The most important thing to understand about your Lameda assistant is not what it can do: it is what it deliberately will not do on its own.
Handles what it knows, hands off what it does not
For routine work, such as answering a product question from your Catalogue, quoting a price, walking a customer through checkout, or answering an FAQ from Knowledge, your assistant acts on its own, instantly, at any hour. For anything uncertain, sensitive, or outside what you have told it, it stops and routes the conversation to you through a Handoff instead of taking a guess.
Why this matters more than raw cleverness
An assistant that answers everything, even things it does not actually know, will eventually get something wrong in a way that costs you a customer or a refund. An assistant that says “let me get my team to help with that” and flags it for you is slower on that one answer but far safer across every conversation put together. When real orders and real payments are involved, being predictable is worth more than being impressive.
- A pricing question your Catalogue answers directly: handled automatically.
- A policy question covered in your Knowledge FAQs: handled automatically.
- A customer asking for a discount, a custom order, or a complaint about a past order: routed to a Handoff.
- Anything the assistant has not been given an approved answer for: routed to a Handoff, not improvised.
What this means for you day to day
The better your Catalogue and Knowledge are kept up to date, the fewer conversations need to reach you personally, not because the assistant got braver, but because you gave it more ground it is confident standing on. Handoffs are not a failure of the system; they are the system working exactly as intended, sending you the conversations that genuinely need a human.
If you want fewer Handoffs over time, the answer is not to change how the assistant behaves. It is to keep expanding what it is allowed to know, through your Catalogue and your Knowledge entries.