Leads and customers almost never mind speaking to an AI over WhatsApp, Instagram, Telegram or a website. Especially if it is good or better than a human at its job. Answering frequently asked questions, checking availablility, adding data to a table, opening a support ticket.
But it gets brutal on call. Almost no one I know has 'been okay' (let alone ejoyed) being dealt by an AI over call. Even 2-3 years after Eleven Labs becoming mainstream, people just do not like talking to AI.
There's a strange, mecahnical consistency to it. And 80% of people can easily tell. Because it makes them feel processed. It makes them feel like the same person it spoke to just before.
You and I have both been on the receiving end of it: the reply that opens with "Hello, how can I help you?". You immediately know there's some GPT in the mix.
The Humanising masterclass comes with your $36 receptionist, and it exists to remove exactly that feeling.
It covers two layers.
The voice layer
If your receptionist speaks, it speaks through ElevenLabs. ElevenLabs is, on the face of it, a text-to-speech service.
What matters for your business is the difference between its defaults and its potential: the same engine can sound like a satnav or like a person, and the gap is entirely in the choices.
The masterclass walks through those choices:
- Which voice suits your business.
- Pacing, and why slightly slower reads as more confident.
- The settings that make a phone reply sound answered rather than narrated.
The Humaniser prompt
The second layer is written tone, and it is carried by something we call the Humaniser prompt: a layer of instructions that changes how your receptionist says everything, without touching what it knows.
- Shorter sentences.
- Real acknowledgements instead of customer-service filler.
- Answers that address the question asked, in the order it was asked.
- Contractions where a person would use them.
The difference is hard to describe and obvious to hear.
Put two replies side by side, one with the layer and one without, and you will pick the humanised one every time without being able to fully say why.
Tone is a dial, not a setting
Here is the part that makes this practical rather than a one-off polish: tone drifts, and businesses differ.
Maybe your receptionist is too formal for your Instagram audience. Maybe it is too breezy for a law firm. Maybe it is perfect in English and stilted in Urdu.
You do not edit a prompt file.
You tell the same Claude Code trainer that configured your receptionist:
- "warmer on Instagram"
- "more formal with anything involving fees"
- "match the customer's language and register"
The trainer adjusts the personality layer, and because your receptionist is one agent on many channels, the tweak is live everywhere at once.
Tuning tone (just like adding tools or updating behaviour) becomes a conversation, not a 30 day project.
Which means you actually do it.
The knowledge makes your receptionist correct. The humanising makes people stay in the conversation long enough to benefit from the correctness.
You need both, and both are in the $36.
A note on prices: every price in this series is the price inside its funnel, where each purchase unlocks the next. Bought on its own, outside the funnel, the same product can cost up to 30% more.
Next up: where every one of those conversations ends up, your CRM.