Home · Business · Sales · Lead qualification
As of 13 August 2026, AI can only partly qualify leads from your website live chat.
Most people should hand this to a purpose-built tool.
Can you do it?
15 minutesto a draft.
30 minutesto something you’d act on.
Cost, all in£0
Skill neededpower-user
Who has to check ita colleague
What the alternative costsTidio Lyro is a purpose-built website live-chat agent that answers customer questions from your content.
If this goes wrong: a promising enquiry is labelled unqualified or sent to the wrong person, and the opportunity may be lost before anyone notices.
What to actually do
Use a tool built for this
The route this page recommends
Hand it to a person
Second choiceA person who owns the outcome does this end to end, worth it when the failure is dear.
Do it yourself
The distant thirdA chat interface, power-user skill, and roughly 30 minutes until you can act on the result.
How to actually do it
- Open your current website copy, FAQ, product or service information, price list and contact or booking details, then remove anything out of date.
- Write down the qualification rules in business terms, including the locations served, customer types accepted, budget or timing requirements, disqualifiers and cases that require a human.
- Define the hand-off routes in a table, with one named team or person for each type of enquiry and a separate route for complaints, uncertainty and urgent requests.
- Paste the source material, rules and routes into a chatbot with the prompt, then ask it to produce the flow, labels, hand-off summary and ten test messages.
- Compare every drafted answer, price, policy and routing rule against your live website and internal records, deleting anything the source material does not support.
- Run the ten tests plus recent anonymised chat examples through the proposed flow, checking that good leads are retained, poor fits are handled politely and unclear cases reach a human.
- Configure the approved questions, labels, hand-off message and human fallback in your live-chat system, then have a colleague monitor the first conversations and amend the rules when a decision is wrong.
Prompt
Design a live-chat lead-qualification flow for [BUSINESS NAME]. Business and offer: [DESCRIBE WHAT YOU SELL, WHO BUYS IT, SERVICE AREA, PRICE RANGE IF RELEVANT, AND CURRENT AVAILABILITY] Qualification rules: [A LIST OF THE FACTS THAT MAKE AN ENQUIRY A GOOD, POSSIBLE OR POOR FIT] Questions the chat may ask: [LIST THE QUESTIONS AND ANY QUESTIONS IT MUST NOT ASK] Routing: [STATE WHICH TEAM OR PERSON SHOULD RECEIVE EACH TYPE OF ENQUIRY, AND WHAT COUNTS AS AN URGENT HAND-OFF] Source material: [PASTE YOUR CURRENT WEBSITE COPY, PRODUCT INFORMATION, FAQ, PRICE LIST AND CONTACT DETAILS] Create: 1. A concise opening message. 2. A question-by-question chat flow using plain UK English. 3. A qualification label with clear rules for good fit, possible fit, poor fit and unclear cases. 4. A short hand-off summary containing only information the visitor provided. 5. A fallback message for uncertainty, refusal to answer, complaints or requests outside the supplied information. 6. A test set of ten varied visitor messages, including ambiguous and unsuitable enquiries, with the expected label and route. Do not invent prices, availability, policies, customer facts or claims about our service. Do not present a guess as a qualification decision. Ask for only information needed for this flow, avoid sensitive personal data, and direct a visitor to a human when the rules do not resolve the case. Mark every statement that needs confirmation against the source material.
Open it prefilled in ChatGPT or Claude, or copy it into Gemini, which takes no prefill link.
What it gets wrong
- AI cannot decide what a valuable lead means when your sales criteria are incomplete or competing.
- It cannot read an uncertain visitor's intent as reliably as an experienced salesperson who can ask follow-up questions and notice commercial context.
- It cannot know whether your current stock, capacity, territory or pricing has changed unless you update its source material or integration.
- It can collect personal information unnecessarily unless you restrict the questions and set appropriate privacy and retention controls.
- It cannot take responsibility for a missed opportunity or a misleading answer; that remains with your business.
What caps this at PARTLY: judgement under ambiguity, context depth and consent and privacy.
How we scored this
Five axes, each scored nought to two by hand: ten means AI carries the task cleanly, and the thresholds that turn a total into YES, PARTLY or NO are published in the methodology. Each axis name links to its definition.
| Axis | Score (0–2) |
|---|---|
| Output | 2 |
| Inputs | 1 |
| Verification | 1 |
| Liability | 1 |
| Effort delta | 2 |
| Total | 7 / 10 |
The methodology and its thresholds are published in full.
FAQ
- Can AI qualify leads in live chat?
- Yes, it can ask set questions, label replies, summarise the enquiry and send it to the right sales route. It works best when your qualification rules and source information are explicit, with unclear cases sent to a person.
- What information does an AI chatbot need to qualify leads?
- Give it your current offer, service area, customer criteria, availability, disqualifiers, qualifying questions and hand-off routes. Do not rely on it to infer missing prices, policies or capacity.
- Can AI replace a salesperson on website live chat?
- Not for the whole sales conversation. It can absorb the repeatable first stage, but a salesperson still needs to handle ambiguous buying signals, objections, complaints and enquiries that do not fit the rules.
- Is it safe to use AI for live-chat lead qualification?
- It can be used for a defined first pass if you limit the information collected, test the routing and provide a human fallback. You remain responsible for missed leads, inaccurate answers and how visitor data is handled.
Nearby answers
- Can AI qualify my website form submissions?YES
- Can AI remove spam leads from my sales pipeline?PARTLY
- Can AI score my inbound website leads?YES
- Can AI ask qualifying questions to website visitors?YES
- Can AI turn customer enquiries into qualified leads?YES
- Can AI forecast revenue from my sales leads?PARTLY
Assessed by gpt-5.6-luna (gpt-5.6-luna) on 2026-08-13, second-checked by an independent model. Wrong somewhere? Email [email protected] and it gets re-checked.
The newsletter
AI news, new answers and product picks, straight to your inbox.