Knowledge updates, unresolved questions and monitoring need an accountable owner.
Define what the assistant may answer and when it must hand over
A chatbot should use approved information for a small, clearly defined purpose. Start with questions it may answer, data it may access and actions it may take. A fluent response is not evidence that the answer is correct or authorised.
Choose a small useful task, such as explaining service scope or routing an enquiry. Keep uncertain, sensitive or account-specific requests on a human path unless the agreed system supports them safely. Source ownership matters because outdated information can produce misleading answers.
Test unanswered questions, conflicting sources and requests outside scope. Include a visible handover route and a method for reviewing failures. Model and messaging-provider capabilities, fees and delivery responsibilities need confirmation before the build is proposed.
Free AI project review
Start with a clear scope and a practical next step.
Tell us what you want to improve and share the relevant website or project details. We will review the starting point, recommend a suitable scope and provide a custom quote.
The initial review is free. Delivery, third-party charges and ongoing support are quoted separately.
Process
How the work is delivered
Review the starting point
Review the current setup, examples of the problem and the information needed to agree the work.
Agree scope and sign-off
We start with real examples of the task and approved information. A small, clearly defined pilot tests usefulness, failure cases and the level of human oversight before wider rollout.
Deliver and review
Evaluation includes inaccurate answers, unavailable tools and uncertain requests. Provider charges, monitoring and updates are defined separately; the business keeps an accountable human owner.
Hand over the next steps
Review the agreed outputs together. Confirm what your team maintains and what needs a separate ongoing arrangement.
What progress should mean
Measures that match the work
Test answer accuracy, unresolved questions, handover success and response time with realistic examples. Conversation volume alone does not establish customer satisfaction.
A record of delivery
See what was completed, reviewed and still dependent on access or approvals. Project delivery and business results are reported separately.
Clear limits in the data
We explain what can be verified and where a result is only an estimate. Contact clicks, conversations and sales are different actions.
Scope and responsibilities
You keep ownership of your business accounts and approved deliverables under the agreed contract. Any third-party licences, subscriptions or usage fees are identified before you commit.
Can a chatbot handle every customer question?
No. Set a supported knowledge scope and escalation rules. Some questions require a person, current account data or specialist judgement. The sign-off set should include cases the assistant must decline or hand over, not only questions it can answer.
Can the chatbot answer from our documents?
Where the agreed setup supports it. Documents need approval, suitable access controls and a freshness process; using documents does not eliminate inaccurate answers.
How do we keep answers current?
Name the source owner and update route. After a material change, recheck representative questions and remove superseded information from the approved source set.
Will the bot invent answers?
AI can make mistakes even when it uses approved content. We test representative questions, set limits and provide a handover route when an answer is uncertain.
How is AI chatbot development priced?
We quote against your goals, current setup and agreed scope. The quote separates delivery from third-party costs and ongoing support. You receive a custom quote within 2 working days.
How long does the work take?
The timeline follows the size of the agreed scope, access, content and approvals. We confirm milestones in the quote; a simple change and a full implementation need different schedules.
How will we communicate?
Your point of contact agrees a review rhythm with you. We use email, WhatsApp and scheduled calls, with decisions and scope changes recorded clearly.
What do you need to get started?
Share your current website or system, your main goal and examples of what is not working. We then agree the access and information needed for discovery.
Is everything included in the free review?
No. The free initial review helps define the next step. Detailed research, design, implementation and ongoing support are included only when stated in your quote.
Explore the related services
Let’s agree the right next step.
Tell us your goals and current setup. We’ll recommend a practical starting point and provide a custom quote.
Custom quote within 2 working days