Home · Business · Sales · Lead qualification

YES

As of 13 August 2026, AI can detect fake leads for your business.

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 ityou

What the alternative costsA sales team can check leads manually against your CRM, contact records and qualification rules, while Apollo.io provides a prospect database with AI outreach sequences and enrichment.

If this goes wrong: you reject a genuine prospect or spend sales time pursuing a fabricated enquiry.

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 or lead form export and create a copy containing the lead ID, submission time, name, company, job role, email domain, phone number, source, message and any existing status fields.
    2. Remove unnecessary personal information and exclude passwords, payment details and private notes before pasting the data into a chatbot.
    3. Write down the rules for a qualified lead, including target customer type, service area, minimum information and reasons a lead may be spam, then paste those rules into the prompt.
    4. Add a small set of previously confirmed genuine and fake leads, labelled clearly, and paste the redacted lead table into the prompt.
    5. Run the prompt and export the results with the lead ID, risk rating, evidence, missing checks and recommended action kept beside the original record.
    6. Compare every high-risk flag against the original CRM record, your form logs and the email domain, and mark unsupported or ambiguous flags for human review.
    7. Contact or otherwise verify medium-risk leads using your normal business process, then update the CRM with the outcome before rejecting any lead automatically.

    Prompt

    Assess the following business leads for signs that they may be fake, low-quality or unsafe to pursue. Do not claim that any lead is definitely fake unless the evidence proves it. For each lead, return: 1) a risk rating of low, medium or high, 2) the specific evidence for the rating, 3) which checks are missing, 4) the safest next action, and 5) whether a human should review it before rejection. Check for invalid or suspicious contact details, disposable or mismatched email domains, repeated or near-duplicate submissions, inconsistent names and company details, implausible requests, missing information, copied wording and behaviour that conflicts with the qualification rules below. Treat unusual wording or a personal email address as clues, not proof. Never invent facts, search results or contact details. Do not use sensitive personal data to infer whether someone is genuine. Preserve the lead ID and quote the relevant input field for every finding. End with a separate list of leads that must not be rejected automatically. Business qualification rules: [PASTE YOUR RULES]. Known genuine lead examples: [PASTE EXAMPLES OR WRITE NONE]. Known fake or spam lead examples: [PASTE EXAMPLES OR WRITE NONE]. Lead data: [PASTE A REDACTED EXPORT OR TABLE].

    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 cannot prove that a real person submitted a form or intends to buy from you.
  • It cannot access your CRM, form logs, email verification service or call history unless you provide the relevant data.
  • It cannot reliably distinguish an unusual genuine enquiry from a carefully written fake one using text alone.
  • It cannot set the acceptable false-positive rate for your business or decide how many genuine leads you can afford to lose.
  • It cannot take responsibility for rejecting a lead, contacting a person or retaining their personal data.

Even on a YES, the friction has a name: judgement under ambiguity, private data access 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)
Output2
Inputs2
Verification1
Liability1
Effort delta2
Total8 / 10

The methodology and its thresholds are published in full.

FAQ

Can AI tell if a lead is fake?
It can identify warning signs and rank leads for review, such as duplicate submissions, suspicious domains and inconsistent details. It cannot prove that a lead is fake from a form or message alone, so do not reject genuine-looking leads automatically.
How do I use AI to check leads?
Give it your qualification rules, redacted lead records and examples of confirmed genuine and fake leads. Ask it to show the evidence for each flag and separate leads that need a human check from those that can be deprioritised.
Can AI filter spam leads from my CRM?
Yes, if you export the relevant CRM fields and give the model clear screening rules. Keep the original records, check the model's evidence against your CRM and form logs, and require human approval before deleting or rejecting leads.
What is the best AI tool for lead qualification?
Apollo.io is a purpose-built option because it combines a prospect database with AI outreach sequences and enrichment. It can support qualification and data enrichment, but you still need your own rules and human checks for suspicious leads.

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.