Home · Business · Sales · Lead qualification

PARTLY

As of 13 August 2026, AI can only partly qualify leads on the phone.

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 neededchat-fluent

Who has to check ita colleague

What the alternative costsA human sales representative using your phone system and CRM is the alternative; the available tool information gives no price for it.

If this goes wrong: a promising lead is rejected or an unsuitable lead is passed to sales, wasting time and potentially losing revenue.

What to actually do

  1. Use a tool built for this

    The route this page recommends

  2. Do it yourself

    Second choice

    A chat interface, chat-fluent skill, and roughly 30 minutes until you can act on the result.

    How to actually do it

    1. Open your CRM and write down the fields a qualified lead must meet, including customer type, problem, authority, timescale, budget and next step.
    2. Gather your current product information, price list, service limits, approved claims, common objections and escalation contacts.
    3. Paste that material into the prompt and ask the chatbot to produce the qualification playbook and scoring rubric.
    4. Open a blank call script or CRM template and transfer the questions, answer fields, pass criteria and hand-off rules into the order you will use on calls.
    5. Role-play the script with a colleague using three different lead profiles, then record which answers the rubric classifies incorrectly or leaves ambiguous.
    6. Compare every product claim, price and qualification rule in the draft against your current source documents before using it with a lead.
    7. Use the approved script on a real call, capture the lead's answers in the CRM, and ask a colleague to check the transcript and qualification decision before progressing or rejecting the lead.

    Prompt

    Create a phone lead-qualification playbook for [COMPANY] selling [PRODUCT OR SERVICE] to [TARGET CUSTOMER]. Use only the information I provide and mark missing information as [NEEDS CONFIRMATION]. Produce: 1. a short opening that asks permission to continue, 2. the qualification questions in a natural order, 3. follow-up questions for vague answers, 4. clear pass, nurture and reject criteria, 5. a concise objection-handling section that makes no unsupported claims, 6. a call-notes template, and 7. a final scoring rubric from 0 to 10. Do not invent customer needs, product capabilities, prices, legal claims or buying signals. Keep questions open enough to reveal urgency, budget, decision process, current solution, timescale and operational fit. Include a hand-off rule for any question the caller cannot answer. Here is the information to use: [PASTE YOUR IDEAL CUSTOMER PROFILE, PRODUCT INFORMATION, QUALIFICATION RULES, PRICING, COMMON OBJECTIONS AND CRM FIELDS].

    Open it prefilled in ChatGPT or Claude, or copy it into Gemini, which takes no prefill link.

  3. Hand it to a person

    The distant third

    A person who owns the outcome does this end to end, worth it when the failure is dear.

What it gets wrong

  • AI does not reliably detect hesitation, sarcasm or a politically phrased refusal during a live phone conversation.
  • It cannot know whether a lead's stated budget, authority or urgency is genuine without evidence from the conversation and your business context.
  • It cannot take responsibility for rejecting a lead, making a promise or deciding that an exception to your rules is commercially sensible.
  • A generic model cannot access your current CRM history, pricing changes or internal approval rules unless you provide them.
  • Automated notes can preserve the wrong interpretation of an answer, so a colleague still needs to check important qualification decisions.

What caps this at PARTLY: judgement under ambiguity, real time truth 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.

AxisScore (0–2)
Output1
Inputs2
Verification1
Liability1
Effort delta1
Total6 / 10

The methodology and its thresholds are published in full.

FAQ

Can AI make qualification calls for me?
Not reliably as a complete replacement for a salesperson. AI can prepare the script and analyse supplied call notes or a transcript, but the live conversation, judgement and responsibility remain with your team.
Can ChatGPT qualify a lead over the phone?
ChatGPT can create the questions, scoring rules and follow-up prompts for a phone call. It cannot independently know whether a lead is being truthful or whether an unusual prospect is worth progressing, so a salesperson must conduct or check the call.
What should an AI ask when qualifying a sales lead?
Ask about the customer's problem, current approach, urgency, budget, decision-maker, buying process and timescale. Turn the answers into explicit pass, nurture and reject rules that match your product and sales process.
Can AI update my CRM after a qualification call?
It can draft structured notes and suggest CRM fields from information you provide. You should check the transcript and decision against the call before saving or triggering a sales action.

Nearby answers

All Lead qualification rulings

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.