Lead qualification framework: how to define MQL and SQL so marketing and sales stop arguing about lead quality

The short answer

A lead qualification framework is a written agreement between marketing and sales on three things: what counts as a qualified lead, who decides, and what happens next within a set time. Define four stages (enquiry, marketing qualified, sales accepted, sales qualified), score each lead on fit and intent, and record a reason whenever a lead drops out. Then lead quality stops being an argument and becomes a number you can improve.

This guide is for founders, CEOs and heads of marketing who spend money on leads and cannot tell whether the leads are poor or the follow-up is. It gives you the definitions, a scoring table you can copy, a worked example with clearly labelled assumptions, a handoff agreement template and a 30 day plan. It works whether your leads arrive through a website form, WhatsApp, calls or lead ads, and whether you have a team of two or twenty.

Why lead quality becomes an argument

The pattern is the same in most growing companies. Marketing reports leads, cost per lead and form fills. Sales reports calls made and deals closed. In between sits a gap nobody owns: which of those leads should have been called, and in what order.

Three things go wrong in that gap:

• The same word means different things. To marketing, a lead is anyone who left a phone number. To sales, a lead is someone who might pay. Both teams use the word "qualified" and mean different things.
• Nobody records why a lead was dropped. "Not interested" covers a wrong number, a student, a competitor, a buyer with no budget and a good prospect who needed a call back next month. Without reasons, you cannot fix anything.
• Speed is nobody's job. A Harvard Business Review article from 2011, The Short Life of Online Sales Leads by James Oldroyd, Kristina McElheran and David Elkington, reported that most companies are not responding nearly fast enough to online enquiries. The study is more than a decade old, but the habit it describes is easy to find in any sales team today: leads sit in an inbox until someone has time.

If you pay for leads through Google or Meta, the cost is real. Every junk lead you cannot identify is budget the platform will keep spending to find more of the same. Our guide to Google Ads lead quality shows how to send the quality signal back to the platform, and that only works once you have a definition to send.

The four stages and who owns each

Most CRMs use a lifecycle model. HubSpot's documentation, for example, lists the stages in order as Subscriber, Lead, Marketing Qualified Lead, Sales Qualified Lead, Opportunity, Customer and Evangelist, and defines a marketing qualified lead as a contact that your marketing team has qualified as ready for the sales team, and a sales qualified lead as a contact that your sales team has qualified as a potential customer. You do not need seven stages. For most Indian SMEs and startups, four are enough.

The SAL stage is the one companies skip, and it is the one that stops the arguing. It forces sales to either accept a lead or say why not, in writing, on the same day. A rejected MQL with a reason is feedback. A silently ignored MQL is a wasted rupee.

If you are a small team where the founder takes every call, collapse this to two stages: "enquiry" and "qualified". Keep the written definition and the reason codes. Those two are the parts that matter.

Score on fit and intent, not on activity

Many scoring models reward clicks: opened an email, visited the blog, downloaded a guide. Those are weak signals on their own. A student can read ten articles. A procurement head can send one message that says "need a quote by Friday". Score two things separately.

• Fit: is this the kind of buyer we serve well and profitably? City, company size, industry, role, the service they need.
• Intent: how clearly have they asked for a conversation, and how soon do they want to act?

Keep the model small enough to apply by hand. If your team cannot score a lead in under a minute, they will stop doing it.

A scoring table you can copy and adapt

The weights below are an illustration built by Be12 Media for a service business, not an industry benchmark. Change the criteria to match your own best customers, then test them against your last 30 closed deals.

Thresholds (illustrative): 60 or more is an MQL and goes to sales at once. 30 to 59 goes into nurture with a follow-up sequence. Below 30 is closed with a reason code, with no further effort. Review the thresholds every month against what sales actually closed.

One rule to protect speed: a lead that explicitly asks for a call or a quote is an MQL regardless of score, unless it is spam or clearly outside your service. Do not make a person who wrote "please call me" wait for a score.

The five questions sales should ask on the first call

Scoring decides who gets called first. The first call decides whether the lead is an SQL. Frameworks such as BANT (budget, authority, need, timeline) and CHAMP (challenges, authority, money, prioritisation) are lists of questions, not scoring systems. Use them to structure a conversation, and write the answers into your CRM fields. Five questions cover most service businesses:

1. What are you trying to achieve? The need, in the buyer's own words.
2. What have you tried so far, and what happened? This tells you whether they are a first-time buyer or a switcher, and what they distrust.
3. Who else is involved in the decision? Authority, without sounding like an interrogation.
4. When would you like this to start, and what is driving that date? Timeline and urgency.
5. Do you have a range in mind? Budget. Offer ranges if they hesitate, and never push for a number on the first call.

A lead becomes an SQL when you can write a one line summary that includes the need, the decision maker, a rough range and a date. If you cannot, it is still a SAL, and the next step is another conversation, not a proposal.

The handoff agreement: a template

This is the part that makes the framework work. It is one page, signed off by whoever leads marketing and whoever leads sales (in a small company, that may be the same person wearing two hats, and writing it down still helps). Copy this and fill in the numbers your team can actually meet.

On speed, do not copy a number from a blog post. Pick a target your team can meet on a normal Tuesday, measure it, then tighten it. The evidence that quick replies matter is consistent in direction, and the HBR study above is one well known source, but your own data (reply time against won rate) is the only number that applies to your business. If you do not have the first-reply system in place, start with our guide to instant lead response.

Reason codes: the cheapest data you will ever collect

When a lead leaves the funnel, force a choice from a short list instead of free text. HubSpot's default lead status options, for instance, include Unqualified, Attempted to Contact, Connected and Bad Timing, which shows the idea: a status that says what happened. Build your own list of eight to ten reasons. A starting set for Indian service businesses:

• Spam, duplicate or wrong number
• Not reachable after the agreed number of attempts
• Outside our service or location
• Budget too low for what they want
• Not the decision maker, and no access to one
• Timing: revisit in a set number of months
• Chose a competitor (record who, if they say)
• Chose to do it in house
• Price objection after a proposal

After a month, count them. If "not reachable" is the largest bucket, your problem is speed or phone number quality, not ad targeting. If "budget too low" dominates, your ads or landing page are attracting the wrong buyer, and the fix is in the message and the price signal. If "outside our service" is high, your ad copy is too broad. Each pattern points to a different fix, which is why reasons beat a single lead quality score.

Worked example: a Chennai B2B services firm

Assumptions (illustrative, not a client result): a Chennai IT services firm runs Google Ads and a LinkedIn page. It receives 120 form and WhatsApp enquiries a month. The founder and one sales executive follow up. Marketing reports "120 leads". The founder believes about half are useless but has no data.

The firm sets the framework above in a week and runs it for a month. Here is how the funnel could read, with the numbers invented purely to show the method:

Now the firm can ask better questions. What is the cost per SQL by source, not cost per enquiry? Which campaign sends the 34 closed leads? If the job seekers all come from one search term, a negative keyword fixes it. If many MQLs come from LinkedIn and few SQLs, the audience may be right but the message is not specific enough. None of these insights exist while every enquiry is simply "a lead".

To calculate cost per SQL, divide the month's spend on a source by the SQLs that source produced. Compare it with your CAC payback period to decide whether the channel earns more budget. Lead volume alone cannot tell you that.

Qualifying leads that arrive on WhatsApp and calls

In India, many serious enquiries never touch a form. They arrive as a WhatsApp message or a missed call. Treat them as leads on the same terms:

• Create a CRM record for every WhatsApp enquiry and call, with the source set to "WhatsApp" or "phone", so they appear in the same funnel and the same reports.
• Use a short WhatsApp greeting message that asks one qualifying question, such as what they need and by when. The answer is intent data.
• Log missed calls and call back with a scripted opening. A missed call is an enquiry with a lower score until a person speaks to the caller.
• Keep consent in mind. Record how and when a person agreed to be contacted, and what for. India's data protection rules are phasing in, and our DPDP Rules checklist for marketing leaders explains what to record.

If you do not yet have a place where all of these land together, that is the first problem to solve, before scoring. A CRM that captures calls, WhatsApp and forms in one pipeline, such as Be12 CRM, makes the stages and reason codes above a drop-down menu rather than a spreadsheet chore. Read our guide to choosing CRM software for a small business in India if you are comparing options.

Close the loop with your ad platforms

Once you know which leads become SQLs and customers, send that result back. Google Ads and Meta can optimise toward the outcome you report. If you only report a form fill, they will find more form fillers, including the 34 job seekers in the example above. Report the SQL, or the won deal, where the platform allows it, using the server side methods covered in our guides to Conversions API and enhanced conversions and Google Ads lead quality. Check the current platform help pages before you build, because the options and requirements change.

A 30 day plan to put this in place

Common mistakes to avoid

• Too many stages. If your team cannot explain each stage in one sentence, remove stages until they can.
• Gating by budget too early. Many buyers will not state a budget before they trust you. Use budget to qualify on the call, not to reject a form.
• Scoring on vanity activity. Email opens and page views alone do not make a buyer. Weight explicit requests highest.
• Letting sales ignore leads silently. Require an accept or return with a reason. Silence is the most expensive outcome.
• Never re-testing the score. A scoring model is a hypothesis. Compare scores with closed deals every month and adjust.
• Counting leads instead of revenue. The report that matters shows SQLs, opportunities and won revenue by source, with cost against each.

What to do this week

You do not need new software to start. Open a spreadsheet, list your last 30 closed deals and 30 lost enquiries, and write what separates them. Turn that into four stage definitions and a ten line scoring table, then agree the handoff speed with whoever calls your leads. Within a month you will know which source produces buyers, which produces noise, and which part of the funnel is leaking. To go deeper on what comes after qualification, read our guides to lead nurturing and instant lead response, or see how our marketing automation service can set up the system with you.