How to Qualify Leads in DMs (BANT and Other Frameworks for Setters)

Qualifying a lead means finding out, before you book a call, whether they have the problem you solve, a reason to fix it now, the ability to pay, and the authority to decide. In DMs you do that through conversation, one question at a time, reacting to each answer, so it feels like interest rather than an interrogation.

Qualification is where I've seen the biggest gap between setters who fill calendars with good calls and setters who fill them with no-shows and "just curious" people. It's also the part I spent the most time on when designing flows in ReplySetter. Here's how I approach it.

Why qualification matters more than booking rate

It's tempting to judge a setter on how many calls they book. But an unqualified call:

• wastes your closer's time,
• is more likely to be a no-show,
• drags down close rates and morale.

A slightly lower booking rate with much better-qualified calls is almost always more profitable.

The classic frameworks

BANT

The oldest one, from IBM:

• Budget: can they afford it?
• Authority: are they the decision maker?
• Need: do they have the problem you solve?
• Timeline: when do they want to solve it?

CHAMP

A reorder that puts the problem first:

• CHallenges: what problem are they facing?
• Authority: who decides?
• Money: budget?
• Prioritisation: how important is this right now?

MEDDIC

Used for complex B2B sales with long cycles. Overkill for most DM setting.

A simpler framework for DM setters

For DM, SMS and chat setting, I use four questions that map to BANT but feel more natural:

1. Goal: "What are you hoping to achieve?"
2. Gap: "What's getting in the way right now?"
3. Timing: "Is this something you want to fix now or later?"
4. Investment: "Is [range] in the ballpark if it's the right fit?"

Plus authority when it matters (B2B, couples, families): "Is anyone else involved in deciding this?"

Example questions by stage

Goal

• "What made you reach out?"
• "If this worked perfectly, what would be different in 3 months?"

Gap

• "What have you tried so far?"
• "What's been the hardest part?"

Timing

• "Are you looking to start this month, or more of a down-the-line thing?"
• "What's making now the right time?"

Investment

• "Our programs usually run [range]. Is that in the ballpark for you?"
• "Have you set aside a budget for this?"

Authority

• "Would anyone else need to be on the call?"
• "Is this your decision or a team one?"

How to ask without interrogating

These are the habits I've found make the difference:

• One question per message. Never stack them.
• React first. "Ah, that's really common with shift workers." Then ask.
• Give a little value. Share a quick insight between questions.
• Use their words. Repeat back what they said.
• Explain why you're asking money questions. "Just so I don't waste your time..."

Red flags and what to do with them

Not booking someone is a good outcome when they aren't a fit. Being kind about it protects your reputation.

Save the answers, every time

Qualification is only useful if the answers reach the closer. Save them to CRM fields:

• goal
• main obstacle
• timeline
• budget range
• decision maker

Then the closer starts the call already knowing the basics, and you can report on lead quality by source.

Making an AI setter qualify consistently

This is where AI is genuinely better than most humans: it asks every question, every time, and saves every answer. The way I set it up in ReplySetter:

1. A qualification step in the flow with objectives: goal, obstacle, timeline, budget. The step isn't done until each one is collected.
2. Instructions to ask one at a time, react to each answer and skip questions already answered earlier in the chat.
3. Fields: each objective is saved to a contact field as soon as it's known, via the CRM actions.
4. Exits: qualified → booking step; not a fit → polite close with a resource; not now → nurture follow-ups.
5. A switch that routes by answer, for example sending leads above a budget threshold to a senior closer's calendar.

Because each step shows what it collected and why it moved on in the trace, I can see exactly where unqualified people slip through and tighten that one instruction.