FAQs

Frequently asked questions

Direct answers to the questions we are asked before every implementation.

What types of businesses do you work with?

We work with independent businesses that answer repeat customer questions - shops, clinics, consultancies, trades, agencies, hospitality venues, and online sellers. The chatbot handles the repetitive queries so you and your small team can focus on the work that actually grows the business.

Which chatbot platforms and LLMs do you support?

We are platform-neutral. We implement on the vendor and model that fit your data, budget, and compliance requirements, including hosted chatbot platforms and custom builds on large language models such as GPT, Claude, and Gemini. We do not resell software or take commission.

How long does a typical implementation take?

Full implementation is done within a week. Discovery and conversation design happen in the first couple of days, integration and configuration follow, then launch and early training before the week is out. Larger multi-channel programmes can run a little longer.

Can you work with our existing helpdesk or CRM?

Yes. We integrate chatbots with common helpdesk, CRM, knowledge base, and commerce systems such as Zendesk, Intercom, Freshdesk, HubSpot, Salesforce, Shopify, and internal APIs, so conversations create tickets, read order data, and hand over cleanly to you or your team.

How do you handle data privacy and compliance?

We map what data the chatbot can read and store before anything is built, apply role-based access, redact personal data where it is not needed, and document retention. Implementations are designed to support GDPR obligations and your internal security review.

Will this replace me or my team?

No. The chatbot handles repetitive, well-documented queries and passes the rest to you. Complex, sensitive, or high-value conversations reach you with full context attached, so you spend your time on the customers and work that matter most.

What does ongoing optimisation look like?

We review real conversation transcripts, identify unanswered or misanswered intents, update content and routing, and retest. In the first 90 days after launch this happens on a weekly or fortnightly cycle, then monthly once performance stabilises.

What do you need from me during the build?

A decision-maker who knows the business, access to your most common questions and answers, and a technical contact for system access. Most clients spend around two to four hours a week with us during the build.

Can the chatbot answer from my own documentation?

Yes. We connect the chatbot to your knowledge base, help centre, product documentation, and policy pages using retrieval, so answers are grounded in your approved content rather than invented by the model.

What happens if the chatbot does not know an answer?

It says so and escalates. We define fallback and escalation paths during conversation design, including handover to live chat, ticket creation, or a callback request, so a customer is never left in a loop.

Still have a question?

Tell us what your customers keep asking and we will come back with a realistic scope.

Book a discovery call