How to Qualify a Sales Lead: 25 Questions Your AI Agent Can Score (Free Template, 2026)

If you want to know how to qualify a sales lead, start with the questions, not the call. Most inbound leads arrive with a name, an email and nothing else. So your team books a call, spends 30 minutes on discovery, and only then learns the lead has no budget or no need.

That is expensive, and it is also slow. In fact, research published in Harvard Business Review found that "most companies are not responding nearly fast enough" to online leads. When qualification waits for a human, response time suffers.

This guide gives you 25 lead qualification questions, grouped by what they measure. Each one comes with a scoring rule, so an AI agent can score the answer the moment it arrives. Then we show how your agent can turn the template into a live form in one message.

How to qualify a sales lead: the scoring model

Qualifying a lead means answering one question: is this person worth a sales conversation right now? Frameworks like BANT (budget, authority, need, timing) and MEDDIC exist to make that answer consistent. However, a framework only helps if you collect the same data from every lead.

That is why the best way to learn how to qualify a sales lead is to turn your framework into a form. Every question maps to one signal. In addition, every answer option carries points. As a result, the total score tells you what to do next.

The three buckets

Here is a simple model you can start with and tune later:

• Hot (70% or more of the maximum score): send the lead straight to your calendar.
• Warm (40% to 69%): follow up by email within a day, then decide.
• Cold (under 40%): thank them, share a resource, and keep them in your CRM.

These thresholds are a starting point, not a benchmark. Once you have a few weeks of answers, compare scores with the deals that actually closed. Then move the lines.

Why percentages instead of raw points

You will not use all 25 questions on one form. Instead, you will pick 8 to 12. So the maximum score changes from form to form. A percentage of the maximum keeps your buckets stable even when you swap questions.

25 lead qualification questions with scoring rules

The questions below are grouped into six signals: fit, need, timing, budget, authority and intent. For each one, the table shows answer options and the points your agent should assign. Replace the example tiers with your own ideal customer profile.

A note on open-text questions: an AI agent can score them too. Give it a short rubric (for example, "10 if they name a concrete cost, 3 if vague, 0 if blank"). Then it reads the answer and applies the rubric.

Fit: is this the right kind of company?

Fit questions filter out leads you could never serve well. Because they are easy to answer, put one or two of them first.

Need: do they have a problem you solve?

Need is the strongest signal for most teams. A lead with a sharp problem and a small budget often beats a big budget with no pain.

Timing: are they buying now?

Timing questions separate active buyers from researchers. Both matter, but they need different follow-ups.

Budget: can they pay for it?

Many leads skip a blunt "what is your budget?" question. So ask about budget in ways that are easier to answer honestly.

Authority: who decides?

Authority questions tell your rep who else must join the conversation. Therefore, they shape the meeting, not only the score.

Intent: how ready are they to talk?

Finally, intent questions measure how warm the lead is today. Put them at the end, when the person already knows what you offer.

How to build a lead qualification form from this template

A list of qualifying questions for sales is only half of how to qualify a sales lead. The form itself decides how many people finish it. So here are the rules we would follow.

Pick 8 to 12 questions, not 25

Start with one or two fit questions, then need, timing and budget. Next, add one authority question. Close with question 24, because it doubles as the call to action. If a question never changes what your team does, drop it.

Use conditional logic to skip what does not apply

Conditional logic lets you hide questions based on earlier answers. For example, if someone picks "Student" in question 3, the form can end early with a friendly message. Similarly, a solo founder does not need the "who else decides" question.

Make most answers multiple choice

Multiple choice is faster to answer and easier to score. Keep open text for two or three questions where the words carry real signal, such as questions 9 and 25. Meanwhile, your agent handles those with a rubric.

Route every bucket somewhere

A score with no action is just a number. Hot leads should see your calendar right away. Warm and cold leads should still land in your CRM, with the score attached, so nobody falls through.

How your AI agent builds and scores it with dForms

dForms is the form builder for AI agents: the one your agent actually wants to use. With one API key, your agent builds the form, publishes it, scores every answer and sends hot leads to your calendar or CRM. It is also free and open source.

Here is how the whole flow works:

1. Connect your agent. Create an API key on the Developers page of dForms, or connect the dForms MCP at mcp.dforms.ai from Claude or another MCP client.
2. Ask for the form in one message. Paste the template questions you chose and the points for each option.
3. Review the draft, then publish. Edits live in a draft until someone publishes. If you want to approve every form yourself, ask your agent to leave it in draft.
4. Read the scored answers. Your agent can list responses and analytics, so you can ask it which leads came in hot this week.

Here is a prompt you can copy and adapt:

Build a lead qualification form in dForms called "Talk to sales".
Use these questions from the 25-question template: 1, 3, 7, 9, 12, 16, 20, 24 and 25.
Give every answer option the points from the template.
Score open-text answers with this rubric: concrete cost or deadline = full points, vague = low, blank = 0.
If someone answers "Student" in question 3, end the form with a thank-you message.
Send leads that score 70% or more of the maximum to my booking calendar.
Leave the form in draft so I can review it before it goes live.

From there, dForms can sync contacts to HubSpot or send each response to a webhook. If you run an AI SDR, read our guide to building a lead qualification form for an AI SDR. And if you want the bigger picture of why agents need their own form builder, see AI form generators built for agents.

How to qualify a sales lead without these common mistakes

Knowing how to qualify a sales lead on paper is not the same as doing it well. Even a good template fails if the setup is wrong. Watch for these five mistakes:

• Asking everything up front. Long forms lose people. Ask the questions that change the decision, and nothing else.
• Scoring every question the same. Need and timing usually predict deals better than company size. Weight them higher.
• Never revisiting the thresholds. Compare scores with closed deals every month, then move the lines.
• Hiding the score from sales. If the rep cannot see why a lead scored 82%, they will redo discovery anyway.
• Treating cold as dead. A cold lead today might be hot in six months. Keep them in your CRM.

Want to see an agent and a person build one together? Read our Claude Duet guide to lead forms.

Frequently asked questions

What are the best questions to qualify a sales lead?

The best questions cover fit, need, timing, budget and authority. Start with role and company size for fit, the main problem for need, and "when do you want this in place?" for timing. Then add one budget and one authority question. Questions 1, 3, 7, 12, 16 and 20 from the template above cover all five.

How many questions should a lead qualification form have?

Most lead qualification forms work best with 8 to 12 questions. That is enough to score fit, need, timing, budget and authority. Beyond that, people tend to drop off. Use conditional logic to skip questions that do not apply.

Can an AI agent score open-text answers?

Yes, an AI agent can score open-text answers if you give it a rubric. Write the rule in plain words. For example: "10 points if they name a concrete cost, 3 if vague, 0 if blank." Then the agent reads each answer and applies the rule the same way every time.

What is the difference between BANT and lead scoring?

BANT is a framework that tells you what to ask about: budget, authority, need and timing. Lead scoring is the method that turns the answers into a number. In this template, each BANT question carries points, so the framework and the score work together.

What score makes a lead qualified?

There is no universal score that makes a lead qualified. A practical start is 70% or more of the maximum score for hot leads and 40% to 69% for warm ones. After a few weeks, compare scores with closed deals and adjust.

How do I send qualified leads to my calendar automatically?

Set a score threshold and route every lead above it to your booking calendar. In dForms, your agent can configure that routing when it builds the form. It can also sync every lead to HubSpot, so warm and cold leads are not lost.

Now you know how to qualify a sales lead with questions your agent can score. Pick your 8 to 12 questions, set the points, and let your agent build the form.