An AI assistant that replies in your brand's voice — with a human in command.
Every unanswered message is a customer lost. Nº 5 installs named assistants that help make sure nothing goes unanswered — on your site and social. The difference: the AI suggests, and a human approves before anything is published.
The problem it solves
Comments and messages arrive at any hour — and most SMEs don't have someone available to reply quickly. Sales are lost to silence. But blind auto-replies are scary too: nobody wants their brand talking nonsense. We solve both sides: speed with human control.
Who it's for
For businesses that receive contacts via comments, DMs or their website and want to reply quickly and in the right tone — without losing control of what goes out. Each client gets an assistant with its own name and voice, aligned with their brand.
Signs you need this
- Messages arrive after hours and go unanswered.
- You take hours (or days) to reply to comments and DMs.
- You lose track of conversations across Instagram, Facebook and the site.
- You want speed, but don't trust unsupervised auto-replies.
- The same questions keep repeating and eat up your time.
How we work
- Assessment — which questions repeat, what tone the brand uses and where messages land.
- Knowledge base — we build what the assistant knows (services, real information, what it must never say).
- Approval flow — the AI drafts the reply; you approve in one click before it goes out.
- Continuous tuning — the assistant improves with history and your feedback.
What it may include
- A named assistant on your site, with your business information.
- Support for handling comments and messages on Instagram and Facebook.
- Human approval flow — nothing public goes out without your click.
- A controlled knowledge base and rules for what it must never say.
- Conversation history so the assistant always replies with context.
What is not automatically included
So there are no surprises: it is not 24-hour human service (the AI collects and drafts at any hour; public decisions keep human oversight), integrations with external systems (CRM, bookings, payments) increase scope, and sensitive cases — complaints, legal topics, concrete commercial requests — are always routed to a person. Anything outside the scope is quoted upfront.
Real proof
We run named assistants across several brands — from Joaquim, the Terrae concierge, to assistants for other brands across two countries. Each replies in its brand's voice, with human approval before publishing. The assistant on our own site is living proof of what we install.
Straight answers
Does the AI reply on its own?
No. It works with a controlled knowledge base and public replies go through human approval. People first, AI as an accelerator.
Is it 24-hour service?
The AI collects, organises and drafts replies at any hour — but the decision and the publishing keep human oversight. We don't promise a person awake at 3am.
Which channels does it work on?
The site chat and support for comments and messages on Instagram and Facebook. Other channels depend on the scope and each platform's permissions.
What if someone asks something delicate?
Complaints, sensitive topics or concrete commercial requests are routed to a person — the assistant doesn't close deals or give terms on its own.
Social media management · Website creation · Meta & Google Ads · Strategy & brand