Lead qualification on WhatsApp, done by an AI before your team sees it
Half of the enquiries a business gets on WhatsApp are not for it, and finding out costs a rep the same twenty minutes as a good one. Linkubit asks the qualifying questions in the first minute — the ones you define — writes the answers onto the lead, scores it, and hands your team a brief rather than a "hi, price?".
Linkubit answers an enquiry on WhatsApp, email or Instagram within a minute, asks the qualifying questions you configured — one or two at a time, in the customer's language — records the answers as fields on the lead, scores it, and escalates to a person when a rule says so. Nothing is quoted before the fields a rule requires are known.
What it asks for
- What they want, in their words
- Budget or range, with 'not sure' as an answer
- When they need it
- Where they are — city, pin code, delivery area
- Who decides, and whether they are the one asking
- How they found you
None of this is hardcoded. It is a schema you edit, which is why a completely different business runs on the same system.
The questions are yours
A schema you edit, not a script somebody wrote
Each field is a column on the lead — type, options, whether it is required. The assistant asks for what is missing and never asks twice. A freight forwarder and a coaching centre run on the same product with different fields.
- Fields fill in as the customer answers
- A tenant's own field wins over anything the product infers
Freight forwarder
Property consultancy
Some enquiries go straight to a person
Escalation is a rule, and it fires before a reply exists
A named topic, a keyword, a value above a threshold, an angry message — any of them hands the conversation to a human with the whole thread and the fields captured so far. The person picks up mid-conversation, not from scratch.
- The assistant stops the moment the rule fires
- The handover note is internal; the customer sees one voice
Sales Agent
Internal note · good running volume, taking this one — Rohan
Rohan · took over
The rule that holds
No quote, brochure or slot before the required fields are known
You mark which fields are required before the assistant may quote, shortlist or book. The block is a rule enforced in code: the tool is unavailable to the model until the fields exist on the record, so a persuasive customer cannot talk it past the check.
Anything else goes to a person
A topic you named, a value above a threshold, a customer who asks for a human, or a question the assistant cannot answer from what you gave it — the conversation is handed over before a reply is written, with the whole thread and every field captured so far.
From "hi, price?" to a lead a rep wants
What an enquiry looks like when the first minute is handled — the greeting, the questions, the record, the score and the handover.
Answered inside a minute, on the channel it arrived
WhatsApp, email, Instagram DM or Messenger — one inbox, one contact, one thread. The reply names the business, answers the actual question if it can, and asks the first qualifying question.
One or two questions at a time
Not a form. The assistant asks in the customer's language, accepts a voice note, understands "around two lakh" as a budget and "next month" as a date, and writes both onto the lead.
The record fills in as they talk
Every answer becomes a field a scoring rule can read. Budget above a threshold, a city you serve, a timeline inside a quarter — each adds to the score you configured.
The rule decides what happens next
Qualified: the assistant continues to a quote, a shortlist or a slot. Needs a person: it hands over with the fields captured and a note. Not for you: it says so politely and records why.
Your team opens a brief, not a chat
The lead sits on the pipeline with its fields, its score and the thread. A rep reads thirty seconds of record instead of ten minutes of messages, and answers the one thing the assistant could not.
The templates it sends
4 WhatsApp message templates from the industry packs that carry this use case, each written the way Meta's review reads it. Open one for its variables, a filled example and what gets it rejected.
| Template | Category | When it goes out |
|---|---|---|
| Missing details to quote | Utility | The enquiry cannot be priced — no volume, no commodity, no Incoterm. |
| Enquiry received | Utility | A buyer or tenant enquires from a portal, an ad or a hoarding. |
| Enquiry received | Utility | A student or parent enquires about a course. |
| Enquiry received | Utility | A buyer enquires about a model from an ad, a portal or a walk-in form. |
Each template opens on its own page with the variables, a filled example and what Meta’s review objects to. Utility and authentication templates cost ₹0.115 a message in India, marketing ₹0.8631 — the pricing guide has the rest.
Questions people ask
Can I decide which questions it asks?
Yes — they are fields on the lead, with a type, options and a required flag, edited on the dashboard. The assistant asks for whatever is missing. There is no script to maintain.
What if the customer refuses to give a budget?
'Not sure' is a valid answer to any field you allow it on, and the assistant moves on. A field you mark required blocks the quote until it is known, and the assistant explains why it is asking.
How does it know when to hand over to a person?
Rules: a topic you named, a keyword, a value above a threshold, a customer who asks for a human, or a message the assistant cannot answer from what you gave it. The rule fires before a reply is written, and the person sees the whole thread.
Does it work on email and Instagram as well as WhatsApp?
The same fields, rules and scoring apply on every channel. A person who emailed and then messaged on WhatsApp is one contact and one lead, not two.
Read next
Ready to put your sales on autopilot?
Start freeNo coding required. No credit card required.