As of 13 August 2026, AI can turn customer enquiries into qualified leads.
Most people should hand this to a purpose-built tool.
Can you do it?
5 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 alternative that provides website live-chat with an AI agent trained on your content.
If this goes wrong: the system sends good prospects to the wrong queue or rejects them, wasting sales time and damaging trust with potential customers.
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 CRM, inbox or website chat export and collect a representative set of recent customer enquiries, removing personal details that are not needed for qualification.
- Write down your ideal customer profile, the conditions for a qualified lead, the reasons to disqualify an enquiry, and the rules for high, medium and low priority.
- Paste those rules and one enquiry into the prompt, then ask the chatbot to return the status, evidence, missing information, follow-up message and routing recommendation.
- Compare every recommendation with the original enquiry and your written rules, correcting any status that lacks direct evidence.
- Give the corrected examples to a sales colleague and ask them to identify inconsistent classifications, unsafe assumptions and unsuitable follow-up questions.
- Revise the qualification rules and prompt from that feedback, then test the workflow on enquiries that include clear leads, incomplete information and unsuitable requests.
- Send only approved follow-up messages, and record the final human decision beside each lead in your CRM so the rules can be audited and improved.
Prompt
You are helping qualify customer enquiries for a UK business. Use only the information in the enquiry and the qualification rules below. Do not invent facts, infer sensitive characteristics, or treat missing information as a positive or negative signal. Business and offer: [describe what we sell] Ideal customer profile: [describe the organisations or people we serve] Qualification rules: [list the conditions that make an enquiry qualified] Disqualification rules: [list the conditions that make an enquiry unsuitable] Priority rules: [explain what makes a qualified lead high, medium or low priority] Follow-up questions allowed: [list the questions the team may ask] Routing rules: [state which team or person receives each outcome] Customer enquiry: [paste the enquiry here] Return: 1. Qualification status: qualified, needs more information, or not qualified. 2. Priority: high, medium or low, only if the status is qualified or needs more information. 3. Evidence: quote or paraphrase the exact parts of the enquiry supporting the decision. 4. Missing information: list only information needed to apply the rules. 5. Follow-up message: write a short, plain-English UK message asking only the necessary questions. 6. Recommended route and reason. 7. Uncertainty: state what could make this decision wrong. Do not contact the customer, promise an outcome, or make a decision where the rules do not support one. Keep personal data to the minimum needed for qualification.
Open it prefilled in ChatGPT or Claude, or copy it into Gemini, which takes no prefill link.
What it gets wrong
- AI cannot know whether an enquirer has genuine buying authority when the enquiry does not establish it.
- AI applies the rules you give it, but it cannot decide whether your ideal customer profile is commercially sensible.
- AI can mistake politeness, urgency or detailed wording for buying intent.
- AI does not carry responsibility for contacting a person, rejecting an opportunity or storing their personal data.
- AI cannot replace a salesperson's knowledge of account history, local relationships and unusual buying situations.
Even on a YES, the friction has a name: judgement under ambiguity, consent and privacy and stakes of error.
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 | 2 |
| Verification | 1 |
| Liability | 1 |
| Effort delta | 2 |
| Total | 8 / 10 |
The methodology and its thresholds are published in full.
FAQ
- Can AI qualify leads from website enquiries?
- Yes, if you give it explicit qualification rules and the information in each enquiry. It can classify the enquiry, identify missing details, draft follow-up questions and route the result, but a salesperson should handle ambiguous cases.
- How does AI decide if a lead is qualified?
- It compares the enquiry with the criteria you provide, such as customer type, need, budget, authority and timing. It cannot reliably establish facts that the enquirer has not supplied, so the output should include evidence and missing information.
- Can AI automatically follow up with qualified leads?
- Yes, a configured chat or sales workflow can ask routine questions and send an approved message. Keep a human review step for unclear answers, sensitive personal data, unusual requests and any message that could make a commitment on behalf of the business.
- Is AI lead qualification accurate?
- It can be consistent against clear rules, but consistency is not the same as accuracy. Test it against real enquiries and have a colleague check borderline decisions before allowing it to route or reject leads automatically.
Nearby answers
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.